80506 Rachid Bennmessaoud, Uzma Basim, Anthony Cholst, and José R. López-Calix Editors Pakistan The Transformative Path Rachid Bennmessaoud, Uzma Basim, Anthony Cholst, and José R. López-Calix Copyright © 2013 The World Bank / Pakistan 20-A, Shahrah-e-Jumhuriat, G-5/1 Islamabad 44000, Pakistan All rights reserved The findings, interpretations, and conclusions expressed in this work are entirely those of the authors and should not be attributed in any manner to The World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, colors, denomina- tions, and other information shown on any map in this volume do not imply on the part of The World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. The World Bank encourages dis- semination of its work and will normally give permission to reproduce portions of the work promptly. ISBN: 978-0-9833517-1-9 Cover art: Jabbar Gull (2011); cover design: Kayhan Suleman The painting featured on the cover portrays the efforts of an ordinary mind reflecting on an inner world and looking toward its next phase of existence. The eclipse announces imminent change. The butterfly represents the transition. And the dispersing clouds symbolize the approaching realization and enlightenment. The Transformative Path calls for a journey of rethinking development choices and breaking away from the trajectory of low growth—toward a more prosperous and stronger Pakistan. Contents iii Foreword vii Acknowledgments ix Abbreviations xi Rethinking Development Policy Choices: Contributions to the Emerging Agenda 1 Three Central Messages 2 The Overall Agenda: Why a Piecemeal Approach Will Not Work 5 A Menu of the Most Urgent Actions 11 Implementing to Succeed 14 1  Building an Efficient Energy Sector 17 Main Issues 19 Policy Options 29 Notes 35 References 36 2  Reinvigorating the Agenda for Open Trade 37 Main Issues 41 Policy Options 49 Notes 52 References 53 3  Enhancing the Business Environment 55 Main Issues 58 Policy Options 65 Notes 70 References 70 4  Reforming State‑owned Enterprises 72 Main Issues 76 Policy Options 80 Notes 83 References 84 5  Revitalizing Manufacturing 86 Main Issues 91 Policy Options 98 iv Notes 101 References 102 6  Increasing Agricultural Productivity 105 Main Issues 107 Policy Options 114 Notes 117 References 118 7  Enhancing Labor Market Conditions for Vulnerable Groups 121 Main Issues 123 Policy Options 130 Notes 135 References 136 8  Recovering Strong Positive Trends in Poverty and Opportunity 138 Trends 139 Human Opportunity Index 145 Conclusions 148 Notes 151 References 151 9  Expanding Access to Quality Education 153 Background 153 Main Issues 155 Policy Options 161 Notes 165 References 165 10  Expanding Quality Health, Population, and Nutrition 167 Main Issues 170 Policy Options 176 Notes 179 References 180 11  Consolidating Social Protection 182 Main Issues 186 Policy Options 189 Notes 194 References 195 12  Managing Natural Disasters 198 Main Issues 200 Policy Options 201 Notes 205 v References 205 13  Revamping Governance 207 Main Issues 209 Policy Options 214 Notes 223 References 224 14  Improving Financial Management 227 Main Issues 230 Policy Options 237 Notes 239 References 239 15  Promoting Efficient Service Delivery with Decentralization 240 A Historical Perspective 240 Main Issues 246 Policy Options 250 Notes 255 References 257 16  Mobilizing Revenue 259 Main Issues 263 Policy Options 275 Notes 280 References 282 About the Authors 285 Foreword vii The World Bank often welcomes incoming governments with a series of diag- nostics studies and policy notes. Building on this tradition, it is my privilege and honor to present these policy notes as a product of a truly collaborative effort of the entire Pakistan Country Team at the World Bank. The notes reflect the knowledge accumulated through decades of policy dialogue, anal- ysis, and operational work, and enhanced by learning experiences in other countries and by working closely with the Government of Pakistan, Pakistani policy makers and economists, development partners, and other stakeholders. The underlying motivation for these notes is the urgent need for Pakistan to move toward a more sustainable and inclusive growth path. On the exter- nal front, the global economic slowdown has made increasing exports and attracting external financing harder, requiring a more competitive economy. And on the domestic front, successive natural disasters, a rapidly growing population, and the serious macroeconomic imbalances have also increased the need for higher economic growth, massive creation of good private sector jobs, and achievement of quality social outcomes. Indeed, even with the recovery of the global economy and the prompt mobilization of interna- tional support, government resources in the next five years will be stretched. A more open economy and greater selectiveness in public resources allocation will thus be essential. The 16 sector-specific policy notes are not exhaustive. Rather, they aim to deepen the dialogue by providing short, focused discussions of the major challenges facing Pakistan. They propose policy options to help accelerate growth, create more and better jobs, foster human capital and inclusion, and strengthen governance. A synthesis note extracts common messages and sets out a menu of key options with an overriding purpose—to help the govern- ment design its own Transformative Path for Pakistan. It is my hope that these policy notes will make an important contribution to Pakistan’s economic development. And I am confident that if we all work together, and with a solid vision and committed leadership at both the federal and provincial levels, the task of revitalizing Pakistan’s economy and social reform agenda is doable. I can also assure you that if Pakistan is ready to move forward on this agenda, it will find an eager partner in the World Bank. Rachid Benmessaoud Country Director, Pakistan World Bank Acknowledgments ix This report was prepared by a core team comprising Rachid Benmessaoud, Uzma Basim, Anthony Cholst, and Jose R. Lopez-Calix. It was a team effort, and we are grateful for the invaluable contributions of the authors of the policy notes: Rashid Aziz and Mohammad Saqib (Energy), José G. Reis, Daria Taglioni, Charles Kunaka, and Upali Nihal Pitigala (Trade Policy); Mehnaz Safa- vian, Sarwat Aftab, and Sarmad Shaikh (Subnational Business Environment); Sarwat Aftab and Sarmad Shaikh (State-Owned Enterprises); Dan Biller and Ernesto Sanchez-Triana (Manufacturing Policy); S. Amer Ahmed and Madhur Gautam (Agricultural Policy); Cem Mete and Rehan R. Jamil (Labor Markets and Vulnerable Groups); John Newman (Poverty); Halil Dundar and Huma Ali Waheed (Education); Shahnaz Kazi, Inaam Ul Haq, Luc Laviolette, and Kees Kostermans (Health); Iftikhar Malik and Lucian Pop (Social Protection); Haris Khan and Marc Forni (Natural Disaster Management); Naseer A. Rana, Iffat Idris, and Irum Touqeer (Governance); Saeeda Sabah Rashid and Uzma Sadaf (Public Financial Management); Hanid Mukhtar (Fiscal Decentralization); and Jose R. Lopez-Calix and Irum Touqeer (Revenue Mobilization). Jose R. Lopez-Calix led the overall coordination of the preparation of this report. This report would have never seen the light without the solid support from all the World Bank’s South Asia Region Sector Directors and Sector Managers who took direct involvement in its production and quality control: Ernesto May, Vinaya Swaroop, and Tony Verheijen (Poverty Reduction and Economic Management); Herbert Acquay, Maria Correia, Simeon Ehui, Karla Gonzalez, Shobha Shetty, Jyoti Shukla, John Henry Stein, Bernice Van Bronk- horst, and Ming Zhang (Sustainable Development); Amit Dar, Pablo Gottret, Jesko Hentschel, and Julie McLaughlin (Human Development); and Henry Bagazonzya and Sujata Nitin Lamba (Finance and Private Sector Develop- ment). We are particularly grateful for the guidance and steadfast encourage- ment of both Isabel Guerrero, former Vice President for South Asia Region, and Philippe Le Houérou, current Vice President for South Asia Region. We would like to acknowledge the priceless suggestions, guidance, and input brought in by peer reviewer and South Asia Region Chief Economist Martin Rama and by peer reviewer Ishrat Hussain, especially in the last stretch. Precious key inputs were also provided by Eric David Manes on exports diver- sification. Valuable inputs at different stages of the work were also provided x by Reynold Duncan and Eugenia Marinova. Our special thanks for produc- tion support go to Shafqat Mirza, Shabnam Naz, and Abdul Qadir in Islam- abad and to Lin Chin, Rui Ma, Muhammad Shafiq, Rita Soni, and Ekaterina Stefanova in Washington, DC. Finally, we would like to thank the publication team. Mehreen Saeed did precious work in helping us design the cover. Bruce Ross-Larson, Jack Harlow, and Rob Elson played a critical role in helping make our arguments, figures, and charts reader-friendly. And Diane Broadley and Elaine Wilson laid out the report. We also express our gratitude for the cooperation and contributions of our government counterparts and development partners in Pakistan during the last years of working together side by side. Their encouragement and technical advice on development-related policies created an exceptional atmosphere for knowledge exchange. Abbreviations xi BISP Benazir Income Support Programme CCI Council of Common Interest CCT Conditional cash transfer CGA Controller general of accounts CIT Corporate income tax CNIC Computerized National Identity Card DDMA District Disaster Management Authority DISCO Distribution company DRM Disaster risk management ERP Effective rate of protection ESCO Energy supply company FBR Federal Board of Revenue FDI Foreign direct investment FDRAP National Future Disaster Response Action Plan FOI Freedom of information GST General sales tax HOI Human Opportunity Index ICT Information and communications technology IPP Independent power producer IT Information technology KESC Karachi Electric Supply Company LGO Local Government Ordinance LHW Lady health worker M&E Monitoring and evaluation MAF Million acre-feet MDAs Line ministries, departments, and agencies MFN Most-favored nation MPA Member of Provincial Assembly MW Megawatt NADRA National Database and Registration Authority NARC National Agricultural Research Centre NEC National Economic Council NEPRA National Electric Power Regulatory Authority NFC National Finance Commission NGO Nongovernmental organization NLC National Logistics Cell xii NSPS National Social Protection Strategy NTDC National Transmission and Dispatch Company NTN National tax number OECD Organisation for Economic Co-operation and Development OTRI Overall Trade Restrictiveness Index PACCS Pakistan Customs Computerized System PAO Principal accounting officer PARC Pakistan Agricultural Research Council PBS Pakistan Bureau of Statistics PDMA Provincial Disaster Management Authority PEFA Public Expenditure and Financial Accountability assessment PEPCO Pakistan Electric Power Company PFM Public financial management PPHI People’s Primary Health Initiative PPP Purchasing power parity PRA Procurement regulatory authority PSLM Pakistan Social and Living Standards Measurement survey R&D Research and development RCA Revealed comparative advantage RTI Right to information SBP State Bank of Pakistan SECP Securities and Exchange Commission of Pakistan SME Small and medium-size enterprise SOE State-owned enterprise SRO Statutory regulatory order Tcf Trillion cubic feet TDS Tariff differential subsidy TFP Total factor productivity WAPDA Water and Power Development Authority Rachid Benmessaoud, Uzma Basim, Anthony Cholst, and Jose R. Lopez-Calix Rethinking Development Policy Choices: Contributions to the Emerging Agenda This synthesis note aims to provide an overall guiding framework to highlight and bring together some of the critical reform priorities identified in the following policy notes (see Box 1). The topics were selected based on ongoing or completed analytical work and where the World Bank has local and international expertise. They have also been informed by the par- allel preparation of a Country Economic Memorandum focused on job-enhancing growth. The policy notes aim to help the incoming government achieve its ambi- tious economic goals. The Pakistan Muslim League Nawaz manifes- to—“Strong Economy, Strong Pakistan�—establishes the overarching goal of breaking out of the trajectory of low growth to reach a GDP growth rate of more than 6 percent. To get there, it calls for increasing investment, with particular attention to energy, agriculture, transport, and cities. It calls for placing the energy sector on a solid footing by reducing losses, corporatiz- ing and privatizing energy companies, and rationalizing power tariffs. And it calls for opening markets and encouraging regional trade. It seeks to place Pakistan on a sound fiscal path with increased tax revenues, by reducing subsidies and losses in state-owned enterprises (SOEs) and limiting govern- ment borrowing. And it seeks to substantially increase investments in human capital—health, education, and social protection. These are admirable goals, and the policy notes aim to provide the government with options to meet these objectives. Challenges are formidable, and a bold approach is needed, focused on growth as well as stabilization. The past five years represent a historic politi- cal achievement with a democratically elected government completing its full term. On the economic front, however, gains have been limited. Economic growth has been modest at best, inflation has been in the double-digits most of the time, and formal unemployment has risen. There is no doubt that Paki- stan has had to manage through several exogenous shocks, such as natural RETHINKING DEVELOPMENT POLICY CHOICES disasters and the ongoing conflict in neighboring Afghanistan. Nevertheless, there has also been marked deterioration in many structural areas under government control, such as revenue collection, energy sector, social indi- 2 cators, governance, and the overall environment for private investment. Moreover, the current level of reserves and the fiscal deficit are unsustainable. They not only constrain the fiscal space for financing development needs, but also place the economy on the brink of a balance of payment crisis for the second time in less than five years. These macro conditions make solving the structural challenges even more pressing. Simultaneous action on various fronts and a break with the past economic trends will be crucial to putting Pakistan back on the path to prosperity. Strong leadership with clear vision, and sustained implementation, will be crit- ical for bringing about economic reforms that have eluded Pakistan. Essential ingredients missing in past episodes of reform have been a strong political will and a consensus to implement difficult policy decisions. Unfortunately, “stop and go� policies are the rule rather than the exception in Pakistan’s experi- ence with structural reform. The first few months of the political transition provide a unique window of opportunity. Making a bold start and appointing a competent team responsible for following through with solid and steady imple- mentation will be essential. Three Central Messages Accelerating growth in Pakistan has been possible in the past, and it can be again. The country has already enjoyed two previous episodes of sustained rapid growth averaging 7 percent a year (fiscal 1963–67 and fiscal 2004–07). Pakistan needs to come back to that trajectory. At that rate, the economy doubles in size every six to seven years, creating enough jobs for its expanding labor force, but also higher skilled and more productive private sector jobs. There could still be exogenous shocks that can, if unaddressed, derail efforts. But experience has shown that Pakistan’s economy can also be very resilient in preserving growth and poverty reduction despite civil conflicts, natural disas- ters, and complex political transitions. Pakistan has many “positives� that can help support accelerated growth and revive the economy. They include the legitimacy of many of its institutions; its capacity to implement national projects when there is a will; the impres- sive growth of remittances from abroad in the 2000s; its natural resource base and huge potential as a food and mining exporter; its strategic location neighboring some of the fastest growing emerging economies worldwide; and the rapid growth of women in the labor force every year. Pakistan also has a large informal sector serving a sizable domestic market concentrated in rapidly growing large and mid-size cities—and significant untapped potential for export growth in new markets, both regional and global. 3 These positives can be unlocked in three ways: • Message 1. Pakistan must aim to achieve higher growth linked to more and better jobs—especially among women and youths. Perhaps the most pow- erful and dramatic reason for Pakistan to change course is demographic. With the sixth highest fertility rate in the world, the country’s population is projected to double by 2025. The youth labor force is rising fast, and Pakistan needs to create about 1.5 million new jobs a year—about twice its present rate—to prevent rising formal unemployment. Growth and employ- ment, inextricably linked, can feed off each other in a mutually reinforcing manner. But unless growth accelerates, employment required to absorb a demographic bulge will not materialize and inequality will increase— aggravating the risk of conflict and social divisions. While the country is already creating low-­ productive jobs, formal and informal, they are not enough to sustain growth in the medium term. To sustain growth, Pakistan needs to create more productive jobs. • Message 2 . Pakistan must strive for inclusion and improving human capital as a cornerstone of rebuilding productivity. Pakistan is making progress in reducing poverty and improving social indicators, but not as fast as other countries in the region. Sustainable growth can best be achieved by pulling people out of poverty and into a growing middle class. In this sense, inclu- siveness is not merely about fairness—it is about creating a more educated and healthy middle class that is employable and more productive and that, as consumers, can drive economic growth. Inclusiveness should therefore be a central objective—reaching marginalized groups and promoting equal access to economic opportunities and government services. Special focus is needed for the conflict-­ a ffected regions of Khyber Pakhtunkhwa, Federally Administered Tribal Areas, and Balochistan, where strengthening human capital is harder but, if achieved, can yield transformational results. • Message 3. Pakistan can sustain its growth, jobs, and inclusion agenda only by enhancing governance and accountability. Good governance is funda- mental to accelerating growth and attracting private investment—by setting simple minimum rules and regulations that can govern key economic activ- ities with greater accountability and transparency. It can also increase the efficiency of service delivery, particularly for the poorest people, especially if equity is a major goal. In the last decade, indicators of many dimensions of governance have deteriorated, with weaknesses in the rule of law and control of corruption—and with few gains in government effectiveness and regulatory quality. Even with some improvements in public financial RETHINKING DEVELOPMENT POLICY CHOICES management, revenue collection has deteriorated and shrunk the already small fiscal space to finance development needs. A significant reversal of these downward trends is essential, as are improving the quality of public 4 administration and completing the decentralization process to promote a more accountable and efficient system of service delivery. It is important to emphasize that implementing these reforms ad hoc and piecemeal will not yield results. The 16 policy notes can be grouped under these three main messages (Figure 1). Each area of reform will reinforce the others—but, if not pursued, will undermine the others. For example, the cur- ­ rent approach to taxes and subsidies not only distorts business activities and creates avenues for corruption—it also reduces the government’s ability to supply electricity. A lack of energy in turn reduces business competitiveness, reduces trade, and slows growth and job productivity. Similarly, governance issues cross sectors, from education and health to trade policy. Partial imple- mentation will therefore undermine desired results. For instance, if prompt solutions are not found to reduce SOE losses, they will continue to constrain the fiscal space and keep jobs unproductive. Or if a solution is not found to reduce corruption in tax collection, overcoming taxpayer unwillingness to register and file will remain an uphill battle. And unless social protection builds effective insurance mechanisms to protect vulnerable populations against Figure Three central messages 1 Jobs and growth • Building an efficient energy sector • Reinvigorating the agenda for open trade • Enhancing the business environment • Reforming state-owned enterprises • Revitalizing manufacturing • Increasing agricultural productivity • Enhancing labor market conditions for vulnerable groups Inclusion and human capital development • Recovering strong positive trends in poverty Policy notes and opportunity grouped by • Expanding access to quality education the three • Expanding quality health, population, and key messages nutrition services • Consolidating social protection • Managing natural disasters Governance and accountability • Revamping governance • Improving financial management • Promoting efficient service delivery with decentralization • Mobilizing revenue natural events, households that climb out of poverty will likely fall back follow- ing the next natural disaster. In this sense, the whole would be far greater than the sum of the parts. 5 The Overall Agenda: Why a Piecemeal Approach Will Not Work For details in each area of reform, readers are directed to the individual policy notes of most relevance to their concern. Each note provides a short analysis of a few key issues, a menu of priority actions, and their potential sequencing in the short to medium term. But it is also critical to look at the benefits of implementing reforms as a whole and see how they interrelate. Simultaneous and persistent efforts will have higher probabilities of changing the dynamics required on most fronts. Jobs and growth To accelerate growth, Pakistan needs to address its interrelated binding constraints. The key constraints are for the most part well known, and addressing them is indeed at the core of the Pakistan Muslim League Nawaz manifesto. They include poor access to steady electricity, macro fiscal vulnera- bilities, low access to finance and credit (especially for small and medium-size enterprises, or SMEs), trade restrictions, business-­ u nfriendly cities to service growth and green manufacturing, excessive business regulations, poorly performing SOEs, and gaps in infrastructure financing. Some of the main interlinkages are as follows. Reliable energy is the cornerstone of a strong business environment. It is not an understatement to say that energy is the top constraint to growth and jobs. If the country had solved the power load-shedding crisis, GDP growth would have been 1.5–2.0 percentage points faster. Without a reliable and efficient source of energy, businesses simply cannot be competitive. And many businesses have either to invest in captive power generation capacity or face closure and possible bankruptcy. But just as significant is the inverse—the importance of the business environment in resolving the energy sector governance dysfunctions. An important goal is to develop lower cost sources. But the amount of invest- ment needed suggests that—without private financing and a strengthened over- all macro, business, and trade environment—this will not be an easy task. An efficient energy sector is necessary to achieve fiscal consolidation and improve human capital. It clearly is not possible to make the fiscal situation sustainable without a dramatic reduction in power subsidies. This then feeds into strengthening human capital in two ways. First, without reducing power RETHINKING DEVELOPMENT POLICY CHOICES subsidies, Pakistan will not have the fiscal space to make the increased invest- ments in education, health, and social protection it so desperately needs. Second, targeted protection to the most vulnerable low-power consumers is in 6 turn critical to creating the political space to reduce subsidies—most power subsidies go to the well off. Fiscal consolidation will help improve access to finance by SMEs. The large and rising fiscal deficits of the past four years have to be financed. This financ- ing adds to a public debt burden that exceeds the 60 percent of GDP ceiling allowed by the country’s Fiscal Responsibility Bill. More important, the sky- rocketing domestic financing of the deficit has contributed to a private sector credit crunch—severely crowding out the private sector. Heavy SOE losses and guarantees for about 2.5 percent of GDP are also worsening the fiscal situa- tion. While large businesses can often obtain nonbank financing, the most vulnerable are the SMEs, ironically the most dynamic and job-creating portion of the business community. Bank lending to SMEs is only 6.6 percent of total advances. But just as important is the reverse flow. The growth of businesses and SMEs is critical not just for growth and jobs, it is also necessary for the government to increase the tax revenues needed to bring down the deficit. Fiscal consolidation is necessary to improve the perception of reduced country risk and raise private investment. A clear sign of this linkage is the Emerg- ing Markets Bond Index, which rose above 1,000 basis points in March 2013 (later falling, following the election). Coupled with a downgrade of Pakistan’s sovereign bond ratings in 2012, this has made external finance extremely expensive for Pakistan, resulting in a sudden stop in external flows and requir- ing excessive government borrowing domestically. Bringing the macro stance into greater sustainability would help improve the perception of an invest- ment-friendly country with much lower risk—and to bring down interest rates and inflation for businesses. To unlock the manufacturing potential, attention needs to be devoted to urban development, green manufacturing, and infrastructure financing. Current urban development strategies and zoning regulations do not foster produc- tive or commercial activity. As a result, there is a lack of multifunction urban areas that emphasize density and commerce. Manufacturing continues to be heavily concentrated in low value-added consumer products that attract little investment. To take advantage of the important benefits that agglomerations produce for job generation, cities should become growth-friendly clusters that foster both competition and learning. This should in turn help introduce new more competitive technologies, including those that meet basic environmen- tal and social standards increasingly required for exports. Equally important is that cities be linked to each other to get the most out of agglomeration, connecting industrial clusters. This requires upgrading, extending, and reha- bilitating transport infrastructure to build trade corridors (within Pakistan and between Pakistan and its neighbors). 7 Improved trade policy is necessary to foster export (and rural) diversification, business competitiveness, and revenue. The trade regime is plagued with priv- ileges: 9 out of 10 manufactured products granted statutory regulatory order (SRO)–based tariff exemptions benefit a single local monopoly producer. And contrary to the most-favored nation (MFN) declared tariff rates, there are myriad tariff slabs. Simplifying the tariff and trade regulation regime and reducing SROs would help reduce Pakistan’s anti-export bias and create a level playing field for competition—without which Pakistan businesses will fall behind. While some businesses would surely face stronger competition, the sooner Pakistan integrates into the regional market, the faster its businesses will become competitive overall and benefit from the high growth rates of its large neighbors. Businesses would also benefit from the greater availability of trade finance. Trade logistics businesses such as trucking, rail, roads, ports, trade finance, warehousing, and storage would also benefit. And farmers would gain greater access to advanced technology and new markets. But just as important, increased trade and a simplified tariff regime should increase government revenue, enabling the government to invest in further trade-en- hancing improvements, and contribute to expanding agricultural productivity. In all these efforts, it is important to focus on youths and women. Overall economic growth and even high levels of education have not translated into high levels of employment for these two vulnerable groups. Similarly, skills training and temporary jobs are not widespread enough to transition them into the workforce. Providing incentives for greater involvement of the private sector in program design, on-the-job training, and temporary initial place- ments is critical for achieving the required expansion rates in high-quality jobs. Self-employment and entrepreneurship programs would also be important. Inclusion and human capital Poverty diagnostics, an important tool for decision making on improving inclusion and human capital, needs to be strengthened. Outdated data and inaccuracies make it difficult to diagnose poverty, and it is critical that the capacity and autonomy of the Pakistan Bureau of Statistics (PBS) be assured, with its findings made public. In the meantime, the Human Opportunity Index provides a useful way to identify whether there has been progress in both the coverage and quality of service provision. Inclusion policies should RETHINKING DEVELOPMENT POLICY CHOICES ensure equal opportunity for the vulnerable segments of population includ- ing women, underserved pockets in Punjab and Sindh, and conflict-affected regions of Khyber Pakhtunkhwa, Federally Administered Tribal Areas, and 8 Balochistan where outreach and service delivery has become increasingly challenging. One promising option to expand coverage is to contract nongov- ernmental organizations or private providers on a low-cost basis to reach out to population in difficult areas. Accelerated progress in basic education and health services will enable Pakistan to make faster progress in growth and jobs. Businesses need more productive, healthy, and educated workers to grow. But while human develop- ment indicators have improved in the last decade, they still lag countries at a similar income and many of Pakistan’s neighbors. Pakistan ranks 145 of 187 countries in the United Nations Development Programme’s Human Devel- opment Index. It has the world’s second highest out-of-school population (7 million), of which two-thirds are girls. Enrollment and completion rates for primary education are among the world’s lowest. Health coverage is also dis- mal, with under-five malnutrition rates as some of the world’s worst. And the country is also one of the lowest spenders on education (1.9 percent of GDP) and health (0.9 percent of GDP); less than a third of the minimum average for countries at similar incomes per capita. As noted, it will be critical to make progress on reducing nontargeted power subsidies and increasing tax reve- nues in order to have the fiscal space to increase education and health spend- ing. Reform initiatives in Punjab and Sindh have delivered on some counts, but the challenges continue to be huge as these provinces are overspending on staff salaries rather than on nonsalary priority social investment. With the management and financing of social services decentralized to the provinces, national and provincial standards need to be coordinated, and their achieve- ments monitored to address inequities. Developing a system to set standards in hiring, track performance, and monitor learning outcomes should be the cornerstone of quality service delivery. Consolidating progress on social protection will make a huge difference not only for the targeted poor but also for the government to reduce nontargeted subsidies. Whereas Pakistan spends about 3  percent of GDP in untargeted subsidies, the pro-poor–targeted Benazir Income Support Programme (BISP) spends barely a 10th of that figure, with more inclusive results. The BISP has an efficient and modern system for targeted cash transfers to the poor, and its database can be expanded to integrate other social programs under a targeted approach. However, efforts are also needed to consolidate these gains through a coherent social protection framework across provinces, while replacing expensive universal subsidies (power, wheat) with more targeted interventions. This will require an increase in social protection spending but be far less expensive than the current nontargeted programs. Better management of natural disasters is a prerequisite for sustained growth. 9 Pakistan is a disaster-prone country that is extraordinarily exposed to earth- quakes, floods, droughts, and cyclones. The country has had 2 floods in the last 3 years, 14 cyclones in the last 40 years, 2 droughts in the last 12 years, and 2 earthquakes in the last 15 years. This hazard profile, when combined with rapid population growth and urbanization, greatly increases the vulnerabil- ity of human and physical capital. The major challenges in managing natural disasters are threefold: preparedness, well-­ executed emergency response, and strong dedicated institutions and data systems to oversee mid- to long-term rehabilitation. It will also be critical to mainstream disaster risk reduction in urban planning, factory design, and other infrastructure. Governance and accountability Strengthening governance and accountability is the glue that will acceler- ate growth and improve social service delivery. The governance indicators of the Global Competitiveness Report 2012–2013 ranks Pakistan as follows: 109 in transparency of government, 116 in property rights, 119 in irregular pay- ments and bribes, 127 in reliability of police, 129 in favoritism in decisions of public officials, 132 in business cost of crime, and 143 in business costs of terrorism. But not all news is bad. Recent positive developments include the growing entrenchment of democratic processes, the transfer of greater responsibility for service delivery to the provinces, and the exponential growth in electronic media (enhancing transparency and accountability). The governance agenda is multidimensional and includes four priority areas: • Improved public administration and service delivery. Improvements can be achieved by reforming the civil service, building subnational and local capacity to monitor and ensure effective implementation, allowing citizen’s voice and participation in local planning, monitoring, and accountability, and establishing a local government system that has the requisite adminis- trative and financial autonomy. • Transparency. Pakistan needs to consolidate its public financial management system and establish effective right to information regimes at the federal and provincial levels. The former revamps budget and cash management, strengthens the complaint mechanisms at public procurement regulatory authorities with data publicly available, and conducts performance audits of key public transactions. The latter requires passage of right to information legislation, followed by proper implementation and enforcement and aware- ness-raising campaigns for citizens. RETHINKING DEVELOPMENT POLICY CHOICES • Accountability. Setting goals and contracts up front and holding officials accountable should enable the government to regularly report on progress and then reduce excessive control or political interference. For example, in 10 revenue mobilization, this could make a difference in designing a tax sys- tem that is broad, simple, and equitable and that facilitates tax registration and compliance through transparent information technology systems. • Anticorruption . Pakistan should undertake a thorough systemic diagnosis, focusing on key institutions (such as the police, tax administration, and the like), devising carefully deliberated strategies to address the issues, ensur- ing proper implementation of the strategies, and reaching out to the inter- national community to make progress visible and widely supported and appreciated. Stronger public financial management can be a foundation for improving ser- vice delivery and making public expenditures more transparent. The public financial management system is fairly well developed but is not consolidated. The most notable initiatives have been adopting a uniform International Monetary Fund–Government Finance Statistics compliant chart of accounts for budgeting and accounting, separating accounting and auditing functions, introducing a pilot medium-term budgetary framework at the federal level, implementing a governmentwide financial management information sys- tem, enabling automated budget compilation, and establishing procurement regulatory authorities for Punjab and Sindh. Beyond keeping pace with these achievements, the next priorities are to develop an effective decentralized budget management system, improve cash management, and strengthen the capacity of the regulatory authorities for procurement. Decentralization can promote efficient service delivery. The 18th Amend- ment requires greater accountability by provincial governments, which is a strong incentive for them to improve the outreach and quality of services. Although a wide range of central ministries were abolished, a large num- ber of ad hoc functions seem to have simply been assigned to new federal ministries. Further accountability at the provincial level could be devolved toward the local and community levels. And given the fiscal imbalances in the federal budget, the level of budgetary transfers may need to be rebalanced to ensure consistency with the devolution of responsibilities—or new revenues will have to be generated both at the provincial and federal levels. In this regard, the roles and functions of each level of government should be further clarified. Beyond service delivery and accountabil- ity, decentralization holds the promise of stronger revenue mobilization, but provincial rules on expenditures and debt management may need to be revisited. Mobilizing more revenue fairly and effectively is critical not only for growth and macroeconomic stability but also for governance. If Pakistan wants to create the fiscal space required for its development needs—both for infrastruc- ture and social service—it has no choice but to implement a comprehensive tax 11 reform agenda that is perceived as equitable in its policy design and effective and free of corruption in its administration. At less than 10 percent of GDP and decreasing, the tax ratio is among the lowest worldwide. The focus needs to be on making taxation simple, on broadening the tax base, on ensuring compliance, and on modernizing the tax administration—with steady tenure and professional hiring policies in the Federal Board of Revenue (FBR), and enhanced transparency and accountability about its results. On tax policy, the elimination of SROs, exemptions, and zero rates that make compliance low should enable authorities to raise more revenue with equity. Lowering personal and corporate income taxes to international averages could also be considered once revenues start to pick up. Customs tariffs should be simpli- fied to a maximum of three slabs. Here again, governance is critical. Creat- ing a more autonomous, performance-based, and modernized FBR and a joint Finance-FBR Tax Policy Commission would be two important steps toward more effective revenue mobilization. As tax revenue relies too heavily on federal taxation (about 95  percent), while paying too little attention to provincial governments collection potential, provincial revenue administra- tions should also follow such a results-oriented model. A Menu of the Most Urgent Actions A solid start by the government to move forward comprehensively across many areas in the first 100 days will be essential. What has derailed past reforms is hesitant and fragmented reform efforts. A carefully chosen set of front-loaded measures would, by contrast, earn the government the political capital to make further (and earlier) medium-term advances on the reform agenda. The most important reform to initiate is in the power sector, as progress in all other areas, including fiscal management and the private sector, depends on it. Within the power sector, action is needed on several fronts simultaneously. First, a single-point power authority should be established—and managerial autonomy, performance standards, and accountability should be introduced for power utility firms, while considering some for privatization. This approach should provide clear incentives to reduce theft and losses and to increase col- lections. Second, a clear schedule of power tariff increases must be established to bring average notified tariffs closer to determined tariffs alongside visible enhancements on power reliability. The goal would be to enable distribution companies to achieve full cost-recovery and perform basic operations and RETHINKING DEVELOPMENT POLICY CHOICES maintenance (life-line below-cost tariffs on the basic block could be maintained for the poorest). This can be combined with a schedule to improve power reliabil- ity for paying customers and reduce subsidies. Third, while the circular debt is a 12 symptom and can be unwound over time once the system is put on better foot- ing, consideration can be given for taking it off the books of the energy compa- nies and parking it for now. Fourth, a medium-term plan for sequenced priority investment in both domestic gas and hydropower should be initiated, including opening domestic gas to private sector investment. The Water and Power Devel- opment Authority’s financial and planning capacity should be strengthened to enable it to raise investment funds more easily from the private sector. The second most important reform to initiate early on is revenue mobiliza- tion. This, combined with a reduction of power subsidies, is a precondition to creating the fiscal space for increased investments in infrastructure and human development. First, ad hoc tax and tariff reductions and SROs (except priority medicines and food items for the poor) should be eliminated. Second, the tax base should be broadened by using existing information technology systems to identify and register taxpayers and businesses that have not yet been included. This can be combined with enhanced electronic filing of taxes to facilitate compliance. Third, customs tariffs should be sim- plified to a maximum of three slabs; corporate and personal income tax can also be streamlined. Fourth, the FBR should be made more autonomous with greater enforcement capacity. Finally, incentives may be considered to allow provinces to raise own-generated resources, perhaps starting by broadening the coverage of services taxed by the general sales tax, upgrading the urban property tax (combined with an agricultural income tax), and replacing the one-timeregistration tax on motor vehicles with an annual license tax and fixed fuel levy. Excise taxes on discretionary and luxury items could also be considered. The third most important action is to reinvigorate SOE reform and the busi- ness environment. Deploying the private sector as the engine of development is long overdue, and many actions could be done quickly. To create a level play- ing field and improve services, the government should prepare a roadmap to privatize or restructure loss-making SOEs in a short time-frame. To reduce red tape, it should also prepare fairly quickly the ground for a one-stop shop for new investors. On the financial sector, the State Bank of Pakistan (SBP) could monitor and make private SME expansion a priority for banks. Streamlining construction permits would also be important. Improving the registration base for land and housing, and allowing movable collateral, would increase the number of firms and individuals using more collateral to borrow funds. Special programs should be considered for youths and women to ensure they can participate in this growth, through targeted on-the-job training programs and greater microfinance for potential entrepreneurs. The fourth priority should be the regional agenda, focused not only on 13 India but on all regional countries. Completing the trade normalization process with India and granting it MFN status would help Pakistan benefit quickly from the fast growth and large markets. Conservative estimates sug- gest that bilateral trade flows could multiply at least three times, and most observers agree that the growth-enhancing dynamics that this process would unleash would be even more significant for foreign direct investment (especially information technology and manufacturing), services (including financial and tourism), integrated value chains in manufacturing, and power projects. A power transmission link with India, under a power tariff arrangement benefiting both countries, could be implemented in the next 6–12 months with a potential capacity of 1,000 megawatts. And over the medium term, bring- ing electricity and natural gas from Central Asia could have similar potential. Trade reform should be done in combination with improving trade logistics, including upgrading border management, with a focus on the Wagha border. Special economic zones in border areas could also be considered. The fifth priority should be to improve human development. With the management and financing of education and health decentralized to the provinces, there is a need to strengthen national and provincial standards and provide transparent mechanisms for increased budget transfers linked to performance. In education, building on the programs already under way in Punjab and Sindh, there is room to increase support for low-cost private provision of education, particularly in areas underserved by the state, as well as to strengthen accountability and learning assessments to provide incentives to improve education and reduce absenteeism. In health, the most pressing need is to increase spending on nonsalary activities. To fight staff absentee- ism and strengthen outreach, the government could consider contracting out management or greater use of nongovernmental and community organiza- tions. Expanding the Lady Health Workers Program is necessary since the regions not covered are the most disadvantaged. For social protection, the BISP database provides a golden opportunity to contain untargeted subsides and redirect more resources to the poor to help them manage the afteref- fects of reductions in nontargeted subsidies. Pilot BISP programs for pri- mary education and health insurance should be assessed and scaled up, with a contingency fund for natural disasters considered. The sixth priority for making all these efforts more effective is to strengthen governance and accountability. A key first step in reforming the energy RETHINKING DEVELOPMENT POLICY CHOICES sector is to commercialize or privatize energy companies to strengthen their accountability to consumers and their ability to make cost-­ recovery decisions. For decentralized service delivery, a critical element will be to 14 set goals, collect data on them, and hold service providers responsible for meeting them. Greater use of contracting out service delivery in some areas of education and health can help deliver. It is also recommended that ­nstitutions—the FBR, SBP, and PBS—be made more auton- three critical i omous, with greater capacity and enforcement authority. Beyond these steps, greater use of e-services throughout the government can facilitate service delivery with feedback mechanisms. Financial management can be enhanced by improving the capacity of the regulatory authorities for procurement to monitor the performance of public entities and switch to e-procurement. The right to information laws and regulations should be approved. On devolution, there is a need to clarify roles and responsibilities (including budgeting, financial reporting, and taxation powers) between national and provincial authorities to strengthen accountability. Finally, civil service reform should be initiated to reduce turnover, to attract solid and specialized professionals, and to rationalize pay scales and make them transparent. Implementing to Succeed Far-reaching reforms will take both leadership and building consensus. The suggested actions included here should be considered as a menu of options, to be selected and guided by the new administration’s goals and ambi- tions—and by political realities. But the challenge for the government is that incremental (piecemeal) actions will not be enough to change the dynamic. For comprehensive reform, implementation will require strong political will, expert design, and persuasive negotiation. Powerful individuals and well-organized groups with vested interests will continue to avoid changes, as in taxation, agriculture, and trade reform. Thus, a broad consensus with a solid commu- nication effort is needed as part of the change management process to over- come these interests. The near-­crisis situation in Pakistan can also become a positive motivator for change. And early gains (low-­hanging fruit) can provide momentum for further progress. Achieving success will also take competent and skilled managers and expert designs. Political leadership at the top is essential for commitment and credibility. But vision and political decision need to be complemented by skilled managers and persuasive negotiating capacity to champion and design reforms through the complex political and social milieu. All successful past reforms in Pakistan featured strong support from top leaders, essential for reaching political support, and from highly respected professionals, both for their academic credentials and deal-making skills. Solid technical design also makes the acceptance of reform more likely, especially if it rightly identifies winners and losers and provides for proper contingency actions to minimize opposi- 15 tion. Financial support from donors is secondary, and not determinant; if it forces reform with no ownership, it can even be counterproductive. But it can prove useful when complemented by solid institutional ownership and timely advice and technical assistance. A bold start, with a properly sequenced rollout, will enhance the likelihood of success. A carefully chosen set of front-loaded measures would earn the government the political capital to make further (and earlier) medium-term advances. A possible sequence would be for the government to begin with measures that consolidate fiscal sustainability, set new growth dynamics into motion, and address the most pressing social commitments. Phasing out expensive subsidies, initiating energy and tax reform, and taking key actions in social protection—like consolidating and expanding the BISP—should also not be delayed. These steps might be followed by a thorough exercise review- ing spending priorities, consolidating ongoing sectoral reforms (especially for education, energy, and agriculture), and introducing institutional reforms (especially those for public financial management, trade, and vocational train- ing). These reforms would stimulate exports, strengthen governance, and give the new government an image of commitment, participation, and transpar- ency—which would help it win support for further reforms. Special attention should be given to the most vulnerable people or areas. Many of the reforms noted here are most likely to benefit those already well integrated into Pakistan’s economy. Large groups of people—particularly the poor, youths, and women—will need special measures to ensure that they too can participate in and benefit from these reforms. So, too, will special mea- sures be needed to ensure that the most conflict-affected areas are not left out. This will require a deeper understanding of the drivers of conflict, so that they can be addressed. Above all, sustained implementation matters. There is no substitute for politi- cal will to make bold decisions. But the key to success will be implementation, implementation, and implementation. It does no good, and probably some harm, to take half-hearted steps that are then partly reversed when facing obstacles. Most reforms, because they are complex and intertwined, will need concerted actions by multiple official entities across several years. A cross-sec- toral reading of the policy notes would facilitate collective thinking on the complexity of the issues—and highlight the need for strong and coherent RETHINKING DEVELOPMENT POLICY CHOICES collective and comprehensive government action. To facilitate reform imple- mentation and coordination across multiple entities, the government could consider creating small monitoring cells or performance management units 16 at the Prime Minister and Chief Minister offices to follow up and manage the overall progress on targets. A possible list of the 12 most urgent transformational reforms to Box initiate 1 • Strengthen power sector governance. Establish a single point power authority and appoint profes- sional management in the boards of power distribution companies, bring average notified power tariffs close to determined tariffs along with visible reduction in power outages, and create incen- tives for private investment in gas fields. • Attract private investment. Restructure and privatize selected power distribution companies and other loss-making SOEs, and expand commercial credit to SMEs. • Manage fiscal crisis. Eliminate tax exemptions and zero rates (and related SROs); identify and reg- ister 1 million new taxpayers using information technology systems and databases, and increase fuel taxes and special excise duties. • Reorient public spending. Phase down untargeted subsidies to power and SOEs, and reassign funds to key infrastructure spending and maintenance (power, water irrigation, and roads) and targeted social spending. • Reinvigorate trade liberalization and competitiveness. Eliminate trade-related SROs and simplify customs tariffs to a maximum of three slabs. • Strengthen economic governance. Strengthen the autonomy and enforcement capacity of the FBR, SBP, and PBS, and ensure stability of tenure and accountability of key government positions. • Enhance business environment. Establish a one-stop shop for investors in obtaining licenses and registration, and streamline construction permits and property registration procedures. • Tap regional potential. Complete normalization process with India granting it MFN status, and sign a power transmission connection and trade agreement. • Improve social service delivery. Double funding for education, health, and nutrition nonsalary bud- gets, set a system for tracking teachers’ professional recruitment and performance, and use low-cost private outsourcing mechanisms. • Protect the poor and vulnerable. Consolidate gains under the BISP through a coherent social pro- tection framework across provinces, increase the BISP budget and extend pilot programs on educa- tion, child labor prevention, and health insurance, use the BISP scorecard to redirect expensive power subsidies to poor households, create contingency fund and micro-insurance programs against the risk of natural disasters, and make poverty data available to the public. • Address corruption. Approve right to information law, set e-procurement to make public biddings transparent, perform audits of major procurements, and establish public procurement regulatory authority two-tier complaint redressal mechanisms. • Complete devolution. Establish financial and administrative autonomous local governments, and clarify roles and responsibilities among national, provincial, and local authorities (budgeting rules, financial reporting, and taxation incentives). Rashid Aziz and Mohammad Saqib Building an 1 Efficient Energy Sector Pakistan’s energy sector is facing a serious crisis. Key challenges include large and growing shortages of energy, high energy costs, and inefficiencies that prevent the sector from financing all its costs. It therefore relies heavily on government support, through subsidies and funding for almost its entire investment program. Specific actions to overcome the dire situation can be classified along three main lines. The first are actions to overcome the investment deficit, which requires least-cost investment plans based on low-cost supply sources (notably hydropower), market competition, more efficient energy use, and incen- tives to manage consumer demand. Second, steps to improve sector finances are needed. Key actions include modifying the current universal national tariff, implementing cost- recovering tariffs, allocating additional gas for power generation, and improving utilities’ operational and commercial performance, including commitments to pay their suppliers (particularly the Water and Power Development Authority, for hydropower) in a timely manner. Third, to improve sector governance, managerial autonomy and account- ability need to be introduced in power utilities. Performance contracts or other tools for monitoring and improving the performance of the public utilities—for example, outsourcing management, leasing, sale of shares along with management control, and privatization— should be considered, and the approach deemed most appropriate should be introduced. The problems and potential solutions to the recovery of the Pakistan energy sector are all well known. What is needed now is leadership—that is, a lead authority empowered to imple- ment and monitor the pace of reforms—and a road map of sustained actions. The energy sector is in serious crisis. Performance is far from satisfactory, and major issues need to be resolved. Key challenges include substantial and growing shortages of energy (6,000–7,000 megawatts [MW]—a third of peak demand for electricity; and up to 1,500 millions of cubic feet per day [mmcfd]— around 25 percent of peak demand for gas), as well as financial constraints that prevent the sector from financing all its costs. As a result, the sector relies heavily on government support—through subsidies amounting to about 2 percent of GDP from the budget to cover operating costs and funding for almost the entire invest- ment program. This situation is unsustainable, as energy shortages are constrain- ing the growth of productive activities (resulting in a loss of GDP of 2 percent or more a year, according to preliminary estimates), employment, and exports, while budgetary subsidies are diverting resources from other high-priority activities. building an efficient energy sector In recent years, the government (with donors) has prepared strategies and plans to address these issues. For example, the Asian Development Bank prepared a comprehensive assessment of the energy sector in 2010 (under the Friends of 18 Democratic Pakistan umbrella), analyzing the sector’s challenges and recom- mending ways to address them (ADB 2010). Similarly, in September 2010 the government prepared a power sector reform plan that aimed at addressing governance and efficiency issues, regulatory challenges, investment require- ments, and power sector finances. Implementation of these reform plans has, however, been incomplete at best. Pakistan’s per capita energy consumption is on par with other countries at a similar stage of development (Table 1.1), and energy access has expanded rapidly. The number of electrified households rose from nearly half of households (7.8  million) in 1996 to two-thirds (13.4  million) in 2006 and to three-quarters (19.0  million) in 2011.1 The greater access comes from a vigorous rural electrification program—financed by government resources— which the government continues to pursue, largely for social reasons. But while network expansion brings social and equity benefits, it has also increased the energy shortages. Although about 3,000 MW of generation capacity has been added since 2008, because of inefficiencies and the network expansion this addition has not reduced the overall shortfall—on the contrary, it has increased. The energy intensity of Pakistan is relatively high due to inefficiencies in not only supply (for example, public sector generation plants depict heat rates of Table Comparative energy use, 2011 1.1 Electricity Energy Fossil fuel energy Electricity access Energy use GDP per unit GDP imports, net consumption production (share of per capita of energy per capita (share of use) (share of total) (TWh) population) (KGOE) (PPP $ per KGOE) (current $) Selected countries in ascending order of GDP per capita Nepal 12 12 3 44 341 3.5 535 Bangladesh 17 71 42 41 209 7.9 675 Pakistan 24 62 94 62 487 5.4 1,017 Vietnam –11 70 95 98 681 4.7 1,224 India 25 73 960 66 566 6.0 1,375 Sri Lanka 44 44 11 77 478 10.7 2,400 Indonesia –84 66 170 65 867 5.0 2,952 China 9 88 4,208 99 1,807 4.2 4,433 Region South Asia 25 70 1,120 62 519 6.0 1,253 TWh is terawatt hour; KGOE is kilograms of oil equivalent; PPP is purchasing power parity. Source: World Development Indicators database. 12,000–14,000 British thermal units per kilowatt hour [btu per kWh], whereas efficient gas fired generation plants should achieve heat rates of 8,000 btu per kWh or below) and transmission/distribution, but also in the use of energy across different sectors. 19 Main Issues A large and growing shortfall of energy supply with a shift to a poor high-cost fuel mix Installed capacity stagnated over 2000–08, with only two additions (325 MW at Chashma nuclear plant in 2002 and 1,450 MW at Ghazi Barotha hydropower plant in 2004). By comparison, installed capacity increased from 3,711 MW to 5,201  MW (40  percent) in Bangladesh over the same period, and from 97,885 MW to 143,061 MW (46 percent) in India. After 2008, some 3,000 MW were added, taking capacity to about 22,500 MW by the end of 2011. Around 2,500 MW (10 percent or more of total installed capacity) is unavailable, how- ever, as lack of maintenance and rehabilitation has reduced the capacity of public sector plants. This capacity is further eroded in winter, when reduced water flows mean that about 5,000 MW of hydropower is unavailable. Total generation has therefore stagnated at 94–98 billion kWh since 2006/07, despite the additions; the newly installed capacity could have generated around 10  billion kWh, reducing the peak deficit by a comparable amount. The deficit arose because investments fell heavily. They were needed to enhance supply and upgrade the networks to cater for the increased demand. From a peak of 26 percent of total investment and 51 percent of public invest- ment in the mid-1990s, the share of energy (including power) investments had declined to 4 percent and 26 percent, respectively, by 2009/10. During this period, private investment (except for expenditures to complete plants that were initiated during the 1990s) was essentially zero. A number of factors led to the fall in investments. Some of the reasons why investments declined are historical, but they are still relevant and continue to impact on investments: • While there is consensus on the need to develop hydropower resources, actual investments are often impeded by concerns of the constituents over the distribution of water resources that would be made available through large reservoirs. This issue is termed a “trust deficit� in the Bank’s water sector strategy report (Briscoe and Qamar 2005). Essentially, the provinces are concerned that they are not receiving their due allocations under the Water Accord of 1991 and have reservations about the efficacy of federal building an efficient energy sector agencies (such as the Indus River System Authority) responsible for imple- menting the accord, the mechanisms set up for monitoring such allocations (for example, the telemetry systems installed by the Water and Power Devel- 20 opment Authority [WAPDA] over the last decade), and the operations of the entities responsible for managing water rights and allocations (for example, WAPDA). • The climate for private investment was hurt in the late 1990s. Pakistan went through a rather acrimonious debate with the sponsors of an independent power producers (IPP) program involving threats of, and actual, cancella- tion of licenses of some IPPs; accusations of corruption in the IPP program; and a widespread perception that those projects entailed higher power prices than required—and therefore a reversal of such programs was in order. More recent efforts to mobilize private investments for power generation have thus not attracted major international investors. The government recognizes that the public sector alone cannot finance the large investment requirements, and so may need to review the incentives for private sector participation to avert any further shortfalls in private investment in power. • The surplus that emerged following the commissioning of IPPs in 1997– 2001 led to some complacency in government circles. They felt that invest- ments to enhance capacity were not really a high priority. Given that power generation projects have long lead times and that power demand follows income growth with a lag, the country is still facing the impacts of this attitude—and the resulting decline in investments. The shortfall in supplies (during peak periods) widened over 2008–11, from around 4,000 MW to about 7,000, or about one-third of peak demand (Figure 1.1). The shortfall is essentially managed by load shedding—rotating Figure Demand has risen faster than capacity, 2008 –11 1.1 Hydro Thermal (WAPDA) Thermal (KESC) Nuclear Peak demand Actual production 25,000 20,000 Megawatts 15,000 10,000 5,000 0 2008 2009 2010 2011 Source: World Bank staff estimates. shutdowns of supply feeders. These disruptions are hurting industrial, com- mercial, and other productive activities, and their impact is estimated at around 2 percent of GDP (Ministry of Finance 2012). 21 The historical pattern of development—which has yielded a suboptimal energy mix and heightened the economy’s vulnerability to world market oil prices— has increased generation costs by three to four times. Generation costs account for around two-thirds of power supply costs. The generation mix has shifted from two-thirds hydro and one-third thermal in the 1980s to only 30 percent hydro and 70 percent thermal today. Also, there has been a shift from domestic low-priced gas to imported, higher priced, dirtier furnace oil (Figures 1.2–1.4). Figure Generation costs 1.2 Fuel Operations and maintenance Capital 20 U.S. cents per kilowatt hour 15 10 5 0 CCGT-LNG/ Diesel New New Hydel Nuclear Wind imported gas engine–fuel supercritical stream RoR oil coal– red –fuel oil boiler Note: Estimated cost of power generation alternatives for residual fuel oil is $710 per ton and for imported coal is $120 per ton at delivered prices. Source: Tetra-tech and Hagler Bailly Pakistan 2012. Figure Power generation fuel mix, 1990 –2011 1.3 Hydro Gas Oil Coal Nuclear Imports 100 Billions of kilowatt hours 75 50 25 0 1990–91 1995–96 2000–01 2005–06 2010–11 Source: Pakistan Energy Yearbook various years. building an efficient energy sector Figure With stagnating supply, gas diverted from power, 1990 –2011 1.4 Domestic Commercial Industry Cement Fertilizer (feedstock) Fertilizer (fuel) Power Transport (compressed natural gas) Unaccounted for gas 22 4 Billion cubic feet per day 3 2 1 0 1990–91 1995–96 2000–01 2005–06 2010–11 Source: Oil and Gas Regulatory Authority. The cost of generation of existing hydro plants is PRs 1.6 per kWh and for new hydro plants it is estimated to be between PRs  2.50–2.60 per kWh; by com- parison, generation cost of gas based combined cycle plants is PRs 4.50 per kWh, nuclear PRs 4.20 per kWh, coal PRs 8.10 per kWh, fuel oil PRs 16.75 per kWh, and diesel around PRs 21.00 per kWh. By moving to thermal generation, Pakistan is not benefiting from the cheapest supply sources—hydro and coal (see Figure 1.2). Some of these consequences are self-inflicted, as invest- ment decisions have been held up due to disagreements between provinces on water-resource use and mining rights. Even within oil and gas, reduced gas availability ensures that power generation is not based on the cheapest fuel sources—in fact, it is based on fuels (fuel oil, diesel) that are the most expensive and volatile in price movements. Nor is the allocation of lower cost gas for power generation based on least-cost criteria: while public sector plants installed 15–20 years ago receive gas, some private plants—which were installed in the last five years and incorporate more modern and efficient technolo- gies—do not. Hence, there is an overall inefficiency. The transition to thermal generation (and within thermal, from gas to fuel oil) has thus increased gener- stagnating supply, gas diverted from power, 1990 –2011 ation costs by three to four times; also, power generation costs move in tandem with fluctuations in international oil prices. While natural gas can significantly reduce generation costs in Pakistan, in recent years the available gas has been reallocated to other activities and sectors— at the expense of power generation. Pakistan attracted investment in ­ gas exploration in the 1990s, raising proven reserves more than 50 percent— to 32 trillion cubic feet (Tcf)—over 1996–2006. Gas production doubled to 3,800  mmcfd and sales to about 3,347  mmcfd. Much of this additional gas was allocated for power ­ generation—rising from 511 mmcfd during 1995–96 to 1,343 mmcfd during 2005–06. But this success was not sustained: reserves have since fallen (27 Tcf in 2011), production and sales have stagnated at about 4,000 mmcfd and 3,400 mmcfd, and the allocation for power generation has 23 fallen to 924 mmcfd. By the end of June 2011, gas demand had increased to 6,000 mmcfd. Despite stagnant supply, another 1.8 million consumers were added over 2006–11, bringing the total to 6.2 million. In parallel, and partly due to the large expansion of the networks, the performance of the gas utilities has deteriorated. Unaccounted for gas rose from 188 mmcfd in 2006 (about 7 percent of production) to 300 mmcfd in 2011, or more than 10 percent of available gas (global best practice is 1 percent). Also, gas has been gradually diverted from power. Indeed, power generation receives about 400 mmcfd less gas today than in 2006. Thus, only 26 percent of power generated in 2011 came from gas, down from 41 percent in 2006. The high costs are also driven by high transmission and distribution losses: 25 percent in 1996, 24 percent in 2006, and still 24 percent in 2011. But the distribution companies (DISCOs) show wide variation in losses: from 10 per- cent for Islamabad Electric Supply Company (Table 1.2) to 35  percent for Peshawar Electric Supply Company (PESCO) in 2011. Another issue is that the DISCOs do not collect all the bills issued—of the PRs 705 billion billed in 2010/11, some PRs  83  billion (12  percent), or nearly $1  billion, was not Better distribution companies have reduced losses to good-practice Table levels among developing countries, 1999, 2006, and 2011 1.2 (percent) Distribution company 1999 2006 2011 Islamabad Electric Supply Company 13.3 9.7 Gujranwala Electric Power Company 16.5 10.2 12.0 Lahore Electric Supply Company 20.2 13.1 13.3 Faisalabad Electric Supply Company 12.7 11.6 11.2 Multan Electric Power Company 24.1 20.5 18.3 Turkish Electricity Distribution Company, Istanbul 16.2a 11.0b North Delhi 27.1 12.5 Eastern Power Distribution Company of Andhra Pradesh 12.0 7.0 Dakshin Gujarat Vij Company 24.3 12.1 Edenor, Buenos Aires 10.2 11.1 12.6 PLN, Jakarta (Java only) 17.5a 8.8 Source: NEPRA (Pakistan); TEDAS (Turkey); Edenor (Argentina); PFC (India); PLN (Indonesia). a. Data are for 2005. b. Data are for 2008. building an efficient energy sector collected. Again, there is wide variation—four or five of DISCOs have achieved collection rates of 95–98 percent, while others have rates of just 60 percent. 24 Boosting energy efficiency could be one of the most cost-effective ways to narrow the power deficit in Pakistan and to address both peak load and energy shortages. Current “low-hanging� options with available technologies range from improvements in lighting and air-conditioning in buildings to street lighting and water pumping in municipalities—and waste heat recovery in industries. Although the importance of energy efficiency has been recog- nized in Pakistan, and Enercon was established in the early 1990s, there has been relatively limited implementation of actions on this front (such as large- scale deployment of energy efficient lamps). As a result, the vast potential for energy savings across the buildings, industry, transport, and agriculture sectors, estimated to be in the range of 20 to 30 percent, remains largely untapped. Strained finances Power tariffs (despite recent rises) are still far short of supply costs. The aggre- gation of tariffs determined by the National Electric Power Regulatory Author- ity (NEPRA) can be taken as a proxy for full cost recovery. The tariffs notified by the government represent the reduced cost to consumers. The difference represents a steep financial shortfall relative to the cost of supply (Figure 1.5). At its peak (up to about July 2010) the shortfall was nearly 35 percent of supply costs; for the 2011/12 determinations, it was still around 25 percent of total supply costs. A major part of the current shortfall comes from the fact that tariffs were not adjusted between November 2003 and February 2008, while Figure Progressive tariff adjustments, February 2008 –June 2012 1.5 NEPRA-determined estimated average tariff Government of Pakistan–noti ed estimated average tariff 13 Average tariff, PRs per kilowatt hour 11 9 7 5 3 February January January January January June 2008 2009 2010 2011 2012 2012 Source: World Bank staff estimates. supply costs rose sharply—oil prices (around 75 percent of supply costs) peaked at $150 a barrel in June 2008. While the government has aggressively adjusted tariffs since 2009—and changes in supply costs are promptly incorporated in consumer tariffs—the overhang from 2003/04–2007/08 has not been overcome. 25 The difficult financial picture is further complicated by uniform national tar- iff fixed at the cost of the most efficient DISCO, even though supply costs vary widely by province and DISCO. The government makes tariffs uniform by noti- fying the lowest determined tariff for each class of consumer to all DISCOs. The sum total of notifying a uniform national tariff equivalent to the costs of the most efficient utility implies that all other utilities automatically receive a subsidy through the tariff differential subsidy (TDS). In 2012, this amounted to PRs 465 billion (about 2 percent of GDP). Another contributing factor is low collections. Even if tariffs are insufficient for cost recovery, DISCOs should collect all the revenues they bill. However, sub- stantial accumulated arrears or receivables have developed within some DISCOs (discussed above). Due to inadequate tariff adjustments and insufficient cost recovery, most DISCOs continue to suffer financial losses and are compelled to defer investments for system enhancement, efficiency improvements, and the like. A tariff structure that does not promote energy conservation has also become a major barrier to pushing demand-side energy efficiency in Pakistan. Finally, NEPRA establishes performance targets for each DISCO, but some DISCOs do not meet them and thus require additional support.2 (Other DISCOs, however, achieve good levels of technical and financial performance.) This lack of cost recovery is a huge burden on the federal budget, aggravat- ing macroeconomic imbalances. The government provides subsidies from the budget for various (Table 1.3). The largest component of budgetary subsidies is the TDS. Since 2002/03, the government has kept power tariffs below the cost of supply3 and committed to pay the difference as a TDS. Yet the volume of such subsidies is unsustainable. The shortfall in DISCO revenues—which the government finances as TDS—has amounted to PRs 250–400 billion (roughly $3.0–4.5 billion) annually in recent years. The lower level is achieved in years when the government does not take actions to offset previous years’ accumu- lated deficits and payment shortfalls. In years when such overdues are also tackled, the level of TDS has approached the higher end of the range. The TDS has also been very unpredictable. Each year, the budgeted level is exceeded by a factor of three or four. This places a high burden on the fed- eral budget—which cannot be s ­ ustained—and adds great uncertainty to the budgeting process. building an efficient energy sector Table Federal government subsidies, 2008/09–2012/13 1.3 (PRs million) 26 2008/09 2009/10 2010/11 2011/12 2012/13 Budget Revised Budget Revised Budget Actual Budget Revised Budget Power sector 88,412 111,640 66,703 179,526 87,317 334,927 147,288 464,256 185,287 Ex-WAPDA DISCOs 74,612 92,840 62,903 147,005 84,000 288,927 122,700 419,018 134,970 Of which Tariff differential 65,000 82,000 10,000 77,000 30,000 238,827 50,000 412,018 120,000 KESC 13,800 18,800 3,800 32,521 3,317 46,000 24,588 45,238 50,317 Of which Tariff differential 12,000 17,000 2,000 31,700 2,000 46,000 24,000 45,000 50,000 WAPDA is Water and Power Development Authority; DISCO is distribution company; KESC is Karachi Electric Supply Company. Source: Budget documents (various years); World Bank estimates. Finally is the issue of “circular debt.� With revenue and resource shortfalls, the DISCOs build up arrears in payments to the National Transmission and Dispatch Company (NTDC). The arrears force NTDC to delay payments to its power producers (including WAPDA, for hydropower supplied to NTDC), which then build up arrears to their fuel suppliers, refineries, and so on. This is often referred to as circular debt, though the arrears buildup is one- directional—from consumers to DISCOs and transmission companies, gener- ators, and ultimately to fuel suppliers. The Ministry of Water and Power defines circular debt as the bills unpaid by NTDC/Pakistan Electric Power Company (PEPCO) to other energy suppliers. Based on this definition, the ministry estimated circular sector debt, as of June 30, 2012, at PRs  461  billion.4 However, this is only “current� circular debt, as it does not include debt swaps or other financing that the government resorts to for clearing arrears. For exam- ple, PRs 142 billion of term finance certificates were issued during 2011/12; similarly, PRs 302 billion of previously accumulated DISCO debts was consol- idated into government debt. Total liabilities of the sector as of June 30, 2012, could thus be around PRs 600 billion. The Planning Commission estimated the circular debt (as of September 2011) at PRs  270  billion for PEPCO and PRs 27 billion for KESC, for a total of PRs 297 billion. It also reported that circular debt is increasing by PRs 30.5 billion a month because of shortcom- ings in NEPRA’s method to determine actual cost of service, underbudgeting of subsidy payments, and failure to collect revenues. 5 The large and growing subsidy requirement, along with circular debt, has two important impacts. First, it crowds out borrowing by the power and other energy sector entities for investment in new capacity and blights the further development of existing projects. If the amount of subsidies paid from the budget during 2007/08–2011/12 (PRs 1.2–1.4 trillion) had been invested in power generation, it could have financed 4,000–6,000 MW of new hydro or thermal capacity. This should be compared with the 3,470 MW added over 2008/09–2010/11 and the need for as much as 8,000  MW today. Second, 27 circular debt creates severe liquidity shortages in the sector as a whole, such that suppliers refuse to provide fuel, generating plants stay idle, maintenance programs lag or are not implemented at all, and spare parts are unavailable when equipment breaks down (NEPRA 2011). But it must be stressed that the circular debt is a symptom of the inefficient and unsustainable energy system that can only be addressed once the underlying causes have been resolved— otherwise­it would just reemerge. Weak governance Overall sector governance remains weak with diluted responsibilities across ministries. The government, as owner of the power and gas utilities, exercises almost no control over the managements of public utilities. This is due partly to the limited capacities of the sector ministries, but it also high- lights the lack of a reform champion or leader, who is fully empowered to implement the reforms. Past efforts at sector reform had limited impacts or were abandoned in between. Even when PEPCO was empowered to fully imple- ment the reforms, the desired result was not achieved because it got bogged down in running the companies, rather than restructuring them (develop management capacities, systems, and human resources) and preparing them to operate as autonomous, commercial entities. To address these governance and high-level leadership challenges, the Friends of Democratic Pakistan report recommends merging the sector ministries and regulators and consolidat- ing the authority and responsibility for completing the reforms in one office (a senior energy advisor). Equally, the government has not announced its vision and strategy for the energy sector. As a result, investment decisions are not driven by rigorous analyses (such as least-cost planning). Over time, the capacity of the Plan- ning Commission and the utilities to conduct such analyses has eroded. The process for approval of public sector projects has also been undermined. Operational and technical performance can be improved across a wide spectrum, in technical areas (losses, voltage and other indicators of service quality), commercial areas (collections), and other areas. While some util- ities in the power sector have achieved operational and technical perfor- mance standards that compare favorably with utilities elsewhere, most fall far short of these standards. The key reasons the utilities are not complying with building an efficient energy sector mandated or internationally accepted levels and standards of performance include: • The sector’s unbundling—which was designed to improve performance 28 gradually by, for example, enhancing private sector participation, increas- ing commercial pressure, and promoting greater accountability—has not been completed. At times, pressures from within the utilities have fore- stalled the process;6 at other times, action plans or programs prepared with inputs and the consensus of all stakeholders have not been implemented.7 • There is limited progress in operating the sector on commercial princi- ples and in assigning responsibility (along with accountability) to boards of directors and managements of the utilities. The government is involved heavily in the utilities’ day-to-day operations, with two effects: boards of directors and managements have no incentive to perform their functions efficiently or to the best of their ability, and the government cannot hold them accountable for improving the utilities’ performance. • In 2009/10, the government signed performance contracts—initiating a process for holding the utilities accountable for performance—with all the DISCOs. But the contracts were not extended to subsequent years, and an important opportunity for strengthening governance and accountability was missed. • Managing the power sector along commercial lines was introduced unevenly, creating efficiency gaps, as illustrated by the use of IPPs. The solicitation of IPPs in the 1990s was based on a fixed price, and the government guaranteed the offtake of power by NTDC or WAPDA and fuel supply by Pakistan State Oil or gas companies to the IPPs. This approach was appropriate at the time, as procedures to select IPPs based on competi- tion on the price of power had yet to be developed. Also, direct contracting between IPPs and power consumers (as merchant plants or for part of their output) was not standard industry practice. However, both these concepts are now well established and are used for most IPP solicitations. Pakistan’s IPP policies have not, though, evolved to introduce or apply these concepts. As a result, private sponsors are not required to compete on the price of power that they offer, and the government continues to take on contingent liabilities—as guarantees for the offtake or fuel-supply obligations of their public sector counterparts. Other options for improving operational and technical performance include decentralization of responsibilities (for the DISCOs) to the provinces—and privatization of the utilities. The former would require significant strength- ening of the provincial governments’ capacity to manage the utilities. While privatizing the utilities remains a primary objective of the government, it needs to be recognized that in the current ­situation—the severely constrained financial situation of the power utilities—it would be difficult to attract private sector interest in owning and managing those utilities without some form of discount on the asset values and share prices. Since some privatization transac- tions have, in the past, been overturned due to such discounts, a consensus on 29 the approach for privatization needs to be developed before such transactions can be taken to market. Policy Options High-level leadership and management of the reform program is required. While addressing the challenges and constraints facing power and overall energy development requires actions across several fronts, strengthening governance and leadership should be an overarching objective. The govern- ment needs to establish a high-level and fully empowered structure (which could be a single ministry or a dedicated task force reporting to the Council of Common Interests or the Cabinet) to manage the transition. The respon- sible ministry or task force would need to be empowered to act on behalf of sector institutions, take the required administrative actions, address conflict- ing objectives, priorities, and points of view, and provide regular updates to the Council of Common Interests or the Cabinet. The ministry or task force would also need resources to conduct critical analyses and reviews and be guided by the vision or goal that the government e ­ stablishes—for example, for the sector for the next 15–20 years. Reduce supply shortfalls The decline in investments has to be reversed and investments need to be prioritized through long-term and least-cost analysis. Enhancing supply and overcoming the current shortages requires significant investments over the medium and long term. The focus of these investments has to be on low-cost sources of supply—particularly hydro, coal, and gas. Energy and power invest- ments, because of their long gestation and large financing requirements, need to be carefully designed and implemented, requiring integrated energy planning as a forecasting tool, with the capacity to prepare and update forecasts or plans. 8 Such forecasts for the power sector were traditionally prepared by NTDC, with the Planning Commission’s Energy Wing providing the macroeconomic and multisector framework.9 Over time, the capacity of NTDC and the Energy Wing to do this has eroded. The Asian Development Bank has provided an Integrated Energy Model to the Planning Commission for preparing comprehensive forecasts, impact analyses, and the like, though input requirements for the model are high. Whatever the mechanism, there is a need to develop and sustain capacity in the utilities (including NTDC) building an efficient energy sector and the government for preparing comprehensive, integrated energy develop- ment plans and for updating them regularly. 30 Investments in hydropower generation need to be accorded the highest priority. This requires enabling WAPDA to finance part of its investment through internal cash generation, according priority to WAPDA in the alloca- tion of official foreign currency financing, and strengthening its capacity to access domestic and international capital markets. More effective use of existing generation capacity must also be ensured. Much of the installed capacity in Pakistan is not available or only partially available— primarily due to a lack of maintenance, but also because most of the public sector plants were installed 20 or more years ago and now need to be replaced or rehabilitated. Rehabilitation and upgrading of such plants, including their conversion to coal, should be accorded a high priority—particularly for meeting the growth in demand over the next two to three years. Continuing efforts are required to develop domestic resources—coal, gas (conventional, shale and tight gas, liquefied natural gas, and pipeline imports) and renewables—as inputs for power generation. The success of earlier policies in promoting exploration and development of gas high- lights the need to further improve incentives. While the government has enhanced the producer price for natural gas under the 2012 Petro- leum Policy and announced premiums for shale and tight gas develop- ment, it should also recognize that the gas pricing policy places a cap on the reference (fuel oil or crude) price, which is well below current world market prices. These imperatives for enhancing domestic gas supplies (and the policy measure to be used for encouraging further exploration—that is, further adjustments in the parity price) need to be weighed against the implications of such pricing policies. Specifically, an increase in the party price would reduce the comparative advantage of gas-based generation projects—and may ultimately defeat the purpose of lowering power generation costs. Regional energy trade offers opportunities for reducing the supply deficit. Pakistan can supplement domestic energy (power as well as gas) supplies through imports. There is a strong complementarity between the hydropower generation capacities of Central Asian countries, which peaks in summer, and seasonal demand patterns in Pakistan. Imports from Central Asia can contribute toward meeting the peak summer demand in Pakistan; it could also be a source for low-cost power for the country. Similarly, power imports from India—because of the close proximity of high consumption areas in Pakistan (such as Lahore and Faisalabad) to India—could alleviate the shortages in those areas. Efforts need to be made to expedite such projects. Improving energy efficiency is a “least-cost� option for curtailing the deficit, 31 investment requirements would be reduced correspondingly. This requires increased efforts by Enercon, to pursue strategies to scale up demand side energy efficiency measures across various sectors, in coordination with WAPDA, DISCOs, private sector entities, service providers, local financial institutions, and other stakeholders. There are significant opportunities for tapping into energy savings across the buildings (public and private/residen- tial), industry, transport, and agricultural sectors, which could be achieved through a combination of policy-based and market-driven interventions and associated financial incentive tools and capacity-­ building measures. They include robust energy-­ e fficiency building codes, minimum energy performance standards for appliances, energy service companies (ESCOs), energy audit capacity building, and dedicated energy efficiency credit lines and funds. Global experience can be used and adapted to apply practical, }effective, and proven delivery models in Pakistan, which could go far in improv- ing energy efficiency and thus mitigating the energy shortages and crises, especially in the short term. Improve sector finances A framework is urgently needed to reduce the across-the-board subsidies and the sector’s financial deficit. Key actions to improve sector finances include moving to cost-covering tariffs—for example, in three years, along with targeted subsidies for life-line consumers; eliminating the TDS and address- ing, in parallel, the issues arising from the policy of uniform national tariffs; revamping the tariff notification process, to make it (essentially) automatic; and improving the DISCOs’ collection performance. • While the government has increased tariffs substantially since 2007/08, these actions are seen as ad hoc measures to reduce the sector’s deficit at any one point of time. A firm commitment to implement cost-covering tariffs over the medium term—along with programs to shield the poor from further tariff increases—is therefore strongly recommended. • In the absence of an announced timetable to eliminate the TDS, the utilities face no compulsions to reduce costs or increase collections, as the perception (and most often, the reality) is that the government will cover the deficit. Announcing a firm timetable for eliminating the TDS is there- fore a high priority; it will also mean savings of PRs 300–400 billion a year in the federal budget. The first action under this timetable should be to raise tariffs to 85 percent of the Minimum Determined Tariff for each building an efficient energy sector consumer category, starting July 1, 2013. The government will, in parallel, need to address policy issues to maintain uniform national tariffs. • Various mechanisms could be used to reconcile the differentiated 32 cost of supply with this policy. They include enacting surcharge and subsidy schemes, using the bulk sale and power purchase price to equate consumer level tariffs, allowing some DISCOs to charge the actual cost of supply (which implies very small variations in tariffs), and directly subsidizing the remaining DISCOs for supply costs in excess of this level. • Also, the government claims it notifies tariffs at the level of the lowest cost DISCO. This principle is, however, not always followed. The Planning Commission has estimated that if the government adjusted all tariffs to those of the lowest cost DISCO, the DISCOs could achieve PRs  98  billion in additional revenues, and the TDS requirement (PRs 592 billion) could be reduced by about 15 percent. • The tariff notification process must also be revamped. First, the NEPRA Act, or NEPRA’s Tariff Rules, should be amended to require that NEPRA complete its determinations within six weeks after petitions are filed; currently it takes six to nine months. The time taken by Ministry of Water and Power to notify tariffs adds further delays. The cumulative impact (such as penalties and financial costs that the DISCOs have to incur, which NEPRA does not recognize or accept) is estimated to be around PRs 81 billion for 2012/13. • Finally, requiring the DISCOs to improve performance—notably, collection of the amounts that they bill to ­ consumers—will reduce the subsidy requirement, and improve the DISCOs’ cash flow. For 2012/13, the DISCOs will lose PRs  85 billion in revenues solely because they do not comply with NEPRA’s target for collections. Improving the DISCOs’ collection performance will, however, not be possible unless the govern- ment enforces strict discipline on all ministries, departments, and agen- cies—both federal and provincial—to pay their electricity bills on time. In addition to improving collections, the DISCOs and NTDC need to adhere to commercial norms for payments to power suppliers—particularly for payments to WAPDA. Other actions that should be considered include: • As an interim step, publish the TDS amounts (both budgeted and actual) separately for each DISCO to identify the DISCOs and geographic areas benefiting from this subsidy. • Strictly use the most efficient plants first, and allocate natural gas to them (as power generation costs account for the bulk of power supply costs). • Improve the utilities’ operational performance and manage consumer demand more efficiently. This requires reducing theft and technical losses and improving metering. A 1 percent reduction in losses would generate PRs 9–10 billion in additional revenues or reduce costs by a similar amount. Improving efficiency in supply and at the consumer and end-user level would 33 also reduce power supply deficits and improve reliability of the system, though these approaches may face financial, institutional, and market barriers that need to be addressed through appropriately designed policies and programs. Key components of an energy efficiency improvement roadmap should include comprehensive legislation to promote energ y eff iciency, strength- ened institutional set up and coordination, loss-reduction programs for DISCOs and public generation companies, demand-side management through DISCOs, building codes and equipment efficiency standards, and promotion of energy savings performance contracting by ESCOs. Such a strategy for scaling up energy efficiency should be accompanied by a comprehensive and targeted consumer awareness program and capacity building of a range of stakehold- ers from energy auditors, to ESCOs and to financial institutions, which will help mainstream energy efficiency as an integral component and business line within the sector framework and operations. Stronger energy efficiency policy signals will also attract the private sector, including appliance and equipment manufacturers, building contractors, industries, ESCOs, and local commercial banks. Strengthen sector governance The government must announce its vision for the sector and designate a high- level champion or leader to manage the reforms. This announcement has to be complemented by amendments (wherever required) in the current rules of business, to enable the high-level champion to make all the required decisions on behalf of the sector ministries and agencies to implement reforms. It also requires decisions on such issues as: • Merging the Ministry of Water and Power and Ministry of Petroleum and Natural Resources into a single Ministry of Energy; merging NEPRA and OGRA into a single Energy Sector Regulator. • Strengthening regulatory framework—updated rules and regulations, capacity building of regulators, and so on. Other measures to improve sector governance include: • Introducing legislation, along with enhanced investigation and prosecution, to combat energy theft. This will also require enhancing, over time, the capacity of the DISCOs to investigate and submit cases for prosecution— and of the police and judicial system to handle a significantly larger number of cases. building an efficient energy sector • For all public sector companies, developing performance standards, which could be enforced through annual performance contracts or through more rigorous and timely monitoring of utility performance; ensuring compli- 34 ance with such standards; and introducing reward and penalty schemes for boards of directors and management that comply with (or fail to comply with) such contracts. The responsibility for monitoring these performance contracts could be assigned to the boards of directors of the companies or (to further enhance transparency) to NEPRA. In the latter case, NEPRA’s performance monitoring capacity would need to be enhanced. • Granting the public sector entities that can operate independently the required autonomy and accountability to operate along commercial lines. The government maintains strict administrative and operational control over all companies. Some companies, such as Islamabad Electric Supply Company, Faisalabad Electric Supply Company, Gujranwala Electric Power Company, and Lahore Electric Supply Company, can manage their affairs—including power purchase costs, investments, debt service, and operating costs—within the resource envelope provided by the determined tariffs. The government should cease intervening in the day-to-day management of these companies as a first step and hold the boards of directors and managements accountable for ensuring efficient electricity supply. Gradually, such delegation should be extended to other companies, when they demonstrate the ability to manage their affairs independently. The government needs to introduce market conditions for power, where commercial principles and trade among generators, distributors, and initially large customers would cater to all service requirements. Introducing market forces in the power sector should also be considered, which would absolve the government from responsibility for managing utilities, reduce government liabilities and subsidies, and allow the private sector to take over commercial activities (which it is best suited to perform). Introducing competition in the solicitation of new IPPs could be the first step. For this purpose, the policy and security package documents would need to be updated to allow for competi- tion on the price of power and energy, enable the level guarantees and con- tingent liabilities required to analyze specific investments, and amend the selection ­ criteria—to allow selection based on lowest energy cost, minimum contingent liabilities and commitments, and the like. The government will need expert advice and assistance, for amending the security package and selection process along these lines. The government has announced a new exploration policy for natural gas, for which there may be room to improve the incentive structure. The policy aims to enhance local gas production by attracting investments for exploration and includes an increase in the wellhead producer price. This up-front incen- tive is reportedly attracting interest from local and international companies. The government may be reluctant to offer the full parity price, as that may generate windfall profits for exploration companies. However, mechanisms 35 can be designed for the potential exploration companies to share some of the upside potential with the government. Without more domestic gas, Pakistan will need to import the required fuels, and thus the opportunity cost of forgoing gas development is high. In parallel, the government should consider introducing market mechanisms in the gas sector. They could involve encouraging exploration companies to seek customers independently and enabling them to transport that gas through the gas transmission companies via an open-access framework. The above three pillars for reforming the energy framework are mutually reinforcing. Adding low-cost generation and improving governance are essential elements of financial sustainability. And without financial sustain- ability it would be difficult to achieve good governance or attract the invest- ment needed for adding low-cost generation. Notes A. Ahmad, K. Saeed, M. Saqib, and L. Wang (Energy Unit, South Asia Sustainable Development Department) also contributed to this note. 1. However, according to household income and expenditure surveys (see Ministry of Finance 2011), 91 percent of households report electricity as their source of lighting. 2. Without government support to offset these performance failures, the DISCOs’ revenue and resource shortfall would become a charge on their equity. Applying this formula over time, a number of DISCOs erode all their equity. The support provided by the government can therefore be considered as recapitalizing those companies. 3. NEPRA-determined tariffs are taken to represent the cost of supply. 4. Data provided to the Asian Development Bank and World Bank team by the Ministry of Water and Power in its communication on September 29, 2012. 5. Planning Commission presentation on circular debt, September 7, 2011. 6. The most notable of such pressures is the incomplete transfer of human resources from WAPDA to the companies. A cadre of senior officials insists that because WAPDA employed them on a “common seniority� pool, they are entitled to serve in any of the companies that were cre- ated out of WAPDA. While this means their expertise can be available to all companies, it distorts the managerial incentives that would arise building an efficient energy sector once these officials opt for a specific company—and demonstrate their allegiance to that company alone. 7. Examples of such lack of implementation include the recommendations 36 of the Energy Sector Task Force and the September 2010 Action Plan prepared by the government. 8. For a recent exposition of the need for such planning, see Alahdad (2012). 9. The last comprehensive National Power Plan was prepared, with assistance from the Canadian International Development Agency, in 1994; an update (based on a limited set of inputs and assumptions) was recently prepared by NTDC. References ADB (Asian Development Bank). 2010. “Integrated Energy Sector Recovery Report and Plan.� Friends of Democratic Pakistan, Energy Sector Task Force, Manila. Alahdad, Ziad. 2012. “Pakistan’s Energy Sector, From Crisis to Crisis—Breaking the Chain.� Monograph Series. Pakistan Institute of Development Econom- ics, Islamabad. Briscoe, John, and Usman Qamar. 2005. Pakistan’s Water Economy: Running Dry. World Bank Report 44375. Oxford, UK: Oxford University Press. Ministry of Finance. 2011. Pakistan Social and Living Standards Measurement Survey 2011. Islamabad: Government of Pakistan. ———. 2012. Pakistan Economic Survey 2011–12. Islamabad: Government of Pakistan. Ministry of Petroleum and Natural Resources. Various years. Pakistan Energy Yearbook. Islamabad: Hydrocarbon Development Institute of Pakistan. NEPRA (National Electric Power Regulatory Authority). 2011. “State of the Industry Report 2010/2011.� Islamabad. Tetra-tech and Hagler Bailly Pakistan. 2012. “Pre-feasibility Study for Power Import by Pakistan from India.� World Bank, Islamabad. José G. Reis, Daria Taglioni, Charles Kunaka, and Nihal Pitigala Reinvigorating the 2 Agenda for Open Trade Pakistan’s trade performance is disappointing. Despite being in a very dynamic region, the country’s position in world trade has barely changed over the past three decades. The trade to GDP ratio showed only a very small increase in the last decade. Exports generally lack sophistication, and dynamism (as marked by entry and exit in foreign markets and new exports) has declined in recent years, coinciding with the reversal of trade policy to a less open stance since the mid-2000s. Open trade is central to any growth strategy—worldwide, no successful growth story has exclusively featured inward- looking growth. Restoring the country’s place in international markets is thus criti- cal for resuming growth, raising productivity, and creating jobs. To improve its trade competitiveness, Pakistan needs to simplify tariffs and trade regulations in order to reduce the anti-export bias; accelerate deep preferential trade agreements to encourage trade creation; fully normalize trade relations with India to benefit from growth there; and address logistical weaknesses to reduce trade costs. Trade is an integral part of Pakistan’s growth agenda. The country has failed to benefit from the trade bonanza in both South Asia and the world, and its position in global trade has changed little over the past three decades. While its world market share has declined over the last 20 years (despite a moderate recovery after the global crisis), those of Malaysia, Mexico, and Thailand have doubled, and China’s tripled. Aided by large reductions in trade barriers and technological advancements, developing countries— led by China and other emerging markets—have become major drivers of global trade. There is now little dispute that, in the long run, economies more open to trade show stronger economic growth and overall development performance, including export growth (Dollar and Kraay 2002, 2004; Michaely, Papageorgiou, and Choksi 1991; Winters, McCulloch, and McKay 2004). Pressure is growing on developing economies to improve productivity and restructure industry. World trade has undergone a dramatic transformation over the past decade and is increasingly characterized by global fragmen- tation of production leading to greater trade and a wider variety of traded goods and services—a trend that has largely eluded Pakistan. The evolving trade landscape, dominated by emerging economies like China and India, is reinvigorating the agenda for open trade likely to exact more pressure on firms in other developing economies, such as Pakistan, to improve productivity and restructure the industrial sector to enhance their competitiveness so that they can benefit from international market 38 opportunities. Steps to enhance domestic competitiveness are becoming more important—as is how they measure up to competitors’ actions. Pakistan’s recent trade performance is one of stagnation, as indicated by a very small increase in its trade to GDP ratio over the last decade.1 Peer countries have leapfrogged with high growth. Pakistan’s position below the predicted line indi- cates that it “undertrades� relative to countries at comparable GDP per capita incomes (Figure 2.1).2 What is striking, however, is that Pakistan’s average trade- to-GDP ratio for 2008–10 (bottom panel) was only 2.6 percentage points higher than a decade earlier (top panel), in contrast to the shares of its peers. In 1998– 2000, Pakistan’s trade to GDP ratio was barely lower than China’s and much higher than India’s. Ten years on, both China’s and India’s shares nearly doubled Figure Openness to trade 2.1 Trade to GDP ratio, 1998–2000 (percent) 250 Malaysia 200 150 Thailand Vietnam 100 50 China Turkey Pakistan Egypt, United States Bangladesh India Arab Rep. Brazil 0 5 6 7 8 9 10 11 12 GDP per capita, average over 1998–2000 (purchasing power parity, log scale) 250 Trade to GDP ratio, 2008–10 (percent) 200 Malaysia Vietnam 150 Thailand 100 India Turkey 50 China Bangladesh Pakistan Egypt, Brazil United States Arab Rep. 0 5 6 7 8 9 10 11 12 GDP per capita, average over 2008–10 (purchasing power parity, log scale) Source: Authors’ calculations based on United Nations Statistics Division (n.d.). while Pakistan’s remained largely stagnant, increasing only 2.6 percentage points, from 31.5 percent to 34.1 percent. Both other countries also showed higher rates of economic growth. 39 Over 1980–2011, Pakistan’s exports grew faster than the world’s but far more slowly than those of its South Asian peers. Indeed, Pakistan’s exports grew 13-fold, much less than the 30-fold in the rest of South Asia. During this period, the most intense growth of Pakistan’s exports was at the end of the 1980s, the start of the 1990s, and the mid-2000s. The end of the 2000s, however, saw a strong deceleration of export growth, following the drop in demand associated with the industrialized countries’ debt and economic crises, as well as slower growth in South Asia (Figure 2.2). Firm-level data show declining dynamism in export markets. Export entry and exit rates are slower than in peer countries, and new entrants barely offset the loss in exports by exiting firms. Each year, some 16–37 percent of firms exit and another 19–37 percent enter. At first glance, there seems to be sig- nificant destruction and creation of exporters in the data, but entry and exit rates are lower than in other lower middle-income countries (Figure 2.3). As a result, the effect on the number of exporters seems marginal, suggesting that the number of exporters is not expanding much and that Pakistan’s trade openness is stagnating. Dynamism decelerated sharply after 2004, as evidenced by the decomposition analysis of export growth in firm, country, sector, and product extensive and intensive margins presented in Figure 2.4.3 On average, fewer firms entered and exited the export market after 2004. Incumbent exporters gradually became more conservative in export markets, as well as sector and product experimentation. In Figure Export growth, 1980 –2011 2.2 Pakistan Low- and middle-income South Asia countries World 350 300 Index (1980 = 100) 250 200 150 100 50 0 1980 1985 1990 1995 2000 2005 2011 Source: Authors’ calculations based on United Nations Statistics Division (n.d.). reinvigorating the agenda for open trade Exporter entry and exit rates in Pakistan and peer countries, Figure 2004 – 09 2.3 Entry rate Pakistan Bulgaria Colombia Ecuador Peru South Africa 40 0.5 0.4 0.3 Percent 0.2 0.1 0.0 2004 2005 2006 2007 2008 2009 Exit rate Pakistan Bulgaria Colombia Ecuador Peru South Africa 0.5 0.4 Percent 0.3 0.2 0.1 0.0 2004 2005 2006 2007 2008 2009 Note: The above middle-income and medium-size countries were selected from the list of countries available in the firm-level database, which is much more limited than the ones for aggregate data. Source: Authors’ calculations based on World Bank (n.d. a). Figure Decomposition of annual export growth rates, 2002/03–2009/10 2.4 Firm entry Firm exit Country entry Country exit Sector entry Sector exit Product entry Product exit Intensive positive Intensive negative 60 40 Islamabad FaisalabadGujranwala Lahore Sheikhupura Sialkot Hyderabad Karachi Sukkur Multan Rawalpindi Quetta Peshawar 20 Percent 0 –20 –40 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 Source: Pakistan customs data. Islamabad FaisalabadGujranwala Lahore Sheikhupura Sialkot Hyderabad Karachi Sukkur Multan Rawalpindi Quetta Peshawar short, entries and exits by new firms, as well as existing firms in new countries, sectors, and products, all fell in the second half of the 2000s. Many important changes, such as the security situation and trade policy, 41 have taken place in Pakistan over the past decade, which may explain these trends in firms’ export growth. The fact that export growth decomposed by entries and exits at the firm, sector, and product levels have all declined sug- gests a decrease in the dynamism of Pakistan’s exports—and possibly a lack of incentives for the exporters to innovate. Could changes in trade policy have affected this performance? Main Issues The following impediments stand in the way of Pakistan improving its trade competitiveness. Reversion to protectionism Pakistan’s trade policy is reverting to protectionism. Over the last 15 years, it has seen a pendulum-like trajectory. Starting in the mid-1990s, the country embarked on a major trade liberalization program, which included reducing trade tariffs, simplifying the overall tariff structure, and abolishing nearly all remaining quantitative restrictions. Trade policy initially favored multilat- eral and unilateral trade liberalization and aimed at reducing government intervention, in a period when a relatively favorable macroeconomic environment supported trade policy reform and export expansion. In the mid-2000s, the pendulum started swinging back. New import substi- tution policies came in, with the pace picking up after the start of the global crisis in 2008 (Planning Commission 2011). Current trade policies have become more oriented toward providing extra protection to the processing margins of selected local products and producers. The Planning Commission (2011) reports that of the 906 manufactured products granted tariff exemp- tions, 91 percent benefited a single local monopoly producer, and 5 percent two producers, while smaller producers in particular were excluded from the exemptions’ benefits. This recent reversal has made Pakistan the world’s seventh most protected economy, thus risking further undermining its already poor openness. The Overall Trade Restrictiveness Index (OTRI) suggests that the country’s overall restrictiveness has increased. On the OTRI—which ranges from 0 in Singapore and Hong Kong SAR, China, to 16 in the Islamic Republic of Iran—Pakistan reinvigorating the agenda for open trade Box Overall Trade Restrictiveness Index 2.1 Measuring the restrictiveness of trade policy is often a complex task, especially given countries’ multifarious tariff and trade regimes. Standard indicators of restrictiveness do not capture widespread exemptions affecting 42 imports and other policy barriers and are not well grounded in theory (Kee, Nicita, and Olarreaga 2008). The Overall Trade Restrictiveness Index (OTRI) quantifies the uniform tariff that, if imposed on home imports instead of the existing heterogeneous structure of protection, would leave aggregate imports at their current level. It is a rigorous way to calculate and compare countries’ weighted average tariffs, with weights reflecting the importance of each good in total imports and the responsiveness of the import of each good with respect to the relative tariff. Its theoretical foundation, developed by Anderson and Neary (1996, 2003), was first applied empir- ically by Kee, Nicita, and Olarreaga (2008, 2009). Following Kee, Nicita, and Olarreaga (2008), the authors estimate the OTRI for Pakistan and other countries using the latest Center for Global Trade Analysis Project (GTAP) Data Base (version 8, base year 2007) and the current GTAP model (Hertel 1997), using applied tariffs and nontariff barriers. The GTAP Data Base provides a more comprehensive estimate of nontariff barriers than the measures in the United Nations Conference on Trade and Development’s Trade Analysis and Information System database, the latter of which are used in the calculation of OTRI published in World Bank (n.d. b). Figure Estimated Overall Trade Restrictiveness Index, selected countries 2.5 Indonesia Morocco Pakistan Vietnam Bangladesh Nigeria Egypt, Arab Rep. Senegal Sri Lanka Nicaragua Zambia Lao PDR Albania Guatemala China Ukraine Paraguay Malaysia Philippines India Bolivia Turkey Armenia Georgia 0 5 10 15 Overall Trade Restrictiveness Index Note: Green highlights some of Pakistan’s main competitors. Base year is 2007, except for Pakistan, which is 2002. Source: Authors’ calculations based on Center for Global Trade Analysis (n.d.). reached 9.9 in 2010, up from 9.0 in 2004. (Box  2.1 gives further details on the index.) While Pakistan’s OTRI is not South Asia’s highest, with a score of 9.9, it places Pakistan in the 88th percentile, among countries with the most restrictive trade policies (Figure 2.5). Overall trade restrictiveness developments 43 in Pakistan are possibly associated with the increasing complexity of national trade policy and with the reversal of some of the early progress toward reforms. Trade policy complexity The high degree of tariff escalation and complexity in the tariff structure are reflected in the effective rate of protection (ERP). By taking into account protection on both outputs and inputs, ERPs better represent tariff-generated transfers to producers than do nominal rates of protection, which are based on protection of outputs only. Protection on services industries is mostly negative, imports of most basic raw materials are relatively free of import restrictions, and products at an intermediate stage of production have moderate ERPs. By con- trast, import-substituting industries, such as beverages, tobacco, motor vehicles and accessories, and vegetable oils and fats, yield very high ERPs. These results reinforce the earlier assertion about the escalatory tariff structure in Pakistan and provide evidence for the instrumental role of exemptions and high duties to achieve indigenization objectives in several manufacturing activities, and the automotive sector in particular. Pakistan’s pattern of protection discourages the production of high value-added products. ERPs (taking into account protection on outputs and inputs) are typically exaggerated with low value-added activities, such that even low tariffs on final goods can lead to increased returns (Figure 2.6). Sectors Figure Effective rates of protection 2.6 450 Beverages and tobacco 350 250 Percent Leather products 150 Vegetable oils Motor vehicles 50 Mineral products 0 –50 0 25 50 75 100 Value added (percent) Source: Authors’ calculations. reinvigorating the agenda for open trade with high ERPs tend to have low value added, or conversely there is a higher incentive to produce low value-added goods. That sectors with higher ERPs tend to be domestically oriented demonstrates the inherent bias against 44 export-competing sectors. The first-generation reforms in the early 2000s cut heavily into the use of quan- titative restrictions as protective devices, though Pakistan still applies some nontariff measures. The most prominent is the Indigenization Program, with its discretionary control over new entrants to specific industries, because all potential entrants must negotiate with the Engineering Development Board (a separate agency set up in 1995 to manage the large number of industrial “local content� programs originally sponsored by the Ministry of Industry) and agree on an individual indigenization program and how it would be phased in over time. The administration of this—often competing—dual regime discriminates against the import of domestically available materials, amounting to a de facto local content requirement. According to Global Trade Alert, other non- tariff measures include bailouts and state aid, export subsidies, export taxes and restrictions, import bans, sanitary and phytosanitary measures, and trade-defense measures, such as antidumping measures, countervailing duties, and safeguard measures. Although not all these measures are barriers or protectionist, many do have a negative impact on export competitiveness, as rec- ognized in the recently announced Strategic Trade Policy Framework for 2012– 15 (Box 2.2). Poor trade facilitation and logistics A sound trade facilitation and logistical system can play a decisive role in attaining export-led growth—by reducing trade costs, upgrading service quality, improving connectivity between domestic and foreign markets, and moving up the supply chain. Yet Pakistan’s logistics are worse than India’s Box The Strategic Trade Policy Framework for 2012–15 2.2 The new Strategic Trade Policy Framework unveiled by the Ministry of Commerce of Pakistan for 2012–15 recognizes the impact of high tariff protection on export competitiveness and adopts a series of guiding principles to create a more competitive environment (Ministry of Commerce 2012). The framework lays out as guiding principles promoting competitive markets, catering to the changing needs of the economy, and giving consideration to consumer welfare. The measures include short-term tariff reductions for imports of capital goods, as well as medium-term measures such as establishing a Services Trade Development Council (to tap the country’s potential of export services) and an Export-Import Bank. Perhaps still missing is a clearer medium-term strategy to move toward reduc- ing complexity of the current tariff structure. and the South Asian and global averages, according to the 2012 Logistics Performance Index (Figure 2.7). South Asia lags behind all regions except Sub-Saharan Africa, faring in line with its per capita income, and performs marginally better than the low-income country group (Figure 2.8). 45 Despite its low rating, Pakistan’s Logistics Performance Index improved over 2007–12 (Figure  2.9). It picked up sharply in customs and infrastructure but declined in logistics competence. In customs and infrastructure, the public sector played a larger role than in logistics competence, which depends more on the private sector. Figure Logistics performance in India, Pakistan, and South Asia 2.7 India Pakistan South Asia Customs Timeliness 4 Infrastructure 3 2 1 Tracking and tracing International shipments Logistics competence Source: World Bank 2012. Figure Regional logistics performance, 2012 2.8 Upper middle-income countries Lower middle-income countries Low-income countries Logistics Performance Index score, 2012 3.00 2.75 2.50 2.25 2.00 East Asia Europe and Latin America Middle East and South Sub-Saharan and Paci c Central Asia and Caribbean North Africa Asia Africa Source: World Bank 2012. reinvigorating the agenda for open trade Figure Logistics performance, 2007, 2010, and 2012 2.9 2007 2010 2012 Customs 46 Timeliness 4 Infrastructure 3 2 1 Tracking and tracing International shipments Logistics competence Source: World Bank 2007, 2010, 2012. Improving the performance of Pakistan’s seaports and that of the core national trade corridor will benefit the rest of the economy. More than 90 percent of Pakistan’s international trade is transported by sea and passes mostly through Karachi. Pakistan has two other ports, Qasim and the recently developed Gwadar Port, which is near the border with the Islamic Republic of Iran. The three are largely feeder ports connected to hubs in the Middle East (Salalah) and Asia (Colombo and Singapore). Authorities have expressed interest in devel- oping Karachi in particular as a hub and Gwadar as a specialized bulk and tran- sit port. But Pakistan’s ports lie close to a major international shipping lane and in a region with competition. For instance, based on traffic volumes, Karachi ranks 75th globally, exceeded in size by eight other nearby ports. Developing the ports as hubs depends on drastically improving port operational efficiency or focusing on niche markets for specific products. In this regard, authorities propose developing Gwadar as a specialized port for containers but more importantly to handle bulk commodities and transit cargo for the Central Asian republics. Overland movement of cargo in Pakistan depends largely on road transport Sheikhupura Sialkot Hyderabad Karachi Sukkur Quetta Peshawar Islamabad FaisalabadGujranwala Lahore Multan Rawalpindi but faces challenges in quality of service and in border management (which has an interface between public and private sector agencies involved in goods clearance). These topics are now discussed in more detail. Road transport is the dominant mode of land transport, accounting for more than 80  percent of cargo volumes. Despite poor conditions of parts of the road network, Pakistan has some of the lowest road freight transport rates in the world (Figure 2.10), partly due to the industry’s structure. The Figure Selected transport prices, 2007 2.10 Togo–Burkina Faso South Africa–Zambia 47 France China United States Brazil Pakistan 0 2 4 6 8 Average road transport price (US cents per ton-kilometer) Source: Authors’ calculations based on Teravaninthorn and Raballand (2009). majority of trucks are operated informally as small fleets. In 2007, Pakistan had as many as 209,000 registered trucks, more than two-thirds of them rigid trucks. Most vehicles are old and highly fuel inefficient. Truckers keep rates low by maintaining old fleets and by overloading their vehicles. Competition and high operating costs cut heavily into their margins and hamper efforts to modernize their fleets. The Pakistan trucking market continues to evolve with an increasing number of formal companies running large fleets of trucks (more than 50 vehicles). The largest fleet operator is the state-controlled National Logistics Cell (NLC), which owns more than 1,500 trucks. The NLC, owned by the military, previously had a monopoly on Afghanistan transit traffic by road. The NLC suffered from a lack of capacity, and the market was opened up in 2011. Other large fleet oper- ators are petroleum companies, such as Shell, and dry-port operators, notably Sialkot (more than 70 trucks) and Faisalabad. These larger operators all provide modern services, including vehicle and shipment tracking. The independent truckers have also changed how they are organized. Sheikhupura Sialkot Hyderabad Karachi Sukkur Quetta Peshawar Islamabad FaisalabadGujranwala Lahore Multan Rawalpindi Previously they were contracted for individual shipments with rates negoti- ated directly with the shippers or their brokers, and they secured backhauls from the many brokers in the major cities. Over the past few years, some of the larger road haulage companies have begun contracting service to independent truck operators to supplement their own fleet. The government is currently implementing a trucking policy that seeks to modernize the trucking sector, albeit slowly (Box 2.3). While some of these initiatives may raise trucking prices, they also raise quality, especially reliability, which is especially critical for sophisticated reinvigorating the agenda for open trade Box New policy on trucking industry 2.3 In 2007, the government drafted a trucking policy, which is now under implementation. The policy is designed to serve as the basis for modernizing the trucking sector. It has several progressive provisions 48 that, if realized, would raise the trucking sector’s service quality. Key provisions of the policy include: • Enhancing access to finance. The policy seeks to make it possible for trucking concerns to access financing from financial institutions. This is being pursued by designating the sector as an “indus- try,� which under Pakistani law then enables trucking concerns to borrow from banks at less than commercial rates. Limited access to finance has long been an impediment to modernizing and replacing the current fleet of predominantly old trucks. The policy provides incentives for this to happen. • Motor vehicle registration. The provinces handle motor vehicle registration in Pakistan. It is thus usually difficult to obtain current information on the vehicle fleet. The policy proposes to establish a central depository for motor vehicles for nationwide maintenance of data. • Vehicle worthiness testing and certification. The policy provides regular tests of fitness of vehicles and their road worthiness certification. This will be accompanied by training and capacity building of staff involved in the tests. supply chains. Some market segmentation should therefore be expected, with the more sophisticated chains paying more. Customs has been modernizing border management for the past several years, though several challenges remain. Reforms introduced include the use of the Harmonized Code, creation of a Single Administrative Document (Goods Declaration), and electronic submission and processing of declarations at seaports and major land borders. These improvements have cut clearance times and increased the collection of duties and taxes. Until recently, there were two IT systems for submitting the Goods Declaration—the Pakistan Customs Computerized System (PACCS) and the web-based One Customs. In 2012, One Customs was adopted as the sole system and is now being rolled out across the country. This has helped address the problems of duplication of systems development posed by the previous situation. Inspection decisions are often accompanied by informal payments. The rate of physical inspection has been reduced but remains higher than it would be under an effective risk management system. The typical clearance time for cargoes with proper documentation is two days for imports and one day for exports, while that for other cargo is longer (USAID 2010). Customs has a risk management unit that developed some risk profiles and in theory selects the level of inspection applied for individual shipments. However, both the PACCS and One Customs systems perform limited functions, and neither has a risk management module. (PACCS randomly selects containers, but this can be overridden by an officer, creating opportunities for informal payments.) Although the procedures for handling transit cargoes have been simplified, 49 the time required for issuing transit permits is long, and the procedures for inspecting transit cargo are neither transparent nor efficient. Nor does Pakistan have a formal system of Authorized Economic Operators, although informal recognition is given to preferred traders who receive expedited clearance. Modern X-ray scanning has been introduced but only at the Qasim Interna- tional Container Terminal and only for exports to the United States as part of the Customs-­Trade Partnership Against Terrorism. Policy Options The following policy actions aim to stimulate export diversification and expansion. Simplify tariff and trade regulations The priority is to abolish the distorted and highly complex tariff structure. Domestic market protection through tariff protection generates an anti- export bias. Because Pakistan cannot influence world market prices, exporting firms gain no benefit from domestic tariff policies. By contrast, producers for the domestic market enjoy a price advantage over foreign competitors. This leads to market distortions that incentivize producers to focus on the domestic market. Simplifying the complex import tariff and tax regime would also foster certainty, increase transparency, and promote firm dynamism. The remaining regulatory duties also need to be eliminated—and the import tax regime liberalized. While many regulatory duties were removed in 2011, selective protection and concessions on inputs remain a source of economic inefficiency because valuable resources are being diverted—to less productive sectors that are protected from otherwise low international prices and away from the industries that have comparative advantage. Concessions that work as a de facto import licensing regime entail transaction costs to the firms that hope to benefit, to the government bodies that administer it, and to the Engi- neering Development Board. Because the information and transaction costs will tend to be lower relative to their output, the system probably benefits larger established firms and discriminates against small and medium-size enterprises. This “customized� domestic protection entails important costs and risks, including the risk that protected firms have a disincentive to become efficient, often relying on continued protection to remain in business, rendering a loss to reinvigorating the agenda for open trade consumers, and reducing economic efficiency. High duty rates and a regulatory duty imposed only on imports is an extremely inefficient way of cutting imports of “luxury� products and is a counterproductive method for generating govern- 50 ment revenue. The very existence of schemes granting protection to selected sectors and products increases the scope for smuggling and under-invoicing. In the short term, a road map should be outlined for gradually simplifying the tariff regime. It might include an immediate reduction of the top rate to 25 percent and a move to a transparent, three-band structure (25 percent, 10 percent, and 0 percent). In the medium term (three to five years), the government should consider phasing out regulatory duties and eliminating exemptions outside trade agreements and free zones. These steps would have important positive effects in the long run, with transitory adjustment effects on output and employment limited to some sectors and a relatively small impact on tariff revenues.4 Higher value-added export-oriented sectors, such as chemicals and chemical products and machinery and equipment, are likely to benefit even in the short run. Many lower value-added, import-competing sectors, such as processed foods, rubber and plastic goods, and other manufactur- ing, would adjust production downward in the short run, as would many services sectors that temporarily lose out as consumption adjusts to take advantage of lower prices, particularly in traded goods. The medium term should also see a move toward an eventual uniform 10 percent tariff. Estimates calculated for this note found that eliminating all exemptions and moving toward a uniform 10 percent tariff would raise tariff revenues 79  percent and total import tax revenue 36  percent while increasing import prices 1.4 percent. The full effects of reform would start to emerge over the medium term, as economic agents respond to the new incentive environment, shifting resources from less productive to more productive economic activities. Accelerate deep preferential trade agreements Accelerating implementation of deep preferential trade agreements and signing other agreements in that vein are key initiatives in expanding market access. Pakistan has to scale up efforts to benefit from the geographic advantage of being in a high growth region of the world. Signif- icant trade-creation effects are likely to originate from the relatively recent trade agreements with China and Malaysia, due to the deep nature of these agreements and their coverage beyond market access issues. By contrast, regional preferential trade agreements limited to market access are unlikely to bring benefits because of Asia’s deep, fast-paced integration agreements. Fully normalize trade relations with India 51 Fully normalizing trade relations with India, including opening the border, will facilitate deep forms of trade integration. The measures are necessary to benefit from India’s fast growth and to promote complementarities, includ- ing value-chain activities and investment potential. Our calculations using a gravity equation suggest that exports to India are 40  percent below their predicted potential. Similarly, Pakistan is the key missing market for India, underlining that both countries would gain greatly by normalizing their trade relations. These moves should aim at deep forms of trade integration and not be limited to market access. Given that trade is only now starting to be normal- ized, the focus should be on expediting measures to facilitate trade, building on the recently signed agreements on mutual recognition and visas, and improving infrastructure,institutions, services, policies, procedures, and market-oriented regulatory systems. Specific measures for further assessment, potentially on a joint basis with India, include: • Removing impediments at the border, especially Wagah–Attari, and along trade routes. A one-stop border post at Wagah–Attari would have a large demon- stration effect, as would inland container depots on either side of the border. There is also potential for transit agreements to link Pakistan to Bangladesh and to Nepal, and India to Afghanistan (similar to the one negotiated between Pakistan and Afghanistan); for associated infrastruc- ture supporting new trade routes; and for online payment systems. • Further integrating border communities. Populations along the border regions are among the poorest in both Pakistan and India, and so integrating their communities presents particular challenges. More localized initiatives to target them, including border bazaars and other measures to encourage cross-border trade, are yet to be fully explored. Some initiatives along these lines are already in place between Bangladesh and India. Address logistical weaknesses It is critical to improve the regulation of trucking and of the clear- ing and forwarding industries. A wide gap in the legal framework is a liability for goods as they move along the chain—truck operators, for instance, carry only third-party liability. (A draft law on logistical service providers, currently in Parliament, should help improve the quality of other logistical reinvigorating the agenda for open trade services.) Given the presence of numerous small players in this market, a solid legal framework for responsibilities and liabilities, including those of truckers, is paramount. Freight forwarders, most of whom are also clear- 52 ing agents, are the most important intermediaries in trade logistics in any country, and in Pakistan they urgently require capacity building. Improving customs and border management is crucial for lowering logisti- cal costs, particularly port operations and customs procedures. It is import- ant to reduce the current high dwell time for cargo, especially in ports. A unified and comprehensive customs system would benefit trade logistics, allowing the private sector to develop and invest in appropriate interfaces. Also critical is the introduction of an effective risk management system, a formal Authorized Economic Operators regime, and an expedited regime for transit shipments. These measures will allow more cargo to be cleared inland, reducing bottlenecks at seaports and land borders, making clearance procedures more effective, and allowing for greater cooperation between shippers and customs officials. As Pakistan is developing trade links to the Central Asian republics in the north, these systems would be strategic. Notes 1. The trade to GDP ratio is one of the most basic indicators of openness to foreign trade and economic integration. It weighs the combined importance of exports and imports of goods and services in an economy. The ratio gives an indication of the dependence of domestic producers on foreign demand and of domestic consumers and producers on foreign supply. There is a concave relationship between trade openness and per capita income: countries tend to trade more as incomes rise, but at a decreasing rate. 2. An ordinary least squares regression (using cross-country data from the World Development Indicators database) of trade/GDP on the log of per capita income (purchasing power parity dollars), its squared value, the cost of exporting (using a subindicator from Doing Business 2008 to 2010), and population predicts a trade share of 65 percent. The actual trade share of less than 32 percent is therefore extremely low. 3. Export growth is said to take place at the intensive margin when it occurs by selling more of the same products to the same markets, while growth at the extensive margin includes new product discovery, existing products sold to new markets, and new firms entering export markets. 4. An exercise conducted for this note shows that a reform that puts a ceiling for tariffs at 25 percent but retains all other features of the current tariff regime, including exemptions, would reduce tariff revenues 3.7 percent but import tax revenues only 1.8 percent, and import prices only 0.6 percent. References Anderson, James, and J. Peter Neary. 1996. “A new Approach to Evaluating Trade Policy.� Review of Economic Studies 63 (1): 107–25. 53 ———. 2003. “The Mercantilist Index of Trade Policy.� International Economic Review 44: 627–49. Center for Global Trade Analysis. n.d. GTAP 8 Data Base. Purdue University, Department of Agricultural Economics, West Lafayette, IN. www.gtap.agecon. purdue.edu/databases/v8/. Dollar, David, and Aart Kraay. 2002. “Growth Is Good for the Poor.� Journal of Economic Growth 7 (3): 195–225. ———. 2004. “Trade, Growth, and Poverty.� Economic Journal 114 (493): F22–F49. Hertel, Thomas W., ed. 1997. Global Trade Analysis: Modeling and Applications. New York: Cambridge University Press. Kee, Hiau Looi, Alessandro Nicita, and Marcelo Olarreaga. 2008. “Import Demand Elasticities and Trade Distortions.� Review of Economics and Statistics 90 (4): 666–82. ———. 2009. “Estimating Trade Restrictiveness Indices.� The Economic Journal 119: 172–99. Michaely, Michael, Demetris Papageorgiou, and Armeane M. Choksi. 1991. Liberalizing Foreign Trade: Lessons of Experience in the Developing World. Oxford, UK: Basil Blackwell. Ministry of Commerce. 2012. Strategic Trade Policy Framework 2012–15. Islamabad: Government of Pakistan. Planning Commission. 2011. “Pakistan: Framework for Economic Growth.� Government of Pakistan, Islamabad. Teravaninthorn, Supee, and Gael Raballand. 2009. Transport Prices and Costs in Africa: A Review of the Main International Corridors. Washington, DC: World Bank. UNCTAD (United Nations Conference on Trade and Development). n.d. Trade Analysis and Information System database. Geneva. www.unctad.info/en/ Trade-Analysis-Branch. United Nations Statistics Division. n.d. Commodity Trade Statistics database. New York. http://comtrade.un.org/db/. USAID (United States Agency for International Development). 2010. “Pakistan Trade Project (PTP): Dwell Time Study.� Islamabad. Winters, L. Alan, Neil McCulloch, and Andrew McKay. 2004. “Trade Liberal- ization and Poverty: The Evidence So Far.� Journal of Economic Literature 42 (March): 72–115. World Bank. 2007. Connecting to Compete 2007: Trade Logistics in the Global Econ- omy. Washington, DC. reinvigorating the agenda for open trade ———. 2010. Connecting to Compete 2010: Trade Logistics in the Global Economy. Washington, DC. ———. 2012. Connecting to Compete 2012: Trade Logistics in the Global Economy. 54 Washington, DC. ———. n.d. a. Exporter Dynamics Database. Washington, DC. http://data. worldbank.org/data-catalog/exporter -dynamics-database. ———. n.d. b. World Trade Indicators database. Washington, DC. www.world- bank.org/wti. Mehnaz Safavian, Sarwat Aftab, and Sarmad Shaikh Enhancing the 3 Business Environment For developing countries like Pakistan, where small and medium-size enterprises make up a large share of the economy, the quality of the business environment is a critical policy area for governments to drive economic growth, create jobs, and encourage private sector competition. Subnational analysis of the business environment finds that the regulatory and administrative burden of complying with business regulations varies considerably across cities. The policy focus can therefore benefit from designing, implementing, and analyzing reforms through a subnational lens to enhance that environment. At the provincial and district levels, such an approach can maximize the impacts of reforms; at the federal level, it can monitor, compare, and benchmark subnational performance. A country’s regulatory environment affects businesses at almost all stages of the business life cycle. Business registration regulations apply when entrepre- neurs decide to start an enterprise. Tax and customs regulations apply when firms produce and export goods. Contract enforcement laws apply in almost all areas of business operations. Business regulations, when properly designed and implemented, can be a major catalyst for establishing new businesses and expanding existing ones. Conversely, a regulatory environment that impedes businesses creates more incentives for entrepreneurs to seek loopholes and offer bribes—and thus breeds corrupt economic practices (The Economist 2012). In the private sector, such practices have broader implications on entrepre- neurship and innovation; in the public sector, on administrative efficiency and service delivery. The business regulatory environment is particularly relevant for small and medium-size enterprises (SMEs)—the key drivers of competition, economic growth, and job creation, especially in developing countries (World Bank 2013). Pakistan is no exception: SMEs are a key element of its economic landscape, and up to 99 percent of establishments employ 1–20 workers (Pakistan Bureau of Statistics 2005). Collectively, SMEs in Pakistan provide about 78 percent of nonagricultural employment, contribute almost 40 percent of GDP, and account for some 30 percent of manufacturing exports (Jamil Afaqi and Seth 2009). Given the importance of SMEs in the private sector, it is important to understand the quality of the business environment, identify opportunities Enhancing the Business Environment to make reforms with impact, and better position the economy to be more competitive in serving the domestic and external markets. This note high- lights the quality of Pakistan’s business environment, as measured by the 56 World Bank’s national and subnational Doing Business indicators. Reform options pinpoint the key areas where policy changes could have the most impact. And the benchmarking of regulations across the globe shows how Pakistan compares with its neighbors and global competitors. The note discusses the impacts of business regulations across Pakistan and suggests national, provincial, and local reforms in six key aspects of doing business: starting a business, dealing with construction permits, trading across borders, registering property, paying taxes, and enforcing contracts. The note uses international and subnational data — the former to highlight how the country compares with the rest of the world on national business environment indicators and reforms, and the latter to highlight how Pakistan’s cities compare with each other on good practice and poten- tial reform programs. But why emphasize national and subnational reforms? Some reforms fall exclusively under the federal government’s purview (such as registering corporations), while others are dealt with exclusively by the provinces (such as registering sole proprietorships). And for federal regulations, provincial and local implementation is critical to their effectiveness. National and subnational reforms are complementary and can have profound provincial impacts, so this note highlights the reform options at all government—federal, provincial, and ­ three levels of ­ district—all of which affect Pakistan’s business environment. The international comparisons draw from the Doing Business 2013 rankings, and the subnational comparisons are from the Doing Business in Pakistan 2010 report, because subnational reports are produced only at the request of a client govern- ment. In 2012, a review of three of the subnational indicators (starting a business, trading across borders, and dealing with construction permits) showed that there had been no changes or reforms since the 2010 report. Today, Pakistan is ranked 107 of 185 economies on the overall ease of doing business rankings, down from 76 of 181 in 2008 (Figure 3.1). Figure 3.2 shows Pakistan’s 2013 global rankings on each Doing Business indicator. A review of global comparators masks considerable variation across Pakistan’s business environments. This variation highlights the importance of subnational laws and regulations as well as the consistent implementation of those at national level. More important, a review of that variation in Pakistan’s provinces and cities can highlight the country’s good practices, the opportunities for peer-to-peer or government-to-government learning, and the potential for committed provincial Figure Ease of doing business, South Asia, 2008 and 2013 3.1 2008 2013 (improvement) 2013 (decline) 185 57 Rank 1 Sri Lanka Maldives Pakistan Nepal Bangladesh India Bhutan Afghanistan Source: World Bank 2013. Figure Pakistan’s global rankings across Doing Business indicators, 2013 3.2 Pakistan South Asian median 185 Covered in Doing Business in Pakistan 2010 Rank 1 Getting Paying Enforcing Registering Dealing with Starting Trading Resolving Getting Protecting electricity taxes contracts property construction a business across insolvency credit investors permits borders Source: World Bank 2013. and local governments to create world-class, competitive business environments in their areas. Local, district, and provincial reforms are important in many areas of the regulatory environment, though some reforms can of course take place only at the national level, as prescribed by the constitution or relevant legal framework. The 18th Amendment to the constitution, approved in 2010, devolved several powers, rights, and responsibilities to provincial governments and, in turn, to district governments. District governments are closer to the impacts of business regulations and are potentially better suited to design local reforms and to advocate for—and contribute ­ to—­ federal reforms. Several recent enhancements in the regulatory environment focus on imple- menting current reforms more effectively instead of attempting to introduce Enhancing the Business Environment new ones. Local reforms allow considerable peer-learning opportunities among districts. Several districts have better approaches to business regula- tions and more efficient implementation processes than others. Local reform 58 managers can learn from successful practices in other districts to expedite reforms in theirs. No single city does well on all indicators, and best practices vary by city (Table 3.1). It is therefore enough to start the reform process by introducing measures that have succeeded in other cities. In fact, Pakistani cities have a lot to gain from adopting the best regulations and practices that are working elsewhere in the country. A hypothetical city, “Pakistana,� adopting all the best practices identified in Doing Business in Pakistan 2010 would have ranked 69 of 183 ­ economies—16 places ahead of the country’s global position in Doing Business 2010. Main Issues Business regulations for paying taxes Several studies—for example, Bruhn (2011)—have found that lowering cor- porate tax rates can increase investment, reduce tax evasion by formal firms, Table Subnational Doing Business rankings, 2010 3.1 Dealing with Trading Ease of doing Starting a construction Registering Paying across Enforcing City and province business business permits property taxes borders contracts Faisalabad, Punjab 1 2 6 1 3 4 2 Multan, Punjab 2 6 1 7 3 5 4 Lahore, Punjab 3 3 3 4 3 13 8 Islamabad, Islamabad 4 1 8 3 1 11 10 Capital Territory Sheikhupura, Punjab 5 9 8 5 3 7 6 Gujranwala, Punjab 6 13 2 6 3 10 4 Sukkur, Sindh 7 10 4 10 11 3 1 Peshawar, Khyber 8 3 6 9 10 8 8 Pakhtunkhwa Karachi, Sindh 9 3 10 11 11 1 3 Rawalpindi, Punjab 10 8 5 7 3 12 10 Sialkot, Punjab 11 12 11 1 3 5 10 Quetta, Balochistan 12 6 12 13 2 9 13 Hyderabad, Sindh 13 11 13 11 11 2 7 Note: The ease of doing business is calculated as the ranking on the simple average of city percentile rankings on each of the six topics. The ranking on each topic is the simple average of the percentile rankings on its component indicators. Source: World Bank 2010. promote the creation of formal firms, and ultimately raise sales and GDP. Where lowering tax rates might not be viable (based on the tax regime’s objectives), reducing administrative burdens and compliance costs can also lead to more formal firms and higher sales. 59 The administrative burden of paying taxes is far higher in Pakistan than in most South Asian economies  (although Pakistan’s total tax rate of 35 per- cent is lower than the South Asian average of 40 percent). This helps explain why Pakistan’s global ranking on the paying taxes indicator is 162 of 185 economies—its second lowest for the Doing Business indicators, after get- ting ­electricity—against a much higher median ranking for South Asian economies of 106. Entrepreneurs have to make about 47 payments, which require almost 560 hours annually on filing, preparing, and paying taxes. By contrast, the South Asian average is about 30 payments, requiring about 311 hours. Just four payments (12 hours) are required in the United Arab Emir- ates, the best global performer (World Bank 2013). Historically, most taxes in Pakistan have been imposed by the federal govern- ment, leaving very few local variations. But some provincial taxes contribute to differences in businesses’ regulatory burden. Entrepreneurs spend the same amount of time on tax payments (around 560 hours) in most cities (World Bank 2010). The major difference is in Islamabad: with no social security contribu- tions, it is less cumbersome to pay taxes there (35 payments) than in Baloch- istan, Khyber Pakhtunkhwa, Punjab, and Sindh (47  payments; Figure 3.3). Islamabad’s total tax rate is also the lowest, at about 26 percent, against nearly 32 percent elsewhere. Business regulations for enforcing contracts Pakistani businesses identified courts as one of the top 10 constraints to firms’ investment, and as important as political instability and access to finance in 2007, according to the World Bank’s Enterprise Surveys. Apart from Bhutan and Mal- dives, South Asian economies are ranked fairly low on the enforcing contracts indicator. South Asia’s median rank is 146, its lowest for any of the 10 Doing Business indicators measured. Frequent case backlogs, long processing times, and high costs are common in the regulatory environment. Pakistan (ranked 155) is ahead of other large South Asian economies—such as India (184) and Bangladesh (182)—but still has considerable potential for improving, through both its current and future reforms. Resolving commercial disputes through Pakistan’s courts is regulated mainly by a single federal law, the Civil Procedure Code 1908. It takes 47 procedural Enhancing the Business Environment Figure Paying taxes, 2010 3.3 Payments (number) and time (hours) Payments (number) Time (hours) Total tax rate 800 40 60 600 30 Percent of pro t 400 20 200 10 0 0 Islamabad Punjab Sindh Balochistan Khyber Capital Territory Pakhtunkhwa Source: World Bank 2010. steps to enforce a contract, irrespective of the court’s location. However, the time taken and cost incurred for these procedural steps varies widely (Figure  3.4). In Faisalabad, the process takes 730 days; in Peshawar, 2,190— three times as long. Costs are lowest in Sukkur, at about 21 percent of the cost of claims, and highest in Lahore, at close to 43 percent. Some cities have introduced alternative dispute resolution mechanisms to expedite contract enforcement. They seem to have worked: according to an impact evaluation study in 2010, the number of pending court cases fell after such mechanisms were introduced for tax appeals (Gropper 2010). Figure Enforcing contracts, 2010 3.4 Procedures (number) Time (days) Cost Procedures (number) and time (days) 2,500 50 2,000 40 Percent of claim 1,500 30 1,000 20 500 10 0 0 Islamabad Faisalabad Karachi Rawalpindi Sheikhupura Sialkot Hyderabad Gujranwala Lahore Multan Quetta Sukkur Peshawar Islamabad Punjab Sindh Balochistan Khyber Capital Territory Pakhtunkhwa Source: World Bank 2010. Business regulations for registering property Burdensome and costly regulations for registering property can dissuade business owners from obtaining valid property titles. Apart from direct risks 61 of fraud and eviction are other disadvantages, such as the inability to use property as collateral for financing. In countries with more secure property rights, firms can better allocate resources, better protect returns on different types of assets, and thus grow faster (Claessens and Laeven 2003). In Pakistan, registering property takes six procedures and 50 days and costs about 8 per- cent of per capita income, reflecting Pakistan’s rank of 126 of 185 economies. Although all 13 cities analyzed require these six procedures, the costs and time to register property vary widely. Quetta requires the most days (52) and has the steepest costs (11 percent of the property value); Lahore requires the fewest days (30), and Islamabad the lowest costs (7 percent; Figure 3.5). Time varies mainly by differences in administrative efficiency across the revenue offices issuing the proof of ownership and transferring the property title. Local cost varies typically by different stamp-duty rates, which are set at the provincial level.1 Property registration is still manual in most cities, but Lahore and Sialkot have implemented reforms to computerize their land-record and deed-registration systems. Doing Business in Pakistan 2010 highlighted two computerization projects, called Land Record Management Information System and Participatory Informa- tion System, in selected pilot locations in Punjab and Balochistan. Lahore, as seen above, had the fastest processing time for property registration and Islamabad the lowest cost in Doing Business in Pakistan 2010, both far better than Karachi, Figure Registering property, 2010 3.5 Procedures (number) Time (days) Cost 60 12 Procedures (number) and time (days) Percent of property value 40 8 20 4 0 0 Rawalpindi Faisalabad Sheikhupura Sialkot Islamabad Sukkur Lahore Quetta Multan Gujranwala Hyderabad Karachi Peshawar Islamabad Punjab Sindh Balochistan Khyber Capital Territory Pakhtunkhwa Source: World Bank 2010. Enhancing the Business Environment which determines Pakistan’s global ranking.2 In fact, processing time in Lahore and registration cost in Islamabad in 2010 were better than Doing Business 2013 estimates for Karachi three years later. 62 Business regulations for construction permits Construction permits are an important indicator of a country’s business environment—for two main reasons. First, the industry often accounts for a sizable share of the economy—in Pakistan roughly 2 percent and employing nearly 7 percent of the labor force (Government of Pakistan 2012). Second, con- struction regulations help ensure the safety standards for building projects that protect the public while making the permitting process efficient, transpar- ent, and affordable. Doing Business 2013 ranked Pakistan 105 on the deal- ing with construction permits indicator, slightly better than the South Asian median rank of 109. However, the data measure regulatory impacts in Kara- chi, the country’s largest business city but among the weaker city performers. As licenses for site development, building permits, and utility connections are governed locally, time, costs, and the number of procedures vary widely (Figure 3.6). The number of steps to build a warehouse and obtain utility connections ranges from 11 in Multan, Lahore, and Faisalabad to 15 in Sialkot. The time required, however, varies far more: obtaining all approvals takes 124 days in Peshawar but 223 days in Karachi—almost twice as long. Similarly, the costs are Figure Dealing with construction permits, 2010 3.6 Procedures (number) Time (days) Cost 250 1,000 Procedures (number) and time (days) Percent of income per capita 200 800 150 600 100 400 50 200 0 0 Islamabad Lahore Quetta Multan Gujranwala Rawalpindi Hyderabad Faisalabad Sheikhupura Karachi Sialkot Peshawar Sukkur Islamabad Punjab Sindh Balochistan Khyber Capital Territory Pakhtunkhwa Source: World Bank 2010. about 422 percent of per capita income in Multan but close to twice as much in Islamabad (798 percent). For many cities, Doing Business in Pakistan 2010 highlighted provincial and 63 municipal reforms on construction permits that introduced good practices. Over 2006–09, development authorities and town municipal administrations in Lahore, Faisalabad, and Sialkot adopted uniform building and zoning regulations, such as those already in Multan, Rawalpindi, and Gujranwala. Consequently, the time to obtain a building permit fell 22 days in Sialkot and 15 in Lahore. Over the same period, a pilot program to computerize deed registration in Sialkot cut registration time from 13 days to 6, and computer- ized land records in Lahore cut the time to register property from 37 days to 30, making it the fastest process among the cities covered in Doing Business in Pakistan 2010. Multan is another example that demonstrates Pakistan’s potential to compete globally on many Doing Business indicators. The country’s best performer, it adopted uniform building and zoning regulations, specifying processes and documentation requirements. As a result, it would have ranked as high as 56 on the 2010 global rankings, ahead of developed economies like Australia (62), Norway (65), and Italy (85). The variation in time, costs, and number of procedures across Pakistan’s cities offers huge learning opportunities to repli- cate reforms and improve regulatory efficiency. Business regulations for starting a business Reforms making it easier to start a formal business are associated with 5–11 percent increases in the number of newly registered firms and 2–3 per- cent increases in employment (Motta, Oviedo, and Santini 2010). Such reforms can also improve factor productivity, investment rates, and value added per worker (World Bank 2013). If Pakistan is to continue improving competitiveness, easing business entry will be important. On the starting a business indicator, it ranks 98 globally, close to the South Asian median of 95. And its startup cost of about 10 percent of per capita income is far lower than the South Asian median (about 21 percent). However, its time and num- ber of procedures to register a business are among the region’s highest. Pakistan’s business registration system is highly centralized, meaning that entrepreneurs have to go through the same 10 procedures across the country. Yet there are differences in the time and costs incurred (Figure 3.7), due to varyingefficiency of local branches and variations in use and availability of online or automated “e-services.� Business startup time varied from 16 days Enhancing the Business Environment Figure Starting a business, 2010 3.7 Procedures (number) and time (days) Procedures (number) Time (days) Cost 30 30 Percent of income per capita 64 20 20 10 10 0 0 Rawalpindi Faisalabad Sheikhupura Sialkot Islamabad Sukkur Lahore Quetta Multan Gujranwala Hyderabad Karachi Peshawar Islamabad Punjab Sindh Balochistan Khyber Capital Territory Pakhtunkhwa Source: World Bank 2010. in Islamabad to 24 days in Gujranwala, explained mostly by separate post- incorporation registrations for taxes and social contributions. Costs varied from about 13 percent of percapita income in Islamabad to more than 26 per- cent in Sialkot, stemming generally from differences in incorporation fees for online and offline submissions. At the subnational level, Doing Business in Pakistan 2010 noted that reforms in the regulatory environment had made it easier to do business in all major cities for which data were collected previously for Doing Business in South Asia 2007.3 The introduction of e-Services, for example, in 2008 reduced the time needed to register a company with the Securities and Exchange Commission of Pakistan by one day to two days in Faisalabad, Lahore, and Peshawar (Box 3.1). The fees for online business registration were set at half the in-person registra- tion. Karachi reduced the time to import Islamabad FaisalabadGujranwala Lahore Multan Rawalpindi by half over 2006–09 after an elec- Sheikhupura Sialkot Hyderabad Karachi Sukkur Quetta Peshawar tronic data interchange system was introduced. At the provincial level, Punjab cut four days from the business startup process by delegating registration with the Employees Social Security Institution to local districts. Entrepreneurs are expected to benefit from the reforms under way. Coupled with comparatively low startup costs, Pakistan’s regulatory environment for business registration is heading in the right direction. In its next steps, the key reform for Pakistan to leapfrog other countries toward global best performers like New Zealand is a one-stop shop for entrepreneurs (one that fully integrates all business registration and postincorporation procedures on a single plat- form) to incorporate their business. To develop such an entity, the Economic Box Ongoing Regulatory Reforms to Facilitate Business Registration 3.1 Since the Securities and Exchange Commission of Pakistan introduced e-Services in September 2008, the business registration process has become easier—procedures can be completed online; faster— 65 companies can be incorporated in as little as two days; and cheaper—the fees for online registra- tion are about half that of in-person registration. In February 2012, the commission introduced fast-track registration services, which enables same-day incorporation. Reforms Unit of the Ministry of Finance prompted the creation of a working group of officials from the Securities and Exchange Commission of Pakistan, the Federal Board of Revenue, and the Employees’ Old-Age Benefits Institution. Business regulations for trading across borders Doing Business 2013 marks out Pakistan among the economies that have global good practices in the trading across borders indicator, such as risk- based inspections, electronic submission and processing, and an electronic window to government agencies. On this indicator, Pakistan ranks 85, much higher than the South Asian median, at 133. The costs to import and export in Pakistan are only about 41 percent of the South Asian average. Still, the positive impacts of these good practices vary subnationally. Even though reforms at the ports can help trading businesses across Pakistan, recent data suggest that subnational reforms, such as inland clearance facil- ities and improved inland transport, can reduce trade costs for cities other than Karachi. Moreover, the cost of trading across borders shows marked differences, particularly due to a poorly mapped trade logistics process and weak inland transport infrastructure. Table  3.2 illustrates how business regulations affect documentation requirements in Pakistan, as well as the time and cost of importing and exporting a standard container. While time and documentation vary little across cities, inland transportation and relative distance to and from the port of reference (Karachi Port or Port Qasim) are the determining factors in widely dispersed costs. Policy Options General options Engage all levels of government in a holistic reform agenda The Pakistani government has long recognized the importance of smart regulations. The country became one of the top global reformers in Doing Enhancing the Business Environment Table Trading across borders, 2010 3.2 Cost to Documents Time Cost to export Documents Time import 66 to export to export ($ per to import to import ($ per Province City (number) (days) container) (number) (days) container) Islamabad Capital Territory Islamabad 9 22 670 8 21 936 Faisalabad 9 22 639 8 20 739 Multan 9 22 624 8 20 936 Sialkot 9 22 639 8 20 860 Punjab Sheikhupura 9 21 639 8 20 1,088 Gujranwala 9 23 677 8 20 936 Rawalpindi 9 23 685 8 20 1,012 Lahore 9 23 791 8 20 1,088 Karachi 9 22 611 8 18 680 Sindh Hyderabad 9 20 616 8 20 708 Sukkur 9 21 639 8 18 784 Balochistan Quetta 9 23 619 8 21 693 Khyber Pakhtunkhwa Peshawar 9 22 715 8 20 784 Pakistan average 9 22 659 8 20 865 Source: World Bank 2010. Business 2006: Creating Jobs by introducing reforms in starting a business, registering property, protecting investors, and trading across borders. It has since introduced major regulatory reforms to make it easier for firms to start up and operate in at least five areas of doing business. The comparisons among cities within a large ­ country—such as Pakistan—are even stronger drivers for reform than global comparisons, and they promote peer-to-peer learning. The combination of an intense engagement strategy with subnational clients alongside provincial and national governments is important for creating ownership at all levels of government, which is crucial to enable space for reform. Learn from subnational peers Pakistan’s cities can learn from each other and adopt good practices already working in the country. As in Doing Business in Pakistan 2010, if a city in Pakistan were to adopt all the best practices already existing in the six areas of business regulations covered by the report, it would rank 16 places ahead of the country’s global position in Doing Business 2010. Differences across the cities can guide policy makers in areas where improvements are possible, provide for cities to learn from each other, and foster the adoption of good practices already seen in Pakistan. Indicator-specific options 67 Reform options for specific areas of the regulatory environment build on the pillars of multilayered engagement and peer-to-peer learning. For some reg- ulations, such as business registration, the onus of reforms falls on the federal government. For others, such as construction permits and property registration, reforms are handled by local districts. In paying taxes, trading across borders, and enforcing contracts, the reform process requires coordination among federal, provincial, and local district governments. In all cases, however, the local district government is a critical stakeholder for three reasons. It can inform policy making and the design of reforms. It is responsible for their imple- mentation. And it can monitor and evaluate their subnational impacts. The following options aim to cut the time, costs, documents, and number of procedures required for each area. Paying taxes The administrative burden and compliance costs of preparing, filing, and paying taxes are particularly heavy for entrepreneurs who run SMEs. Even where tax rates are low, the indirect costs of compliance can encourage tax evasion or deter smaller firms from entering the formal sector. The following proposed measures would reduce the administrative burden of tax regulations, address the uncertainty of tax rates, and facilitate tax payments: • Simplify the tax system and broaden the tax base by eliminating exemptions and preferential treatments that erode fiscal revenues. • Improve audit capacity through risk-based audit systems. • Provide incentives for local governments to develop their local tax bases. • Expand electronic filing and payment systems to reduce the transaction costs of paying taxes. • Focus on sequencing and communicating reforms—engage stakeholders, reach out to businesses, and bolster public support. Enforcing contracts Globally, increasing the specialization of judges, divisions, or courts in commercial cases has been a common feature of reforms to improve court efficiency in recent years. Many reforms have also been aimed at automating case management systems and enhancing procedural efficiency. And some Enhancing the Business Environment countries, including Pakistan, have adopted and promoted an alternative dispute resolution system so that the private sector can avoid lengthy court trials and the expenses of litigation.4 The following measures are proposed: 68 • Set up specialized courts or commercial divisions in existing courts. • Improve case management, collect statistical data, and monitor impact of reforms. • Reduce the case backlog by dismissing inactive cases and cases with unresponsive parties. • Introduce time limits and procedural deadlines. • Strengthen the alternative dispute resolution system to reduce court burdens and create cheaper, quicker options for resolving disputes. Registering property Doing Business in Pakistan 2010 noted the considerable potential for regulatory reforms in property registration. Land records and filing systems are predomi- nantly paper-based, local officials have little accountability but high workloads, and property registration involves many taxes and fees. The process in most cities is thus slower than the South Asian average benchmark. Reforms like computerizing land records in Lahore and Sialkot can serve as models for other districts and cities. Building on existing reforms, the following measures are proposed: • Simplify or consolidate procedures, eliminating the need for time-consuming nonessential procedures. • Improve the efficiency of the revenue office through computerization and greater accountability.5 • Reduce the number of taxes and fees. • Introduce a flat fee for stamp duty. Construction permits The regulatory environment for dealing with construction permits is highly decentralized. Indeed, most powers and responsibilities vested in district and local governments, enabling many of them to lead and introduce several reforms over the years. Over 2006–09, for example, development authorities and town municipal administrations in Lahore, Faisalabad, and Sialkot adopted uniform building and zoning regulations, allowing for greater consistency in the regula- tory environment. The following measures are proposed: • Review the role of the Patwari and introduce modern land management practices.6 • Review the internal process flows for requesting and obtaining a building permit. • Rationalize inspections. • Issue the completion certificate at the time of final inspection and reexamine its cost. • Review current building compliance and enforcement regulations to 69 improve their effectiveness. • Review the implementation of existing state bylaws with a risk-based system for environmental approvals to increase approval efficiency. Starting a business Doing Business 2013 emphasized three good practices in business regulations for starting a business. They include increased use of online systems, no minimum capital requirements, and one-stop shops. Pakistan’s regulatory environment already allows for online procedures and has no minimum capital requirements. Research shows that the creation of a one-stop shop is associated with an increase in new firms entering the market of roughly 5–6 percent (Motta, Oviedo, and Santini 2010). Pakistan has many reforms to reduce or improve business incorporation procedures, but to be effective such reforms must be complemented by streamlining postincorporation procedures, such as registering with labor departments and tax authorities. The following measures are proposed: • Make online incorporation fully functional for approving company names, registrations, and payments. • Promote use of online incorporation services for approving company names, registrations, and payments. • Eliminate the company seal requirement. • Create a single access point for tax registrations and social security requirements. • Eliminate the requirement for registration with the labor department of the district. • Create an integrated registration system, combining business, income, and sales taxes as well as a social security contribution. • Implement a unique ID concept for businesses that can be referenced by multiple stakeholders. Trading across borders Pakistan has been highlighted in Doing Business 2013 among the economies that have global good practices for trading across borders, such as risk- based inspections, electronic submission and processing, and an electronic window to government agencies. The following measures aim to build on them: Enhancing the Business Environment • Map the trade logistics process between each district and the relevant port to identify issues constraining the efficient movement of cargo, and identify potential areas of collaboration with the federal government to 70 increase efficiency. • Reduce and streamline documentation requirements. • Improve the use of prearrival information. • Strengthen inland clearance facilities. • Improve the electronic data interchange system. • Improve inland transportation infrastructure capacity. Notes This note draws from Kularatne and Lopez-Calix (2012). 1. While Quetta charges a stamp duty of 5 percent of the property value Peshawar and the cities in Sindh charge 3 percent. Islamabad and the cities in Punjab charge 2 percent. In Islamabad, where it is cheapest to register property, businesses are exempt from paying the urban immov- able property tax. 2. The global series of Doing Business reports measures regulatory impacts in the major business city of an economy. 3. Doing Business in South Asia 2007 covered six cities: Faisalabad, Karachi, Lahore, Peshawar, Quetta, and Sialkot. Doing Business in Pakistan 2010 covered 13 cities: Faisalabad, Gujranwala, Hyderabad, Islamabad, Karachi, Lahore, Multan, Peshawar, Quetta, Rawalpindi, Sheikhupura, Sialkot, and Sukkur. 4. For example, a study cited in Love (2011) finds cost savings of 50–60 per- cent and time savings of up to 11 months with alternative dispute resolution relative to court litigation in Columbia. 5. The two procedures that take place at the revenue office—obtaining the land record (fard) and transferring the property title—account for an average of 75 percent of the time required to register property in Pakistan. 6. The Patwari is the local official at the revenue office in charge of issuing the land record, or fard, and completing the property transfer. References Bruhn, Miriam. 2011. “Reforming Business Taxes: What Is the Effect on Private Sector Development?� Viewpoint Policy Journal Note 330, World Bank, Washington, DC. Claessens, Stijn, and Luc Laeven. 2003. “Financial Development, Property Rights, and Growth.� Journal of Finance 58 (6): 2401–36. The Economist. 2012. “Getting Better: Bad Rules Breed Corruption. Cutting Them Costs Nothing.� Doing Business 2013. October 27. Government of Pakistan. 2012. Pakistan Economic Survey 2011–12. Ministry of Finance, Islamabad. 71 Gropper, Akvile. 2010. “ADR in Tax Disputes.� Research Note, World Bank, Washington, DC. Jamil Afaqi, Muhammad, and Nadia Jahangir Seth. 2009. SME Sector: Genesis, Challenges and Prospects. Lahore, Pakistan: Small and Medium Enterprises Development Authority. Kularatne, Chandana, and Jose Lopez-Calix. 2012. “Addressing Regulatory ‘Software’ Barriers to Business Growth.� Policy Paper Series on Pakistan PK 07/12. World Bank, Washington, DC. Love, Inessa. 2011. “Settling Out of Court: How Effective Is Alternative Dispute Res- olution?� Viewpoint Policy Journal Note 329, World Bank, Washington, DC. Motta, Marialisa, Ana Maria Oviedo, and Massimiliano Santini. 2010. “An Open Door for Firms: The Impact of Business Entry Reforms.� Viewpoint Policy Journal Note 323, World Bank, Washington, DC. Pakistan Bureau of Statistics. 2005. “Census of Economic Establishments 2005.� Islamabad. World Bank. 2007. Doing Business in South Asia 2007. Washington, DC. ———. 2010. Doing Business in Pakistan 2010. Washington, DC. ———. 2013. Doing Business 2013: Smarter Regulations for Small and Medium-Size Enterprises. Washington, DC. 4 Sarwat Aftab and Sarmad Shaikh Reforming State‑owned Enterprises State-owned enterprises are a sizable element in Pakistan’s economic landscape—more than 100 of them operate in a wide range of economic sectors, contributing around 10 percent of GDP and representing about a third of stock market capitalization. But many are marred by weak corporate governance, cost-ineffective service delivery, and considerable financial losses. This note highlights the negative implications of large state-owned enterprises and suggests that they are a burden to already strained fiscal resources, deliver poor services, and create market distortions—all of which hold back economic growth and private sector development. Emphasizing the urgency of state-owned enterprise reforms, the note’s policy options propose measures (including commercializing state enterprise) to curtail fiscal costs, professionalize the role of the government as owner, and improve corporate governance and accountability in state enterprises. State-owned enterprises (SOEs) play an important economic role globally. Their weight may vary from country to country but they remain prominent in energy and network industries such as air and rail transport, electricity, gas and water supply, and natural resource extraction. SOEs are also predominant in the financial sector—particularly banking and insurance—and a consider- able and increasing number are also listed on national stock exchanges. They represent a significant part of total stock market capitalization—from around a fifth in Singapore, a fourth in India and Thailand, a third in Indonesia and Pakistan, half in Malaysia, and three-fifths in China (OECD 2010). Pakistan has substantial investment in its SOEs. They contribute about 10 percent of GDP, and the State Bank of Pakistan estimates that the country has about 100 SOEs at the federal and provincial levels (Speakman 2012). These enterprises provide infrastructure services (power, transport, logistics), economic develop- ment services (oil and gas, mining), financial services (banks, insurance). And they are spread throughout manufacturing. Over the past several decades, Pakistan’s governments have followed various approaches to manage SOEs. An Experts Advisory Cell was set up under the Ministry of Industries to monitor and support industrial SOEs in the 1980s, and in the early 1990s a privatization program was launched with a series of strategic sales in the industrial, energy, and banking sectors (including one power plant, two gas distribution companies, and one bank.1 In the early 2000s, the program conducted a series of capital market transactions that helped mobilize domes- tic savings and strengthen the domestic capital markets. These transactions 73 included international market listings in the form of global depositary receipts for United Bank Limited ($650.2  million) and Oil and Gas Development Company Limited ($772.4  million), which allowed the government to tap international institutional investors. Thus far, 167 privatization transactions have been concluded, realizing about $9 billion (Government of Pakistan 2011). A few SOEs incur substantial losses and have a major negative fiscal impact on public finances. This is due mainly to poor labor and capital productivity, obsolete management practices, and inadequate regulatory arrangements for utility tariffs. SOEs can also constrain private sector growth because of poor service provision. They can crowd out private provision in product and factor markets. And they may generate a strong negative image of the public service. The estimated losses of some key SOEs during 2011 were PRs 28 billion for Pakistan International Airlines Corporation, PRs 36 billion for Pakistan Rail- ways, PRs 22 billion for Pakistan Steel, and PRs 300 billion for the power sector. Heavy fiscal deficits in recent years have generated a large public debt. With lit- tle funds from external sources, financing has shifted toward domestic sources. Much higher fiscal deficits than in the recent past amplify the other challenges facing the government, including slow economic growth, recurring and devastat- ing floods, severe energy shortages, high inflation, and a weak security situation. The SOEs are a significant drain on the state’s limited resources. The State Bank of Pakistan in its Annual Report 2011–2012 stated that the fiscal deficit reached 8.5 percent of GDP in 2011/12, or PRs 1,761 billion, deviat- ing from the provisional estimates of 6.4 percent of GDP. This considerable slippage was exacerbated by shortfalls in tax revenues, nonrealization of some nontax revenues, and overruns in expenditures on account of the energy sector’s “circular debt� and SOE losses.2 The fiscal deficit also included one-off payments of PRs 391 billion (1.9 percent of GDP) for settling SOE debts (State Bank of Pakistan 2012). Apart from these direct subsidies, contingent liabili- ties represent an onerous fiscal impact: guarantees issued to SOEs affect the state’s capacity to borrow and thus negatively affect the state in the medium term. These explicit and implicit guarantees include unfunded losses of state- owned entities such as Pakistan Steel Mills, Pakistan International Airlines Corporation, Water and Power Development Authority, Pakistan Electric Power Company (PEPCO), and Pakistan Railways. In 2011/12, the govern- ment issued guarantees aggregating to PRs 203 billion (1 percent of GDP). REFORMING STATE-OWNED ENTERPRISES The outstanding contingent liabilities as of June 30, 2012, stood at PRs 517 billion (2.6 percent of GDP), which includes the stock of explicit debt guaran- tees in domestic (55 percent) and foreign (45 percent) currencies that appear 74 in SOE’s books of account (Ministry of Finance 2013). Other than the publically guaranteed debt of SOEs, the government issues counterguarantees against the commodity financing operations of Trad- ing Corporation of Pakistan, Pakistan Agricultural Storage and Services Corporation, and provincial governments. The Fiscal Responsibility and Debt Limitation Act 2005 stipulates that the issuance of guarantees, including those for rupee lending, bonds, rates of return, output purchase agreements, and all other claims and commitments that may be prescribed from time to time, as well as renewal of existing guarantees, should not exceed 2 percent of estimated GDP in any financial year. Figure 4.1 illustrates the total guarantees outstanding and subsidies issued over the last three years. Guarantees outstanding and subsidies issued to state- owned Figure enterprises, 2009/10–2011/12 4.1 Guarantees outstanding Subsidies issued 750 500 PRs billion 250 0 2009/10 2010/11 2011/12 Guarantees outstanding Subsidies issued 5 4 Percent of GDP 3 2 1 0 2009/10 2010/11 2011/12 Source: Government of Pakistan 2012b; Ministry of Finance Debt Policy Statement various years; Ministry of Finance Fiscal Policy Statement various years. The fiscal support to SOEs in general and to the eight SOEs on the radar of the Cabinet Committee on Restructuring in particular is heavy (Figures 4.2 and 4.3). During 2011/12, the government issued PRs 203 billion in guaran- tees. That year, the total outstanding stock of guarantees declined by PRs 42 75 billion, implying that PRs  245 billion in guarantees either expired or were retired through government resources. Subsidies remain a major drain on the government’s financial resources. In 2011/12, PRs 512 billion (2.5 percent of GDP) was spent on subsidies, out of which PRs 464 billion (2.2 percent of GDP) was for power. (In 2010/11, the Annual fiscal support to state-owned enterprises—new guarantees Figure issued, 2006/07–2011/12 4.2 Amount of guarantees Guarantees 300 3 Statutory limit under Fiscal Responsibility and Debt Limitation Act 2005: 2 percent of GDP Percent of GDP 200 2 PRs billion 100 1 0 0 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 Source: Government of Pakistan 2012b; Ministry of Finance Debt Policy Statement various years. Guarantees and subsidies issued to state-owned enterprises, Figure 2011/12 4.3 (PRs billion) Guarantees issued Subsidies provided Trading Corporation of Pakistan 18.3 Pakistan Agricultural Oil re neries/ Pakistan Pakistan Storage and Services oil marketing companies 7.9 Steel Mill 8.9 International Corporation 18.7 Utility Stores Corporation 2.0 Airlines Others 4.4 Others 1.2 Karachi Electric Corporation 9.5 Supply Company 45.2 Water and Power Water and Power Development Authority/ Development Authority/ Pakistan Electric Pakistan Electric Power Company Power Company 180.4 419.0 Source: Government of Pakistan 2012a; Ministry of Finance 2013. REFORMING STATE-OWNED ENTERPRISES equivalent figures were 2.2 percent and 1.9 percent of GDP; Government of Pakistan 2010.) It is important to eliminate the price differential in electricity tariffs that generate circular debt in order to create fiscal space for develop- 76 ment expenditure. In 2011/12, 21 percent of the government’s fiscal resources were consumed in power subsidies, a sharp jump from 6 percent of total fiscal resources in 2005/06 (Government of Pakistan 2012b; Ministry of Finance Fiscal Policy Statement various years). Moreover, actual expenditure on subsidies has been much higher than budget estimates consistently for the last four years, particularly due to increasing power subsidies. In 2010/11 and 2011/12, actual power subsidies were three and two times the budgeted amounts (Figure 4.4; Government of Pakistan 2010). Main Issues Absence of a legal and regulatory framework for state-owned enterprises The government has taken some initial steps to improve the performance of key SOEs. In January 2010, the prime minister constituted a Cabinet Committee on Restructuring of public sector enterprises to tackle institutions’ financial losses. The Committee identified eight SOEs for restructuring with the objective of improving overall corporate governance of these entities, curtailing financial hemorrhaging, improving service delivery, and reducing the fiscal burden on the exchequer.3 To improve SOE governance standards, the Ministry of Finance consti- tuted a task force of public and private sector representatives to finalize the corporate governance rules for SOEs, drafted by the Securities and Exchange Commission of Pakistan (SECP). These rules are for SOEs with a corporate Figure Power subsidies, budgeted and actual expenditures, 2007/08–2011/12 4.4 Target Actual 500 400 300 PRs billion 200 100 0 2007/08 2008/09 2009/10 2010/11 2011/12 Source: ADB 2012. structure and are based on the best principles of corporate governance of the Organisation for Economic Co-operation and Development (OECD). The task force completed its work and the draft Public Sector Companies (Corporate Governance) Rules were notified by the SECP for public consultation in March 77 2012 and were placed on the SECP’s website. To seek stakeholders’ views, a thorough consultative process was then followed and conferences held in various cities. These rules were formally approved by the government on March 8, 2013, and will be effective after 90 days of the issuance of the notification. Broader SOE reforms are being carried out as part of Pakistan’s development agenda and framework for economic growth. Some of the government’s other SOE reforms are in Box 4.1 (Economic Reforms Unit 2012). The government has started a program for granting stock options to SOE employees under the Benazir Employees Stock Option Scheme. The Scheme was announced in August 2009 and offers 12 percent stock options from the Box Some initiatives for state-owned enterprise reform 4.1 • The Economic Reforms Unit set up in the Ministry of Finance was designated as the secretariat to the Cabinet Committee on Restructuring. • The government has expedited implementation of the Power Sector Reform Plan 2010, which was upgraded under the Power Sector Recovery Plan 2011. Dissolution of Pakistan Electric Power Company is under way and will be completed by the end of September 2013 with transfer of operational functions to National Transmission and Dispatch Company and Central Power Purchase Authority. The Authority is already operating, and boards of directors for the nine distribution companies and Generation Holding Company have been reconstituted. • The board of directors of Pakistan Steel Mills has been strengthened and implementation of a business plan for revitalizing the company, approved by the cabinet, has started. • A restructuring framework for Pakistan Railways has been brought into operation. An asset manage- ment company is being set up for optimum use of the entity’s assets. Priorities are to repair locomotives—commercial borrowing has been obtained to repair 96 locomotives—and to ratio- nalize routes and freight operations. Reconstituting the board of directors and senior management has also been part of the overall restructuring process since early 2012. • A restructuring plan for Pakistan International Airlines Corporation has been finalized, and addresses corporate governance, human resource rationalization, financial and operational restructuring, engi- neering improvements, procurement and logistics, airport services, and dispatch reliability. • Initial restructuring plans for Trading Corporation of Pakistan, Pakistan Agricultural Storage and Services Corporation, and Utility Stores Corporation have been framed. • General reforms such as implementation of rules for regulating public procurement of goods, ser- vices, and works by the Public Procurement Regulatory Authority have been introduced. • The role of the Competition Commission of Pakistan in providing a level playing field has been strengthened through a separate Act. REFORMING STATE-OWNED ENTERPRISES 80 public organizations to the employees of their respective organization.4 Around half a million employees are expected to benefit. 78 Going forward, it is important to focus on SOE corporate governance reforms as this will help improve management and thus in delivery of better and more cost effective distribution of goods and services to the public at large. Currently, there is no diagnostic assessment of the sector and hence no road- map or action plan formulated to move forward in a coherent manner. There is no SOE Act or equivalent law in the country that would lay down the foundation for good corporate governance. Most jurisdictions have a SOE law that provides the framework for how the state, as owner and policy maker, governs its SOEs. Others have developed an SOE ownership policy, laying down the parameters of state ownership, its role in the corporate gover- nance of SOEs, and how the government will implement its ownership policy. However, both the SOE law and ownership policy are missing in Pakistan. Lack of clear ownership role A fundamental tenet of SOE reform is the need to separate the government’s role as owner from its role as policy maker, coordinator, subsidy deliverer (typically handled by the line ministry), and regulator (typically handled by an independent regulator). In some cases, the line ministry’s role is further reduced by moving the subsidy-financing role elsewhere. Lack of clearly defined, explicit, and consistent ownership policy has resulted in vaguely defined roles and responsibilities for the government, leading to conflicts of interests at the government level, ineffective monitoring and evaluation of SOE performance, and weak corporate governance practices at the SOE level. The government has made little progress in separating these roles as in most cases they are with the line ministries, which vary in capacity but do not act as informed and active owners. Little effort is made by line ministries to ensure that SOE governance is carried out transparently, accountably, professionally, and effectively. No formal monitoring mechanism is in place to gauge SOE performance. There are no regular checks on managerial performance as the boards set neither targets nor performance indicators for management. And there is also no clearly defined board nominations policy that ensures that the best possible directors are nominated independently and transparently to serve on SOE boards. Pakistan has examples of institutional governance working well—and not so well. The country can be proud of institutions like the State Bank of Pakistan, but PEPCO—a company set up for two years to facilitate corporatization of power sector entities—outlived its useful life and continued operations for more than 14 years, and with limited effect. PEPCO failed because of a lack of politi- cal will in government, policies incentivizing high-cost production, insufficient 79 autonomy, dependency on legacy staff to implement reforms, and lack of accountability and clear lines of authority, among others (USAID and World Bank 2012). Weak corporate governance standards The corporate governance Report on Observance of Standards and Codes 2005 found that corporate governance was making progress, despite gen- erally low awareness of its importance. 5 This was exacerbated by a prepon- derance of family-owned private companies that tended to avoid modern corporate governance practices. Such a finding is not unusual in emerging markets and typically requires enforcement interventions (regulatory capac- ity, legislative protection, and disclosure requirements) as well as capabil- ity building among directors themselves. For SOEs there was no distinction between the ownership role (referred to above) and the role of directors, yet this is fundamental to effective corporate governance, particularly when minority shareholdings or multiple shareholding interests are at stake (World Bank 2005). Good corporate governance begins with appointing and empowering an independent board of directors. It is inadvisable to include current civil servants or ministers on the board, which is common practice in Pakistan. The directors are generally not aware of their fiduciary responsibilities, and the board structures are underdeveloped, with limited effectiveness of board committees and processes. Finding competent independent directors can also be quite challenging (World Bank 2005). The Corporate Gover- nance Rules approved by the government in March 2013 are a good first step, but the real test lies in their implementation. Further, the SOEs listed on stock exchanges are already mandated to abide by the 2012 Code of Corporate Governance for listed companies.6 Of particular importance in Pakistan is the appointment and role of the CEO. Identifying and selecting a professional CEO of an SOE, independent of government interference, is a challenging but critical factor for any SOE reform. Measures or protection are needed to ensure that only qualified and noncon- flicted directors and CEOs are appointed. Moreover, CEOs should be appointed by the board rather than by the government. Such measures could be part of the implementation mechanism for the new Corporate Governance Rules. REFORMING STATE-OWNED ENTERPRISES Weak disclosure practices and poor control frameworks are two significant corporate governance weaknesses. The disclosure refers to both financial and nonfinancial practices while the poor control mechanism relates to risk 80 management, internal controls and internal audit frameworks. Policy Options Efforts to reform SOEs have stalled in Pakistan for almost five years— contributing to steep fiscal losses and to worsening and cost-ineffective services. Urgent policy measures are needed to improve the efficiency and effectiveness of SOEs, now discussed under three areas. Enhance financial discipline and strengthen legal framework The government has to identify enterprises that need to be privatized and ensure completion of the privatization process within the assigned time- frame. Corporate governance reforms are no substitute for privatization but do improve the prospects of the entity to be privatized. To build public confidence, privatization needs to be transparent. To bring down fiscal costs, it is important to continue the reform process initiated by the government through restructuring boards and managing key SOEs. This step should be followed by corporatization where the SOEs are not already in a corporate structure. The SOEs should be required to prepare business plans. Restructuring will reduce fiscal costs if the newly appointed boards and management are effective and if corporatization brings the entities under the same law (the Companies Ordinance) and under one regulator (the SECP), thus ensuring a level playing field for the private sector. These measures are in line with OECD guidelines recommending that govern- ments simplify and streamline the operational practices and legal form under which SOEs operate. Graduating to the next stage—­ commercialization—will take more time and a substantial commitment of resources. Commercialization, after the conversion of an enterprise to a company, typically includes identifying noncommercial activities, separating them from the enterprise, and financing them separately; cleaning up the balance sheet and establishing appropriate staff levels through financial and operational restructuring; establishing modern systems of professional management (information, human resources); upgrading human resources; and changing the culture. Commercialization will probably lead to a rethinking of efficiency and profitability at these SOEs, highlighting the question of retrenchment. The government needs to ensure that the right mechanisms are in place to manage, retrain, or compensate staff. It is important to move forward coherently. The first step could be to undertake a diagnostic assessment of the sector, including fiscal issues, legal/regulatory framework, current corporate governance practices, key issues, menu of reform options, and preliminary recommendations. Based on this assessment, a road 81 map/action plan can be prepared by the government for SOE reform. It is crucial to have a legal and regulatory framework for SOEs. The new Corporate Governance Rules are the right step, but an SOE Act or equivalent law is needed to provide the framework for how the state, as owner, governs its SOEs. Typically such an Act requires SOEs to be commercially oriented; SOEs to be forbidden to expand or contract the scope of their activities without govern- ment approval;7 SOE obligations to provide timely financial and management information; guidance on dividends; clarity on which activities of government should be turned over to SOEs and those that should not; clear delineation of the roles and responsibilities of the government, boards, and management; and processes for appointing boards of directors. The legislation may also include hard budget constraints that incentivize the SOEs to run profitably. In parallel, the government takes on the obligation to compensate in a timely manner the SOE when it contractually agrees on services from the SOE— nothing breaks down SOE reform discipline faster than nonpayment of debts owed by other government agencies.8 A core question for Pakistan’s authorities is whether they need the equivalent of an SOE Act. A lesser option would be to issue an ownership policy that defines the overall objectives of state ownership, the state’s role in the corporate governance of SOEs, and how it will implement its ownership policy. Professionalize the role of government as owner It is imperative to separate the government’s role as owner from its role as policy maker, coordinator, and regulator. The government should refrain from getting involved in day-to-day management and should allow SOEs full operational autonomy to achieve their objectives by professionalizing SOE boards and holding them accountable through the development of a proper performance monitoring and evaluation system. The government should replicate past models that succeeded but were unfortunately discontinued.9 Appointing and monitoring the boards is integral to monitoring ownership. As the owner, the government is accountable to Parliament, and it needs to appoint directors capable of meeting the owner’s expectations, following a skills-based appointment process. Anything less—such as patronage or repre- sentation—diminishes the skills on the board. There should be a well-struc- tured and transparent board-nomination process. The boards, once appointed, REFORMING STATE-OWNED ENTERPRISES should be allowed to exercise their responsibilities independently. Centralized reporting systems should be put in place to allow regular monitoring and assessment of SOE performance. 82 The ownership function of SOEs in Pakistan lies with the concerned line ministry. Several countries have moved toward creating a centralized owner- ship entity charged with SOE oversight. Ownership models can be grouped in three broad types (World Bank 2012): • The decentralized model, where ownership responsibilities are dispersed among different line ministries. • The dual/hybrid model, where, in addition to line ministries, a second ministry, such as the Ministry of Finance, may also have certain responsibil- ities like approving annual budgets, subsidies, and transactions and moni- toring financial performance. • The centralized model, where ownership responsibilities are consolidated in an entity that is independent or under the authority of one ministry. There is no one-size-fits-all solution for ownership models. But Pakistan can shift from the current decentralized model to either a dual/hybrid model or a centralized model. A centralized model can take various forms like an advisory/coordinating body (India, New Zealand, Norway, South Africa, Sweden), ownership agencies (China, France, Indonesia), or holding/invest- ment companies (Gulf countries, Hungary, Malaysia, Singapore). Determining which model is most feasible in Pakistan needs to be based on a more detailed assessment of the legal framework and the institutional context. Irrespective of the option selected, regular monitoring of financial and man- agerial performance is extremely important and needs to be prioritized. The government should ensure that each SOE has a strategy, clarifying commercial and noncommercial strategic objectives. Also there should be a set of clearly defined key performance indicators in a memorandum of under- standing that is tracked regularly by the government. To introduce more transparency into the SOE portfolio, the government should start an annual SOE publication, which highlights financial performance and information such as key performance indicators and board composition. Improve corporate governance and accountability Effective corporate governance of SOEs can have a positive impact on an economy, especially for a country with many SOEs. The benefits of better governance include improved SOE financial performance, better service delivery, and greater access to capital markets. The Republic of Korea and Singapore are two notable success stories of strong corporate governance and well-run SOEs. In Pakistan, most of the SOEs listed on the Karachi Stock Exchange and that follow the exchange’s corporate governance requirements are profitable. But listing on the Exchange does not mean that SOEs will 83 be free of political influence and continue to be commercially viable.10 For corporate governance framework to succeed for SOEs in Pakistan, the govern- ment must not interfere in the enterprise’s operations. The SOEs should have an autonomous board of directors accountable to the government for results and performance. The CEO should appointed transparently by the board. The approval of the Corporate Governance Rules for SOEs is a good first step, but the real test lies in effective implementation. Also important will be workshops and training sessions to build awareness and understanding of the Rules. Developing an implementation plan that clearly lays down the roles and responsibilities of the line ministries, Ministry of Finance, and SECP is important. The government should also extend the requirements of the Rules to SOEs that are not in corporate structure until all SOEs are corporatized. Internal and external accountability mechanisms are a function of formal relationships and disclosure. Accountability and disclosure to the Parliament through an annual report on SOEs that includes financial and operational reporting for each enterprise can be a start. The Annual Report of the State Ownership Steering Department in the Prime Minister’s Office of Finland is a benchmark of accountability to the National Assembly (Government of Finland 2011). The financial statements of SOEs should follow international financial reporting standards and should be placed on the SOEs’ websites. The performance of the senior management on the targets (for example, key performance indicators) should be evaluated by the board and should be linked to remuneration. A transparent mechanism should be put in place to evaluate the performance of the board. Notes 1. Kot Addu Power Company Limited was privatized in 1993 and Muslim Commercial Bank in 1991. Sui Southern Gas Company and Sui Northern Gas Pipelines Limited were corporatized in 1990s and shares sold on the stock market. 2. Circular debt in the energy sector of Pakistan refers to the increasing receivables in the power supply chain among refineries, marketing com- panies, power producers, distribution companies, and end users. Overall circular debt was estimated at PRs 382.5 billion in July 2012 (State Bank of Pakistan 2012). REFORMING STATE-OWNED ENTERPRISES 3. These were the National Highway Authority and Pakistan Railways— established under the relevant enactments; Pakistan International Air- lines Corporation—a listed company under section 503 of the Companies 84 Ordinance 1984 (the section applies to listed companies governed by spe- cial enactment); Pakistan Steel Mills, Trading Corporation of Pakistan, PEPCO, and Utility Stores Corporation—private companies; and Pakistan Agricultural Storage and Services Corporation—a public company. 4. For more information, see Government of Pakistan (2011). 5. World Bank 2005. This report is based on the OECD principles of good governance. 6. The 2002 Code of Corporate Governance was revamped and brought in line with international practices. The 2012 Code was implemented in April 2012. 7. While the memorandum of association somewhat limits the scope of activity of a corporatized SOE, this issue is rarely monitored. 8. This is often referred to as circular debt. 9. Policy, regulatory powers, and ownership of assets were separated in the Ministry of Petroleum and Natural Resources, until 2008 when it was reversed. 10. The two relatively well-performing companies (Sui Southern Gas Company and Sui Northern Gas Pipelines) listed on the Karachi Stock Exchange for quite some time are now overstaffed and face serious liquidity and recovery problems with unaccounted for gas losses hitting double digits, up from 3–4 percent previously. References ADB (Asian Development Bank). 2012. Asian Development Outlook 2012: Confront- ing Rising Inequality in Asia. Mandaluyong City, Philippines. Economic Reforms Unit. 2012. “Meetings with World Bank South Asia Financial & Private Sector Development Unit.� Ministry of Finance, Islamabad. Government of Finland. 2011. 2010 Annual Report of the Ownership Steering Depart- ment in the Prime Minister’s Office. Ownership Steering Department, Prime Minister’s Office, Helsinki. Government of Pakistan. 2010. “Budget in Brief 2010–11.� Ministry of Finance, Islamabad. www.finance.gov.pk/budget/Budget_in_Brief_2010_11.pdf. ———. 2011. “Privatization Commission Year Book 2010–2011.� Privatization Commission, Ministry of Privatization, Islamabad. www.privatisation.gov.pk/ Year%20Book/Year%20Book%202010-2011.pdf. ———. 2012a. “Budget in Brief 2012–13.� Ministry of Finance, Islamabad. www.finance.gov.pk/budget/Budget_in_Brief_2012_13.pdf. ———. 2012b. “Economic Survey of Pakistan 2011–12.� Ministry of Finance, Islamabad. Ministry of Finance. 2013. Debt Policy Statement 2012–13. Islamabad: Debt Policy Coordination Office. ———. Various years. Debt Policy Statement. Islamabad: Debt Policy Coordination Office. 85 ———. Various years. Fiscal Policy Statement. Islamabad: Debt Policy Coordina- tion Office. OECD (Organisation for Economic Co-operation and Development). 2010. “Corporate Governance of State-Owned Enterprises in Asia: Recommenda- tions for Reform.� Policy Brief presented at 5th meeting of the OECD Net- work on Corporate Governance of State-Owned Enterprises in Asia, Kuala Lumpur, May 24–25. Speakman, John. 2012. “SOE Reform: Time for Serious Corporate Governance.� Policy Paper Series on Pakistan PK 04/12. World Bank, Washington, DC. State Bank of Pakistan. 2012. Annual Report 2011–2012: The State of Pakistan’s Economy. Karachi. USAID (United States Agency for International Development) and World Bank. 2012. “Pakistan’s Power Sector Enterprises—A Case for Reform.� CASA-1000 Joint Working Group Meeting between USAID and World Bank, October 4, Almaty, Kazakhstan. World Bank. 2005. Report on Observance of Standards and Codes: Pakistan. Washington, DC. ———. 2012. SOE Corporate Governance Toolkit. Washington, DC. 5 Dan Biller and Ernesto Sanchez-Triana Revitalizing Manufacturing On a trend decline since 1960, Pakistan’s economic growth is highly dependent on financial inflows of aid and worker’s remittances. Manufacturing has great potential, though, based on a growing labor force and rising urbanization and connectivity, but its poor performance stems from its largely low-skilled labor force, poor commercial environment, and failure to diversify production and climb up the technology ladder. Manufacturing remains heavily concentrated on nondynamic low value-added consumer products that attract little invest- ment. Though manufacturing is improving these factors, it has room to sharply raise its con- tribution to GDP while reducing public “bads� of an unemployed labor force, such as social conflict and the need for widespread social assistance. Equally, negative impacts of indus- trial expansion, such as pollution, need to be mitigated with a combination of economic instruments, public participation, industrial voluntary actions, and command and control regulations. This note discusses ways in which Pakistan can revitalize its manufacturing and promote agglomeration economies, thereby increasing manufacturing’s contribution to medium-term growth and to job creation. Like other South Asian countries, Pakistan is experiencing a shift from an agriculture-based to a services-based economy. But South Asia—unlike East Asia where manufacturing has played an increasing role in recent decades on the path to becoming largely middle income—has seen its manufacturing share of GDP stagnate since 1980 in all major economies except Bangladesh. For Pakistan, trends are complicated by declining growth since 1960 and overreliance on financial inflows (aid and worker’s remittances). At the same time, a young and largely low-skilled unemployed labor force has also been rising. Part of the solution is for Pakistan to follow in China’s footsteps by engaging in low-skilled labor-intensive manufacturing, filling the gap left behind as China moves up the value chain with its increasingly sophisticated labor force and amid pressure from increasing wages. Yet, equally important is to join China in moving up the value chain. Jolting Pakistan’s manufacturing into increasing its share of GDP involves a raft of measures. They include improving skills and the commercial environment; harnessing the positive forces of agglomeration (such as focusing on infra- structure investments and facilitating trade); promoting inclusive green industrial growth (by, for example, addressing bottlenecks in supply chains, taking advantage of cleaner industries and certification for exports, and 87 facilitating entry of women in the labor force); and ensuring that regu- lations are in place and enforced to avoid the negative externalities of agglomeration. Pakistan is in the midst of two key trends that can provide increased manufac- turing potential: significant migration to urban areas and the development of “agglomeration� economies. The first is a physical concept, the second an economic one. Urbanization facilitates agglomeration, as face-to-face exchange of information is important in allowing labor and production to learn from each other and to apply technological advances. It can also reduce transport costs, as moving goods, people, and ideas becomes cheaper. People may take advantage of agglomeration when different ways to connect and exchange information are available (telecommuni- cations, information technology, and so on). Agglomeration economies also help create internationally connected cities, concentrating produc- tion and further facilitating economic growth.1 These trends combined can have a profound inf luence on how natural resources and infrastruc- ture are used, which in turn may inf luence migration, agglomeration, and ultimately the prospects for high-productivity jobs and economic growth (Box 5.1). But if unchecked, urbanization may backfire due to the public bads associated with it—hence the importance of ensuring that such growth is “green� and inclusive (that is, it takes into account environmental and social issues). Countries tend to move up the path of urbanization and economic progress in tandem. As cities catalyze agglomeration, they are important engines of economic growth. Figures 5.1 and 5.2 provide snapshots of this relationship in 1960 and 2011. In 1960, Latin America was fairly distant from South Asian and East Asian economies not only in urban population but also GDP per capita. Half a century later, the Republic of Korea and Malaysia are not only as urbanized as Latin American countries but are also richer. Pakistan, however, has not moved up the curve: although one of the most urbanized countries in South Asia, its urbanization and growth pale in comparison with that in other countries. It has been unable to use its urbanization and agglomeration to generate the growth and high-productivity jobs that should come with them. Equally, megacities like Karachi suffer from the public bads of urbanization, like congestion, pollution, and social conflict. REVitalizing manufacturing Box Infrastructure, agglomeration, and urbanization 5.1 Rising shares of services and manufacturing in an economy are generally associated with urbanization as the economy increases its sophistication through stronger productivity and growth. The box figure presents 88 a simple conceptual framework of various factors. All influence one another in a virtuous circle, and infra- structure services (connectivity) are inputs to the different quadrants and to the circle as a whole. The goal of a virtuous circle Urbanization Infrastructure Green economic services Agglomeration growth (connectivity) High-productivity jobs Figure Snapshot of urbanization, 1960 5.1 5,000 (PPP $, in ation adjusted; log scale) Mexico Chile 4,000 Brazil GDP per capita 3,000 Indonesia Sri Lanka Malaysia 2,000 Philippines Bangladesh Dominican Republic 1,000 Korea, Rep. Afghanistan India Pakistan 0 Bhutan China –1,000 0 10 20 30 40 50 60 70 80 Urban population (percent of total) Note: Regions are in different colors. The size of the bubbles represent country populations. Source: World Bank 2012; www.gapminder.org. As an economy develops and urbanizes, the share of agriculture in GDP inevitably declines while that of manufacturing and services rises. This form of intersectoral change normally represents a gradual shift from low- to high- productivity activities. With country growth, manufacturing tends to diversify Figure Snapshot of urbanization, 2011 5.2 30,000 (PPP $, in ation adjusted; log scale) Korea, Rep. 20,000 89 GDP per capita Mexico Chile Malaysia China Brazil 10,000 Sri Lanka India Dominican Republic Nepal Philippines 0 Afghanistan Indonesia Maldives Pakistan Bangladesh Bhutan –10,000 0 20 40 60 80 100 Urban population (percent of total) Note: Regions are in different colors. The size of the bubbles represent country populations. Source: World Bank 2012; www.gapminder.org. as firms produce and export a wider range of more sophisticated products. At even higher levels of income, specialization again increases but in high value-added and technologically advanced products. Increased product diversification therefore represents an intermediate stage in a country’s economic development. Yet, apart from Bangladesh, manufacturing has been virtually stagnant for the past 30 years in South Asia (Figure 5.3). Agriculture’s loss has largely been services’ gain. 2 While mitigating the potential negative effects, Pakistani manufacturing needs to capture the benefits of agglomeration. Small industrial companies have generally had a more stable and upward trend than larger companies, amid lagging industrial growth generally. Figure Evolution of GDP composition in South Asia, 1960 –2010 5.3 Agriculture Industry Services 100 Islamabad FaisalabadGujranwala Lahore Sheikhupura Sialkot Hyderabad Karachi Sukkur Multan Rawalpindi Quetta Peshawar 75 Percent 50 25 0 1990 2010 1990 2000 2010 1990 2000 2010 1960 1970 1980 1990 2000 2010 2000 2010 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Afghanistan Bangladesh Bhutan India Malaysia Nepal Pakistan Sri Lanka Source: World Bank 2012. REVitalizing manufacturing Growth of small and household firms has been fairly steady since 2000 while large-scale manufacturing saw a sharp increase in 2002–05, followed by a sharp contraction, reflecting weakening aggregate demand, deteriorating security, 90 and power shortages (Figure 5.4). Districts with faster population growth have seen stronger growth in manu- facturing production value, a sign that even though manufacturing is under- performing, agglomeration is an essential element in its growth (Figure 5.5). Districts with a population density of more than 600 persons per square kilometer are characterized by industrial development and by better education, health infrastructure, and sanitation than those in rural areas. 3 By contrast, districts with low population densities—below 30 persons per square kilome- ter—offer few job opportunities and little to no industrial presence (World Economic Forum 2011). Agglomeration’s importance is also seen in location and type of firms. Firms tend to set up in areas with “location economies�—areas that minimize procurement costs (such as those of transporting raw materials to the firm) and distribution costs (such as those of distributing products to customers). These are areas with specialized labor, interindustry spillovers, local trans- fers of knowledge, and access to export markets. Statistical results based on the Ellison and Glaeser Index indicate that 35 percent of industries in Paki- stan are highly agglomerated, 38 percent are moderately agglomerated, and 27 percent are not agglomerated (Ellison and Glaeser 1997). The most highly concentrated industries are ship-breaking, followed by sports and athletic goods. Other highly concentrated industries are those which need to be close to consumers and suppliers.4 Figure Growth of large and small manufacturing in Pakistan, 1950 –2010 5.4 Large rms Small and household rms 30 20 Percent 10 0 –10 1950–59 1960–69 1970–79 1980–89 1990–99 2001 2005 2010 Source: Ministry of Finance 2008. Relationship between value of manufacturing production and Figure population growth 5.5 Population growth, 1981–82 to 2005–06 1.00 Tharparker Quetta Karachi 91 Karak Lahore 0.75 (log scale) Faisalabad 0.50 Kalat Nawabshah 0.25 Toba Tek Singh 0.00 0 100 200 300 400 500 600 700 800 Value of production, 2005–06 ($ millions) Source: LUMS 2011. Main Issues Declining resources for infrastructure to stimulate agglomeration-benefiting industries Weak infrastructure is one of the main bottlenecks for doing business in Pakistan. Access to power, communications, and transport is a top infrastruc- ture concern for businesses worldwide. Pakistan provides relatively low access to services, which in turn inhibits foreign investment. As examples, Pakistan has only 2 fixed telephone lines per 100 people, far below the 70 in China, 17 in Sri Lanka, 16 in Malaysia, and 16 in Indonesia, and less than India’s 3 (World Economic Forum 2011). Road density in Pakistan is also far lower than in India and Sri Lanka. Only about 62 percent of the population has access to power, less than the 99 percent in China and 80 percent in Sri Lanka, while unreliable energy supply in Pakistan has stifled industrial growth (World Bank 2009a). An inefficient transport sector costs Pakistan’s economy 4–6 percent of GDP a year.5 Investment in key areas is declining. Although the share of transport and communication investment in the total rose from 12 percent to 24 percent in 2000–10, the share of industrial investment fell from 38 percent to 20 percent.6 The share of manufacturing recorded a similar decline, driven largely by a steep fall in investment in large firms (Figure 5.6). The Global Competitiveness Report 2011–2012 ranks Pakistan 115 of 142 countries in infrastructure. The biggest obstacles for doing business in Pakistan are gov- ernment instability and coups, corruption, policy instability, inadequate supply of infrastructure, and inefficient government bureaucracy, followed by poor REVitalizing manufacturing Share of manufacturing investment in total fixed investment, Figure 2000 –10 5.6 Large rms Small rms 30 92 20 Percent 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Source: Pakistan Economic Survey (various issues). access to funding. All these factors reduce export competitiveness, hit the economy’s ability to integrate with global supply chains, and increase the cost of doing business (World Economic Forum 2011). Little diversification and low productivity in manufacturing A main reason for industry’s poor performance is its heavy concentration in low value-added consumer products, such as food, beverages, and textiles, which account for more than 70 percent of total exports. These goods are non- dynamic (Table 5.1), offer little possibility for technological improvement, and attract very little foreign direct investment (FDI). Industry has largely failed to move into more sophisticated capital goods or to develop upstream ancillary manufacturing, such as chemicals and engineering. Crucially, manufactur- ing focuses on products that are losing share in the world market. In 2010/11, the share of textiles in the country’s exports was about 50 percent (Planning Commission 2011), but as a share of total merchandise exports it dropped from 1.72 to 1.65 for 2009–11 (WTO 2012). Value added in industry (as a share of GDP) is much lower than that in comparator countries: it ranged from 21  percent to 27 percent in the 1970s and 1980s, constant at about 23 percent in 1994–2003, and around 27 percent since. India has had higher industrial value added since the early 1980s, and even low-income countries have recently over- taken Pakistan.7 Lack of trade competitiveness due to little manufacturing diversification Rapid growth in global manufacturing exports has allowed developing countries—notably excluding Pakistan—to gain a sizable share of the world Table Average product shares in manufacturing, Pakistan, 1970 –99 5.1 (percent) Product 1970–79 1980–89 1990–99 93 Food and beverages 30.45 30.94 22.89 Textiles 27.78 18.14 25.06 Industrial chemicals 11.20 14.29 15.50 Metals and nonmetals 9.10 14.20 13.20 Petroleum and coal 5.27 6.01 3.26 Electrical machinery 3.31 3.26 5.43 Transport equipment 2.99 2.89 3.05 Apparel, leather, and textiles 2.04 2.37 2.80 Nonelectrical machinery 1.84 2.14 2.09 Rubber and plastic 1.80 1.80 1.42 Source: Pakistan Economic Survey (various issues). market. The global share of exports from India, Malaysia, and Thailand, for example, increased considerably over 1974–2008, while Pakistan’s share remained stable for the period, though it declined over 1990–2008 (Table 5.2). Moreover, Pakistan exports a narrow range of manufactured products to which it adds little value. This can be captured by the PRODY index, which associates a certain income to each product, and the revealed comparative advantage (RCA), a measure that is greater than 1 if the country has a comparative advan- tage in producing a given good.8 The results in Table 5.3 suggest that, despite being a major producer of raw cotton, Pakistan does not have a comparative advantage in those textile products that fetch the highest prices. Its exports are instead concentrated in goods produced by low-income countries (raw cotton, cotton yarn, and cotton fabrics). Pakistan’s high RCA in these products is due mainly to government policies that have favored low value-added items over more sophisticated products, such as felt articles or bonded fiber fabric. Table Country export shares in world exports, 1974 –2008 5.2 (percent) 1974 1980 1990 2000 2008 India 0.56 0.43 0.57 0.70 1.32 Malaysia 0.55 0.74 0.94 1.61 1.43 Thailand 0.32 0.37 0.74 1.13 1.25 Pakistan 0.14 0.15 0.18 0.15 0.15 Source: United Nations Statistics Division n.d. REVitalizing manufacturing Table PRODY and revealed comparative advantage in textiles, 2008 – 09 5.3 Product PRODY ($) Revealed comparative advantage 94 Raw cotton 2,036 5.9 Cotton yarn 5,631 86.5 Cotton fabrics 4,541 115.0 Men’s and boys’ readymade garments 6,777 13.2 Women’s and girls’ readymade garments 5,160 5.2 Clothing accessories (knit wear) 9,419 22.5 Articles of felt (technical textile) 22,486 0.1 Bonded fiber fabric 31,250 0.2 Source: United Nations Statistics Division n.d. In sum, Pakistan’s export position in world markets is not very encouraging. This can be illustrated graphically: Figure 5.7 compares the growth rate of a given good in world exports with the corresponding growth rate in Paki- stan’s exports. If the product is in the top right-hand quadrant (the “com- petitive quadrant�), the product is internationally competitive. Pakistan has only 4.8 percent of its exports in that quadrant, compared with India’s 13.8  percent. Manufacturing exports are dominated by textiles (35.8  percent), whose world demand is falling. The underlying reason for Pakistan’s inability to achieve a dynamic competi- tive advantage is the persistent failure to diversify its production and climb up the technology ladder (Lall 2000).9In fact, in 2008 medium- and high-tech- nology products accounted for a mere 9.5 percent of Pakistan’s exports, but Competitiveness and performance of exports, India and Pakistan, Figure 2004 –11 5.7 India Pakistan 2 Annual growth of product’s share in 1 global exports (percent) Chemicals Pharmaceuticals Pharmaceuticals Chemicals 0 Iron and steel Iron and steel –1 Machinery and transport equipment Machinery and –2 transport equipment Clothing Clothing –3 –4 Textiles Automotive Textiles Of ce and telecommunications –5 products equipment –10 0 10 20 30 –10 0 10 20 30 Annual growth of country’s share in product’s global exports (percent) Source: Nabi 2011. 57.6  percent of world exports (Table 5.4). Moreover, the performance of medium-technology exports has not been encouraging, with growth below both the corresponding world growth rate and Pakistan’s total export growth rate. The growth of high-technology exports, by contrast, has been remarkable at 95 17.5 percent a year, although their share in Pakistan’s total exports is negligible and exceptionally low relative to other countries. Overreliance on remittances and official assistance further aggravates the above impediments. The evidence is considerable that remittances result in apprecia- tion of the equilibrium exchange rate, eroding manufacturing international competitiveness (Lartey, Mandelman, and Acosta 2008; Montiel 2006). Such appreciation also stimulates final consumption (of imports), further conspiring against a competitive local manufacturing sector. While the Pakistan rupee has experienced nominal depreciation since the 2008 global financial crisis, it has been largely overvalued for a long time (and remains so in real terms), effectively penalizing Pakistan’s international competitiveness, especially in activities that depend on medium-term investment like manufacturing. FDI in recipient countries has strong links to economic growth, but FDI can be inhibited by an unstable macroeconomic environment. FDI in the last two decades has come in the oil and gas sector and in the power, financial, and telecommuni- cations sectors, shifting away from manufacturing. Less than a fifth of Pakistan’s FDI in 2008 went to manufacturing, reflecting a declining trend that started in 2004. Due to economies of scale and technological spillover effects, FDI in man- ufacturing generates more benefits than in nonmanufacturing sectors of the economy, which in turn promotes economic growth (Wang 2009). To halt this declining trend, it is important to strengthen policies that reduce firms’ costs of Technological level of exports, Pakistan and the world, 1998 – Table 2008 5.4 (percent) Pakistan exports World exportsa Growth Share Share Growth Share Sector 1998–2008 1998–2000 2006–08 1998–2008 2008 Primary 10.1 12.3 12.7 11.2 11.4 Resource-based 23.9 3.5 10.9 10.5 14.8 Low-technology 8.2 74.7 66.7 9.2 16.2 Medium-technology 8.7 8.6 8.1 10.1 35.5 High-technology 17.5 0.8 1.4 9.3 22.1 Total 9.6 10.1 a. Excludes oil. Source: Pakistan Economic Survey (various issues). REVitalizing manufacturing doing business and that keep fiscal and monetary imbalances from becoming too large. 96 Lack of financing The financial market in Pakistan is shallow. Firms rely principally on retained earnings to finance their working capital and investment needs, ranging from 78 percent in Sindh to 88 percent in Khyber Pakhtunkhwa (World Bank 2009b). If firms draw on external finance, they usually resort to banks and trust funds (67 percent), or the Central Directorate of National Savings (22 percent). Further, access to finance is highly uneven across firms, and credit rationing hits small and medium-size enterprises (SMEs)—potentially some of the country’s most dynamic—making it hard for them to grow or improve pro- ductivity. SMEs also face barriers over weak and poorly enforced creditor rights, as well as high costs of borrowing. They are also are often perceived as risky borrowers, both because of their variable rates of return and their lower human and capital resources to withstand economic adversity. And their accounting systems and financial controls are often inadequate, under- mining the quality of their data. Onerous collateral requirements are particularly problematic (Bari, Cheema, and Haque 2005). Most banks require collateral exceeding 100 percent of the loan to mitigate asymmetric information. Yet the principal asset of manufac- turing firms is land, which is in effect removed from the collateral pool by a highly inefficient land market. The main problem is that land acquisition is cumbersome, involving multiple agencies, complex record keeping, and past sales transactions lacking valid conveyance documents. It takes 49 days to register a property and costs about 4.2 percent of its value (LUMS 2011). And vendors rarely transfer formal titles, thus preventing firms from accessing credit and preventing land from being put to its most efficient use. Social conflict and pollution The growth of manufacturing can also lead to public bads that, if not addressed, can act as bottlenecks in the form of social conflict and pollution. Social conflict, heightened by Pakistan’s demographic growth, manifests itself as sectarian or ethnic strife. Yet a growing urban population can become an important economic asset and lead to jobs growth, assuming that appropri- ate policies are in place to mitigate these negative impacts. And this can be a virtuous circle as agglomeration economies in urban areas can themselves mitigate the likelihood of social conflict. Social conf lict affects industrial production at several levels. First, lack of security creates a poor perception of Pakistan in international markets, and buyers in these markets have become skeptical about doing business with suppliers in the country. Second, a perception of lack of security causes the 97 private sector to reduce its economic activity, with implications for business property and assets. Based on a World Bank survey, the proportion of firms considering law and order a major constraint to their business rose from 22 percent in 2002 to 35 percent in 2007 (LUMS 2011). Industrial degradation of the environment presents health risks for the popu- lation and may have impacts for firms trying to access international markets. Ambient air pollution in medium and large urban centers is extremely serious but very little has been done to tackle it. High levels of dangerous pollutants, such as fine particulate matter and sulfur dioxide, are major health risks. At the firm level, particularly among manufacturers, corporate social and environmen- tal responsibility is becoming increasingly important to gaining and maintaining export markets in high- and upper middle-income countries. Some firms for which exports are important have market-friendly mechanisms to signal to their market that they are serious about this responsibility, as exemplified in attaining process certifications such as ISO 14001 (which covers environmental processes). Yet Pakistan (as well as India and Sri Lanka) are far behind China, for instance, which increased its ISO  14001 certifications nearly 60-fold over 2000–07 (Figure 5.8). Moreover, lax enforcement of occupational safety standards makes industrial hazards common. The Karachi garment factory fire in September 2012 was the worst in Pakistan’s industrial history—more than 280 people died—and such events harm the country’s manufacturing reputation. Figure Number of ISO 14001 certifications, selected countries, 2000 – 07 5.8 China India Malaysia Mexico Pakistan Sri Lanka 300 200 Thousands 100 0 2000 2001 2002 2003 2004 2005 2006 2007 Source: ISO Survey of Certifications 2004, 2007. REVitalizing manufacturing Policy Options A combination of the following policies may help improve Pakistan’s 98 ranking in competitiveness and manufacturing’s contribution to GDP. (Table 5.5 summarizes these suggestions.) Build a prudent macroeconomic framework Current nominal exchange rate depreciation linked to macroeconomic difficulties could easily swing toward nominal and real appreciation associ- ated with remittances and other capital inflows. Such appreciation may inhibit exports and thus potential industrial growth, making it critical to reduce fiscal and monetary imbalances in order to build a strategy for sustainable industri- alization. It also makes it all the more important to remove distortions in the credit market, trade policy, and tax regime. Finally, the government could facilitate FDI from Pakistanis abroad by enabling the creation of diaspora- focused venture capital and private equity funds in such areas as information technology, automobile-­ vending industry, medical equipment and services, and hospitality. Table Summary of challenges and suggested solutions 5.5 Sector or issue Challenge Solution Fiscal/trade (S) • Discriminatory tax regime • Make manufacturing activities attractive to invest • Fiscal and exchange rate in by not taxing them net when compared with fluctuations other economic activities • Mitigate by prudent macro policies and remove distortions in the credit market, trade policy, and tax regime Transport (S M L) • Poor transport sector • Efficiently connect cities and industrial clusters • Give increased attention to rail • Modernize trucking fleet • Redefine government’s role to focus on regulation and improve the private investment environment • Adopt new technology Power (S M) • Power shortages • Develop more and clean energy • Exploit power trade with neighboring countries Trade and logistics (S) • Weak cross-border infrastructure • Exploit Gwadar’s strategic location for gas pipeline Social development • Little skilled labor for small and • Improve skills training and education (S M L) medium‑size enterprises Government (M L) • Limited capacity • Strengthen institutional capacity of infrastructure • Burden on the environmental sector and environment agencies management framework • Combine command and control and market-based incentives to address environmental degradation S is short run; M is medium run; L is long run. Source: Authors. Address systemic issues Issues eroding manufacturing competitiveness include a discriminatory tax regime, poor access to credit, and ad hoc incentives such as statutory 99 regulatory orders (Nabi 2011; Planning Commission 2010). In 2009, 40 percent of firms stated taxes as being a barrier in doing business (though a decrease from 47 percent in 2002; World Bank 2009b). The bulk of the country’s total tax reve- nue is collected from manufacturing firms (corporate income tax) and their out- put (sales tax and various excises), while agriculture and most services are out of the tax net, which makes it more attractive to invest in nonmanufacturing activity. Taxes increase the cost of doing business and reduce incentives to invest in manu- facturing (Manes 2009). Pakistan’s tax system needs to be better balanced across economic activities and should be focused on taxing bads. In other areas, increasing access to credit, especially for SMEs, facilitating vocational training though industry chambers, and supporting firms’ certification in international standards of practices (ISO 14001, for example) would help raise manufacturing competitiveness. These wide approaches run counter to the government’s current incentives to manufacturing, which are generally ad hoc measures like statutory regulatory orders, often provided with- out a clear economic rationale. The government should instead focus efforts to improve the overall commercial environment for production and address market failures related to environmental and social degradation. Facilitate agglomeration through better connective infrastructure Cities and industrial clusters need to be better interconnected by upgrad- ing, extending, and rehabilitating infrastructure. In Khyber Pakhtunkhwa, Punjab, and Sindh, spatially connective infrastructure—particularly interre- gional transport infrastructure, as well as information and communication technology services—can help facilitate links between cities and along trade corridors. The planned rail and road infrastructure connecting Gwadar Port with northern Sindh is one example. Investments in freight transport should be developed, coordinated with efforts to establish or strengthen industrial clusters. Synergies between freight transport and cluster development would boost the structural and spatial transformations that Pakistan is undergoing. The potential of freight transport needs to be unleashed. Reforms should prioritize integrating different modes of transportation while emphasizing rail for long distances, where it is more efficient than road transport (adopt- ing a multimodal transport system); modernizing the trucking fleet to reduce environmental and social externalities; redefining the government’s role to focus REVitalizing manufacturing on regulating and attracting private investment and gradually eliminating the biases that distort the market; and fostering new technologies and procedures that add value to the trade and transport sector, including moving from the cur- 100 rent focus of bulk cargo to containerized cargo. As more reliable energy is critical to industry—indeed, energy shortages have hardest—the government should urgently develop new sources of hit this sector ­ energy supply. This entails, in particular, developing cleaner sources of energy as an alternative to its limited supply of gas sources, upgrading distribution networks, investing in thermal and hydro plants, using coal in a clean way as an alternative source of energy, and cooperating with other South Asian countries to exploit sources of energy, particularly gas and electricity from neighbors. Break down trade barriers and improve cross-border infrastructure for the major transport corridors and markets A 2007 study found that trade between India and Pakistan would increase by 405  percent if political and territorial conflicts were resolved.10 The strategic location of Gwadar should be exploited: it serves as an important node for any Iran–Pakistan–India gas pipeline that could be developed. Importing natural gas would partly address the energy concern, given that natural gas demand in Pakistan far exceeds supply—there would still be a supply gap after 2015 even if indigenous gas reserves and planned projects in pipelines in the country materialize (LUMS 2011). Improve skills training and education of workers to better match employment needs and supply in industry Such efforts are crucial given that many SMEs lack access to skilled labor. Pakistan is rapidly urbanizing, and reforms in trade and infrastructure are expected to contribute to this trend, particularly as investments and new employment opportunities will most likely materialize in urban areas. Although urbanization offers benefits, low skills and education can offset them. Strengthen the institutional capacity of provincial environmental agencies to implement appropriate regulatory frameworks Strengthening provincial environmental agencies should be a priority (particularly after devolution due to the 18th Amendment). Such agencies have little capacity to address the many environmental and social issues that arise during construction and operation of transport infrastructure. Raising their capacity to design and implement environmental and social regulations (that address market failures and diminish rent seeking) at the earliest plan- ning stages and to address issues as they arise would generate wide benefits for the population. These steps may also have long-term benefits for indus- trial growth and exports, though such growth—with its countrywide effects— 101 may burden the environmental management framework. Provincial agencies need to improve their capacity to address environmental and social impacts of industrialization as well. Notes The authors wish to acknowledge the research assistance of Ashma Basnyat and the comments provided by Jose R. Lopez-Calix, Javed Burki, and Anthony Cholst. 1. The literature on agglomerations is vast. See, for example, Fujita and Thisse (2002) and World Bank (2009). 2. It has been argued that Pakistan could leapfrog into service led growth. Yet to take full advantage of modern services and spread the benefits across society, Pakistan needs a well-educated and prepared labor force. Moreover, the very long–term future of the country’s economy may well be services. This note argues that the path to services for Pakistan would go through a dynamic manufacturing sector if there is to be a shared prosperity. The services sector in South Asia has also been extensively discussed in recent literature (for example, Ghani 2010). 3. These districts include Karachi, Lahore, Peshawar, Faisalabad, Sialkot, Islamabad, Multan, Swabi, Gujrat, Rawalpindi, Charsadda, and Gujran- wala (World Economic Forum 2011). 4. Including furniture and fixtures, scientific instruments, pharmaceuticals, apparel, handicrafts and office supplies, printing and publishing, pottery and china products, and paper and paper products (LUMS 2011). 5. Ministry of Finance 2011. In many developed countries, transport contributes 6–12 percent of GDP (Rodrigue, Comtois, and Slack 2009). In India, the trans- port sector contributed 5.7–6.4 percent of GDP in 1999–2005 (ADB 2007). 6. See Pakistan Economic Survey (various issues). 7. Even in textile-related industries, a move in the value chain may prove beneficial. For example, 50,000 pounds of cotton fiber creates 400 jobs in each of the three stages of the textiles value chain—spinning, weav- ing, and finishing of cloth. At the next stage (garments manufacturing), the same volume of fiber creates 1,600 jobs. In addition to being the most labor-intensive stage of the textiles chain, garments manufacturing creates by far the largest value addition and is the least energy- and capi- tal-intensive segment of the textiles chain (Nabi 2013). 8. The PRODY index is constructed as the weighted average of the per capita GDPs of the countries exporting a specific product, and thus REVitalizing manufacturing represents the income level (and productivity) of that product. The weights are the RCA of each country in each product (normalized to 1). If most high-income countries have an RCA in the export of a product, 102 the PRODY would be high. (The RCA is the ratio of the share of prod- uct A in Pakistan’s total exports to the share of product A in total world exports. If the RCA is greater than 1, it implies that the country has a comparative advantage in that product.) 9. The product concentration level is measured using the index where k is the number of products that account for more than 90 percent of Pakistan’s exports and Wi is the share of commodity i in total export earnings. The index can take a value between 0 and 1; the closer it is to 1, the greater the degree of concentration. Pakistan’s value is 0.40. 10. Naqvi and Schuler 2007. According to World Bank (2009a), in 2004 Pakistan and India engaged in the “Composite Dialogue� on peace and security issues, including terrorism and drug trafficking, confidence build- ing, economic and commercial cooperation, and friendly exchanges in various fields. References ADB (Asian Development Bank). 2007. “Profile of the Indian Transport Sector.� Manila. Bari, F., A. Cheema, and E. Haque. 2005. “SME Development in Pakistan: Ana- lyzing the Constraints to Growth.� Pakistan Resident Mission Working Paper 3, Asian Development Bank, Manila. Ellison, Glenn, and Edward Glaeser. 1997. “Geographic Concentration in U.S. Manufacturing Industries: A Dartboard Approach.� Journal of Political Econ- omy 105 (5): 889–927. Fujita, Masahisa, and Jacques-François Thisse. 2002. Economics of Agglomeration: Cities, Industrial Location, and Regional Growth. Cambridge, UK: Cambridge University Press. Ghani, Ejaz, ed. 2010. The Service Revolution in South Asia. Oxford, UK: Oxford University Press. ISO Survey of Certifications 2004, 2007. www.iso.org/iso/iso_catalogue/man- agement_standards/certification/the_iso_survey.htm. Lall, Somik. 2000. “The Technological Structure and Performance of Develop- ing Country Manufactured Exports, 1985–98.� Oxford Development Studies 28 (3): 337–69. Lartey, Emmanuel K.K., Federico Mandelman, and Pablo A. Acosta. 2008. “Remittances, Exchange Rate Regimes, and the Dutch Disease: A Panel Data Analysis.� Working Paper 2008–12. Atlanta, GA: Federal Reserve Bank of Atlanta. 103 LUMS (Lahore University of Management Sciences). 2011. “Industrial Policy, Its Spatial Aspects and Cluster Development in Pakistan.� Lahore, Pakistan. Manes, Eric. 2009. Pakistan’s Investment Climate Laying the Foundations for Growth. Volume 1: Main Report. Washington, DC: Poverty Reduction and Economic Management Department Finance and Private Sector Unit, South Asia Department, World Bank. Ministry of Finance. 2008. Pakistan Economic Survey 2007–08. Islamabad: Govern- ment of Pakistan. ———. 2011. Pakistan Economic Survey 2010–11. Islamabad: Government of Pakistan. ———. Various issues. Pakistan Economic Survey. Islamabad: Government of Pakistan. Montiel, Peter J. 2006. “Workers’ Remittances and the Long Run Equilibrium Exchange Rate: Analytical Issues.� Williams College, Williamstown, MA. Nabi, Ijaz. 2011. “Economic Growth and Structural Change in South Asia: Miracle and Mirage.� Development Policy Research Center Monograph. Lahore University of Management Sciences, Lahore, Pakistan. ———. 2013. “Garments Manufacturing as a Growth Driver in Pakistan.� Unpublished manuscript. Naqvi, Zareen F., and Philip Schuler, eds. 2007. The Challenges and Potential of Pakistan–India Trade. Washington, DC: World Bank. Planning Commission. 2010. “Private Sector Development Task Force Report.� Government of Pakistan, Islamabad. ———. 2011. “Pakistan: Framework for Economic Growth.� Government of Pakistan, Islamabad. Rodrigue, Jean Paul, Claude Comtois, and Brian Slack. 2009. The Geography of Transport Systems. London: Routledge. United Nations Statistics Division. n.d. United Nations Commodity Trade Statis- Database—UN Comtrade. New York. tics ­ Wang, Miao. 2009. “Manufacturing FDI and Economic Growth: Evidence from Asian Economies.� Applied Economics 41(8): 991–1002 World Bank. 2009a. World Development Report 2009: Reshaping Economic Geography. Washington, DC. ———. 2009b. Pakistan’s Investment Climate Laying the Foundations for Growth. Volume 1: The Main Report. South Asia Department, Poverty Reduction and Economic Management Department Finance and Private Sector Unit, Washington, DC. ———. 2012. World Development Indicators 2012. Washington, DC. REVitalizing manufacturing World Economic Forum. 2011. The Global Competitiveness Report 2011–2012. Geneva. WTO (World Trade Organization). 2012. International Trade Statistics. Geneva. 104 S. Amer Ahmed and Madhur Gautam Increasing Agricultural 6 Productivity Agriculture remains a socioeconomically and politically important sector in Pakistan, even as its share in the overall economy continues to fall. Sluggish growth in productivity has constrained farm income growth, limiting its potential for reducing poverty. There remains, however, substantial scope for accelerating broad-based agricultural growth to boost returns, which requires stimulating productivity growth through technology and innovation, better water use management, and the right trade policies. More specifically, the national agricultural research system requires fundamental insti- tutional reforms to make it more efficient and effective. For water, the most important intervention is institutional reform of the entire management system, including completing the devolution of authority to the appropriate scale with provision of sufficient resources and capacity building to the devolved authorities. Finally, improving agricultural trade will require removing discretionary instruments like the statutory regulatory orders and simplifying the trade regime. Distortions in domestic commodity markets also need to be removed. Food price stability is important, but it can be achieved through mechanisms more cost-effective than those in use today. Further, food security for the most vulnerable can be supported more efficiently through well-targeted social safety net programs. Pakistan’s real GDP has grown substantially over the past decade, at an aver- age of about 4.9 percent a year. Growth in agricultural value added has been lower at 3.3 percent a year. As is the case in transforming economies, transi- tioning from an agrarian to a developed economy is accompanied by a decline in the share of primary agriculture and a commensurate increase in the share of manufacturing and services. Pakistan has experienced this trend, with the share of agricultural value added in real GDP declining from 46 percent in 1960 to 26 percent in 2000 and to 21 percent in 2010 (World Bank 2011a). Nevertheless, agriculture remains a socioeconomically and politically important sector in Pakistan’s current transformation. Agriculture directly accounts for more than 40 percent of employment, but the sector’s contribution to overall employment is likely much higher, considering the downstream activities through the supply chains, transportation, and processing sectors to which it contributes (World Bank 2011a). Agriculture also contributes substantially, directly and indi- rectly, to foreign exchange revenue. Its exports account directly for more than Increasing agricultural productivity 11 percent of total exports, with exports of downstream industries like textiles accounting for more than another 40 percent (Planning Commission 2009). 106 While the poverty rate has continued to fall in the past decade, agriculture’s poor performance has limited its contribution to poverty reduction. The proportion of the population below the national poverty line declined from 34.7 percent in 2001/02 to 21.9 percent in 2005/06—and to 17.2 percent in 2007/08 (World Bank forthcoming). Of the 12.8 percentage point decline in the poverty headcount ratio between 2001/02 and 2005/06, growth in farm income accounted for 2.8 percentage points (Inchauste and Winkler 2012). When 2001/02–2007/08 is considered, farm income growth accounted for 3.2 percentage points. The Planning Commission (2009) estimates that average yields tend to be far below the achievable upper bound of progressive farmer yields.1 The national average yields of major crops like rice and wheat are only about 55  percent of progressive farmer yields, the highest achievable yields in Pakistan (Figure 6.1). The yield gaps are even greater for some commercial crops, like sugarcane in Sindh (73  percent). Despite the large potential for improve- ment, yield growth has slowed. For example, rice yields grew at an average annual 5.2  percent in the 1960s but just 3.2  percent in the 1990s and 1.7 percent in the 2000s. A similar pattern can be seen with wheat, which had average annual yield growth of 2.9 percent in the 1960s, 2.0 percent in the 1990s, and 1.1 percent in the 2000s. These crops and others represent major shares of the country’s crop pro- duction, and narrowing the yield gap for major cereals (rice and wheat) and Figure National average yields as a share of progressive farmer yields 6.1 75 1.8 2.6 2.1 50 2.9 50.0 Percent 55.0 25 0 Wheat Cotton Sugarcane Sugarcane Maize Rice (Sindh) (Punjab) Note: Numbers above columns indicate national average yields in tons per hectare. Source: Planning Commission 2009. for high-value crops (cotton and sugarcane) would boost agricultural GDP substantially (Figure 6.2). High-value agricultural products, in particular, have been increasing their contributions to Pakistan’s exports. Agricultural exports account for a quarter of annual export revenue, and their share is increasing 107 rapidly. In 1990, for example, dairy, eggs, and meat had virtually no exports (Hazell and others 2011). But in 2011, exports of dairy and eggs were valued at $30.1 million (real 2000 dollars), while those of meat and livestock were valued at $106 million.2 Fisheries grew 45 percent a year over 2008–11, while fruits, vegetables, and oilseeds collectively grew just 15 percent. Main Issues Sluggish Productivity Pakistan’s agricultural output growth rate has been decelerating, reflected in its declining total factor productivity (TFP) growth rate. 3 In the 1980s, average annual output growth reached 4.8 percent, making the country an international leader in agricultural growth. However, the rate slowed to a more modest 3.3 percent over the past decade, driven by more complex changes in input use and TFP. Figure Composition of agricultural production 6.2 Fishing 3% Other eld crops Forestry 1% 8% High-value crops 17% Livestock 52% Cereals 19% High-value crops Cereals Nonirrigated wheat 4% Sugarcane (irrigated) Cotton 28% (irrigated) Rice Irrigated 37% 39% wheat 57% Fruits and vegetables 35% Source: Debowicz and others 2012. Increasing agricultural productivity In earlier decades, TFP growth was responsible for substantial shares of output growth (Figure 6.3). For example, TFP accounted for 44  percent of output growth in the 1960s, 67  percent in the 1980s, and 37  percent 108 in the 1990s. By contrast, TFP now accounts for less than a fifth of the growth. Relative to other countries, agricultural TFP growth since 1990 and especially since 2000 has been very slow (Figure 6.4). Pakistan’s TFP growth has gone from among the best in the world in the 1980s to the lowest among such Asian comparators as Bangladesh, China, India, and Sri Lanka. Until the mid-1990s, Pakistan had even higher TFP growth than China and almost the same rate as B­ razil—two of the world’s outstanding long-term agricultural performers. Since the mid-1990s, however, Pakistan’s TFP Decomposition of output growth in agriculture by land, input Figure intensity, and total factor productivity, 1961–2009 6.3 Total factor productivity Input/area New land Irrigated 100 Percent of output growth 75 50 25 0 –25 1961–70 1971–80 1981–90 1991–2000 2001–09 Note: Land quality is adjusted by converting rainfed cropland, permanent pasture, and irrigated cropland into rainfed equivalents. Source: Fuglie 2012. Average annual total factor productivity growth for agriculture in Figure Pakistan and comparator countries, 1961–2009 6.4 Pakistan India Bangladesh Sri Lanka China 6 4 Percent 2 0 -2 1961–70 1971–80 1981–90 1991–2000 2001–09 Source: Fuglie 2012. has been f lat while the comparators (including high-performing Indone- sia, Thailand, and Vietnam) have done markedly better. With TFP growth progressively slowing, output growth has been driven increasingly by input use (fertilizer, labor, livestock, and machinery) and irrigation, highlighting 109 the importance of using these inputs efficiently and sustainably. Land ownership and poverty have a strong inverse relationship. The majority of the rural poor are landless or own very small plots. Anwar, Qureshi, and Ali (2004) estimate the poverty headcount of rural nonfarm households at about 48 percent, second only to that of landless farmers (55 percent) but greater than that of farmers with less than a hectare (32 percent). Poverty among farm households with more than a hectare is virtually absent. Moreover, agricultural performance is constrained by the highly unequal distribution of land in rural areas. In 2000, 61 percent of farm households owned less than 2.0 hectares of land (15 percent of total land holdings). Only 2 percent of households had holdings greater than 20.2 hectares, but in stark contrast they accounted for 30  percent of total land holdings (World Bank 2007). Yet past evidence suggests that land productivity on smaller farms may be higher than that on larger farms and that small farms generate higher profits per hectare than large farms (World Bank 2004). This suggests that greater access to land by small- holders leads to higher overall agricultural productivity. Further, the land mar- ket rigidities that perpetuate the historical inequity in land distribution also seriously affect the development of the nonfarm sector (Safavian, Aftab, and Shaikh 2013). Land is rarely bought and sold, so the status quo of unequal land distribution tends to hold, and land rental markets are highly inefficient. The inequality in landholdings by province remained mostly unchanged from the 1970s to 2000 (World Bank 2007). The low rate of transactions is largely attributable to high transaction costs and, possibly, speculative prices in excess of the dis- counted value of potential agricultural earnings from the land. An important constraint for the landless is a lack of access to credit due to a lack of collat- eral—land is the most commonly accepted collateral for formal loans. Given the land limitations, increased crop yields accounted for much of the past growth and can be attributed to major scientific breakthroughs in technology. Major contributions came during the Green Revolution, from the investments in agricultural research undertaken by the national agricultural research system. While helping achieve national food security, past research efforts focused primarily on technologies relying on modern inputs but did not attend to issues of sustainability and efficiency, such as inte- grated crop management, soil health, economical use of inputs and resources, Increasing agricultural productivity and the balancing of external input use with internal nutrient sources. About 52 percent of public research expenditure is on crops, 25 percent on natural resources, about 14 percent on livestock and fisheries, and about 9 percent on 110 social sciences (World Bank 2011c). Public agricultural research has historically been successful. Estimated internal rates of return from investments in it have ranged from 57 percent to 65 per- cent, with most of the returns from Green Revolution research (Ahmad and Nagy 2001). After a period of nationalization of large and medium-size private agribusinesses in the mid-1970s came a slow process of their denationalization and deregulation. Investment in private agricultural research and development (R&D) was thus curtailed severely for a long time, with only recent outreach to the private sector through programs like the Ministry of Science and Technolo- gy’s Science and Technology for Economic Development program. However, the current agricultural research system labors under severe techni- cal capacity constraints. Pakistan’s public investment in agricultural research has been on the decline, and in 2009 stood at about 0.21 percent of agricul- tural GDP and ranked at the bottom of agricultural R&D spending as a share of agricultural GDP in the region (Figure 6.5; ASTI 2012). The Planning Commis- sion (2009) notes that persistent funding constraints may have contributed to limited technology advancements (by limiting research activity)—for example, the wheat cultivar that has dominated production since 1991 is susceptible to rust disease. Another critical constraint is the limited human resource capacity: only 15 percent of agricultural research staff hold PhDs, lower than in the rest of South Asia (Beintema and others 2007). Qualified researchers are discouraged Agricultural research and development spending as a share of agricultural GDP in Pakistan and comparator countries, 1996 – Figure 2009 6.5 Pakistan India Bangladesh Sri Lanka Nepal 0.6 Percent of agricultural GDP 0.4 0.2 0.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: ASTI 2012. from public research agencies because of institutional disincentives, such as few promotion opportunities and low salaries. The technical capacity constraints are compounded by inefficiencies generated 111 by a complex institutional environment. Beintema and others (2007) identified 111 agencies involved in agricultural R&D, employing more than 3,600 research- ers and spending nearly PRs 2.4 billion (2000 PRs). Of these, 37 were federal agencies, 44 were provincial agencies, 17 were higher education institutions, and 13 were private entities. The Pakistan Agricultural Research Council (PARC) coordinates the activities of a large network of public national and provincial agricultural research bodies, institutes, and experimental stations. PARC does not conduct agricultural research itself, but it is responsible for the adminis- tration of the National Agricultural Research Centre (NARC). Since the 18th Amendment to the constitution passed in 2010, the public agricultural system has been devolving from the federal to the provincial level, creating new oppor- tunities for reenergized public agricultural research. With the research agenda moving down, agricultural research can potentially have a greater focus on the needs of local farmers and environmental conditions, though challenges of coordination, duplication, and cost-effectiveness could arise. Inefficient Water Use Management A critical factor in improving crop yields (besides technology) is water availability and the performance of irrigation. Pakistan’s irrigated land as a proportion of cropland is the highest in South Asia, with about 95  percent of arable land equipped for irrigation.4 Historically, Pakistan’s rate of irriga- tion expansion was the slowest in South Asia. Indeed, in the 1970s the average annual rate was 1.2 percent, against a South Asian average of 2.2 percent. In the past decade, however, Pakistan’s rate of irrigation expansion accelerated to 1.4 percent a year, exceeding the 1.1 percent South Asian average.5 But cropland expansion has been slow—a phenomenon common to the region. From the 1970s to the 1990s, Pakistan maintained a cropland expansion rate of 0.6  percent a year, even when such growth was almost stagnant in South Asia as a whole. In the past decade, though, the country’s cropland has been shrinking at an average 0.4 percent a year, faster than the 0.1 percent in South Asia as a whole.6 Despite the high irrigation intensity, average farmers’ access to water is less than it could be due to limitations of the water allocation system. At the farm level, access to canal water is determined by physical location along the canal and through the warabandi water allocation system of administratively set Increasing agricultural productivity rotations. Access to canal water then becomes contingent on access to land and the location of that land. There might not be enough water by the time it gets to land at the tail end of distributaries or watercourses, especially if upstream 112 farmers are accessing water illegally (Yu and others 2012). The irrigation system is highly inefficient, with steep seepage losses in almost every component of the delivery system. Of the 106 million acre-feet (MAF) of irrigation water that goes into the system, only 41 MAF reaches crops, a loss about 61 percent of the water delivered at the head. Of this 106 MAF, about 25 MAF are lost in watercourses and 17 MAF in fields, the most vulnerable components of the irrigation system (Yu and others 2012). Improving the efficiency of water use is therefore a high priority — not only to boost current productivity but also to mitigate the impending risks associated with climate change (Box 6.1). Yu and others (2012) simulate various climate change scenarios for 2020–80 and estimate the sensitivity of the 2008 benchmark economy to future climate risks. They find that, faced with future climate risks, overall GDP could decline 1.1 percent from its benchmark year value and agricultural GDP 5.1 percent. They also find that improving canal system efficiency to save just an additional 12 MAF could not only mitigate the Box Water management reform—promising but incomplete 6.1 Recognizing the importance of a robust and efficient water management system, the government pushed through reforms in the 1990s. These reforms restructured the public irrigation departments to provincial irrigation and drainage authorities; to area water boards to manage main and branch canals; and to farmers’ organizations and water user associations to manage distributor and minor canals. They aimed to enhance water use efficiency, streamline water resource management, and facilitate user participation. These reforms have not been completely successful, however, owing to problems at the provincial and local levels. Provincially, the devolution of autonomy from the irrigation departments is incomplete. For example, in Punjab’s public irrigation department and public irrigation and drainage authority, the secretary of the department is also the managing director of the authority. Another example is in Sindh, where even though the posts of irrigation department’s secretary and irrigation and drainage authority’s managing director are held by different people, the latter has a direct reporting relationship to the former. Locally, the farmers’ organizations do not have the resources or the capacity to fulfill their roles. Nor do they have input into the Abiana setting process, even though they might be responsible for collecting charges. Farmers’ organizations also vary widely in their role as charge collectors, because this role is determined by man- agement that these organizations may not have any voice in. The lack of clarity in the role of the farmers’ organizations and their widely varying mandates by local government has contributed to the Abiana col- lection inefficiencies that are damaging the system’s financial sustainability. Source: World Bank 2011b. adverse impact of climate change but further boost overall GDP 0.9 percent and agricultural GDP 4.2 percent, on average, relative to the benchmark. Financial unsustainability is another critical challenge to the irrigation system 113 under the current water management system. The canal irrigation manage- ment system recovers only a quarter of its annual operating and maintenance costs, with the shortfall expected to increase with rising costs and stagnant Abiana (water charges) per acre of crops irrigated (Planning Commission 2012). The collection rate of assessed Abiana is also low—at only 60 percent of assessed values. The Planning Commission (2012) estimates that the overall budget gap is about PRs 5.4 billion annually, with the system subsidized by the federal government. The current Abiana for different crops might also be distorting farmer decisions. The national average Abiana per acre in 2000–09 was PRs 126–214 for cotton, PRs 185–428 for sugarcane, PRs 125–210 for rice, PRs 69–136 for maize, and PRs 75–131 for wheat. But though rice requires 60 percent more water than cotton—both major export crops—their irrigation charges per acre are about the same, and so may not reflect the crops’ relative profitability, leading to possible overproduction of rice. Policy Distortions to Trade International and domestic trade are critical to improving agricultural pro- duction but face challenges to growth. Policy changes since 2006 have steadily eroded the effects of trade liberalization that Pakistan implemented over 1996–2003. During this period, the government simplified the tariff struc- ture and abolished its state trading monopolies for agricultural products. But it introduced exceptions in 2006 and reversed several of the more important liberalizing reforms in agriculture, particularly for wheat, sugar, and fertilizer. The use of statutory regulatory orders (SROs) has also expanded since 2006. SROs and new regulatory duties have been used to provide exemptions to nor- mal tariffs in some cases and to raise tariffs in others. The resulting trade regime has thus become highly discretionary and uncertain, leading to input- price distortions and highly variable output prices. The expanded ad hoc use of SROs also has fiscal implications, as preferential provisions provide the ben- eficiaries of the orders with special tax and duty concessions and exemptions, leading to a loss of potential tax revenue (Lopez-Calix and Touqeer 2013).7 In most years, major crops like wheat, rice, sugar, and cotton are implicitly taxed by the various price distortions introduced by policies. The policy-in- duced implicit tax on crop production serves to depress production, despite Increasing agricultural productivity implicit net input subsidies. Basmati rice, for example, had negative effec- tive rates of protection in 2008–10, when farm income would have been 21–40 percent higher under a no-intervention regime (Valdes and others 114 2012). The case of sugar is also illustrative: the surge in the world price of refined sugar raised the parity price, but the increase in the general sales tax applied to sugar offsets higher border prices. Sugar’s parity prices are roughly twice the observed farmgate prices, with this price wedge discour- aging production. The benefits of some domestic trade policies have also been unclear, as illustrated by the public procurement of wheat. Government procurement of wheat is extensive, involving federal, provincial, and district agencies. The government sets the procurement price with targets that the Pakistan Agricultural Storage and Services Corporation and provincial food depart- ments are responsible for meeting. Provincial governments (mainly Punjab and Sindh) and the corporation procure about 20  percent of total wheat production each year (Prikhodko and Zrilyi 2012). This federal and provin- cial procurement is absorbing the price transmission that would otherwise prevail in open markets, contributing to a price stabilization effect. Indeed, because the government controls domestic wheat prices and procurement volumes, as well as the international wheat trade, price transmission from world to domestic markets is minimal. The impact of these procurement policies on consumer welfare is ambiguous, and they can become fiscally unsustainable while also leading to perverse out- comes like subsidized exports. All procured wheat is bought and then sold to flour millers in the same wheat-marketing year, with the government absorbing the costs of procurement, storage, and financing. Millers can buy the subsi- dized wheat at below market prices and then sell the flour at open market prices, which would be the prices for consumers. This price stabilization role is perhaps one reason for the recent rise in wheat stocks, which has led to exports at subsidized prices in years of high wheat production. Policy Options Broad-based agricultural growth can be achieved through narrowing the wide yield gaps and diversifying toward high-value agricultural products. This growth can improve the agricultural incomes of farmers (and especially of smallholders), as well as improve rural incomes more generally, through higher returns on land and labor—the latter benefiting the many rural landless poor. Actions are needed in the following key areas to enhance agricultural growth and improve farm incomes (Table 6.1). Table Policy matrix 6.1 Objective Short-run action Medium- to long-run action Improve agricultural Initiate reform of the national agricultural Carry out reforms of the national agricultural 115 productivity research system to make it more efficient research system (clarify mandate, shift and effective personnel from federal to provincial Develop plans for building scientific research institutions, shift budget, provide appropriate capacity performance incentives) Increase budget for agricultural research Plan and implement long-run capacity-building program for scientific research capacity Improve water use Identify mechanisms for institutional reform Implement institutional reform efficiency of the management system: Provide sufficient federal and provincial • Complete devolution of authority to the resources for transition and capacity building appropriate scale (including provincial Set up a third-party watchdog to evaluate the and farmers’ organization) state of institutional reform and to monitor for • Clarify the roles and mandates of each rent seeking authority Remove protection Identify timetable for removal of statutory Remove statutory regulatory orders, reduce variability and bias regulatory orders, tariff reduction and and harmonize tariffs, and dismantle export against agricultural harmonization, and export-barrier removal barriers exports Identify World Trade Organization– compliant instruments that may be appropriate to use instead, such as special safeguard mechanisms Reduce distortions Identify minimum volume of public wheat Implement rules-based adjustable tariffs to in domestic grain procurement (federal and provincial maintain designated price bands markets while programs) Develop and roll out social protection protecting food Identify floor and ceiling prices to follow programs for food security, with clear triggers security world prices for wheat prices and graduation requirements Identify food-insecure groups for social protection programs Improve agricultural productivity Substantial reforms to the national agricultural research system are needed. First, the system requires fundamental institutional reforms to make it more efficient and effective. With efforts under way to develop provincial agricul- tural research institutions, the role of the PARC and the NARC needs to be adjusted to exploit their comparative advantage of being a federal insti- tution able to facilitate federal funding, intraprovincial knowledge, and capacity building. Second, with the shift in primary activities from federal to provincial levels and from policy coordination to agricultural research, there is a need to reflect these activities in human resource and performance incentives. This may require moving personnel from the center to provin- cial institutions, or even changing the composition of the staff, to increase the proportion of scientific research staff, for example. Third, these reforms will require additional spending in agricultural R&D, whether for supporting agricultural research in provincial research centers or capacity building of Increasing agricultural productivity science staff, with the exact composition of the additional spending depend- ing on the nature of the institutional reforms. 116 These reforms to the R&D architecture, by their nature, would be very wide ranging and require substantial groundwork prior to execution. The first step (of two)—a stocktaking of the current agricultural research system—would need to include a detailed institutional audit that examines the system as a whole and to clearly delineate the roles, functions, and mandates of the public federal and provincial bodies that govern and conduct agricultural research. More broadly, this stocktaking would also need to account for the current roles of (and environment for) private R&D, including those of domestic and mul- tinational agribusinesses. It should then lead to a strategic road map for over- hauling the national agricultural research system, with particular emphasis on future budgets, human resources, and capacity building. In keeping with the spirit of the 18th Amendment, this strategic planning would need to have the input and buy-in of provincial and local government institutions and should not be left to just the PARC and the NARC. The second step would be to roll out the appropriate reforms over the next one or two budget cycles. Improve water use efficiency The most important intervention would be institutional reform of the entire water management system. Given the system’s high dysfunction, clarifying the institutional environment would be a prerequisite for any other intervention under consideration, such as revising the Abiana. The reforms to the water management system include completely devolving authority to the relevant scale, clarifying the roles and mandates of each authority, and providing suffi- cient resources and capacity building to allow the devolved authorities to fulfill their mandates. As with the reforms to the national agricultural research system, reforms to the whole water management system will require action over multiple years and will need to be carefully considered. Water management systems show wide divergence in budgets, capacity, and extent of devolution from the fed- eral to provincial level. The reforms need to first identify their current state, from public irrigation departments down to the farmers’ organizations and water user associations, which will help clarify the roles and mandate of each authority and outline a devolution plan for each area where devolution has not occurred (such as the public irrigation and drainage authority still managing public irrigation departments). For entities that require capacity building and management reform (such as farmers’ organizations), budgets to train and support personnel are needed. Improve international trade in agricultural products The trade regime must be simplified. This will require removing unpredictable and discretionary instruments like the SROs, shifting to a lower set of uniform 117 tariffs, and simplifying the trade regime by removing alternative trade policy instruments like export taxes. These three measures would reduce uncertainty, volatility, and the policy bias against agricultural products like rice and sugar. Valdes and others (2012) also point out that equalizing tariffs across agricul- tural products, while necessary, is not sufficient for equal effective protection across products, because protection or support in the input markets could still be substantial, at varying levels. Their study argues that the best approach to reducing the variation in effective protection across outputs is to also reduce the variation in protection of all inputs, including raw materials, capital, and tradable inputs. From a practical perspective, the measures will require a realistic timetable, as well as instruments compliant with the World Trade Organization that may still be able to protect national interests. Improve domestic trade of agricultural products while protecting food security Distortions in domestic markets of commodities like wheat need to be removed. The simplest set of reforms would be to reduce the wheat procurement volume while designing and implementing complementary social safety net programs. The wheat procurement contraction would reduce the effective subsidy to wheat producers and thus the fiscal burden. If food price stability is important, price bands can be implemented using rules-based adjustable tariffs that set floor and ceiling prices to follow world prices.8 In parallel, social safety net programs that target food-insecure groups can be established, with clearly defined triggers and graduation requirements. Notes 1. “Progressive farmers� refers to farmers in Pakistan who have achieved high crop yields applying the available technology and management practices. They thus provide good benchmarks of what is currently achievable. 2. Estimated from the United Nations Commodity Trade Statistics database. 3. The statistics in this section are authors’ calculations based on Fuglie (2012). The TFP estimates in Fuglie (2012) are based on agricultural data on comparable output as well as inputs (labor, land, livestock, machin- ery, fertilizer, and land quality adjustments due to irrigation) for more than 170 countries. Since TFP estimates are sensitive to model specifica- tion and the level of aggregation of inputs included in the calculations, Increasing agricultural productivity the agricultural TFP estimates in Fuglie (2012) for Pakistan may differ from the estimates from other studies at a more aggregated level, such as Lopez-Calix, Srinivasan, and Waheed (2012), which takes a more mul- 118 tisectoral perspective and uses different input definitions. The trends in both studies are similar, however—rising in the 1980s and declining thereafter. 4. All estimates in this paragraph are authors’ estimates based on Fuglie (2012). 5. Pakistan has an irrigation potential of 21.3 million hectares of land, of which 19.3 million are equipped for irrigation: 35.9 percent for surface water, 21.4 percent for groundwater, and 41.3 percent for a mix of surface and groundwater (FAO 2010). 6. All estimates in this paragraph are based on Fuglie (2012). 7. Agricultural incomes are not taxed, and taxing farm income has the potential to increase the tax base and increase revenue. However, Aftab and others (2010) suggest that the poverty impacts of taxing agricultural incomes may be negative, as the tax incidence will be sensitive to the income distributions across agricultural households. Agricultural income taxes may potentially increase poverty if applied too broadly to all house- holds instead of only large farmers. 8. Valdes and others (2012) suggest this approach for wheat and sugar, based on the moving averages of border prices. By keeping the basis of the price band delinked from domestic prices, the policy remains a World Trade Organization–­ compliant instrument as a variable levy. References Aftab, Safiya, Caesar Cororaton, Sohail Malik, and Hans G.P. Jansen. 2010. “Domestic Terms of Trade in Pakistan—Implications for Agricultural Pricing and Taxation Policies.� Report 68404, World Bank, Washington, DC. Ahmad, Mumtaz, and Joseph G. Nagy. 2001. In Private Investment in Agricultural Research and International Technology Transfer in Asia, ed. Carl E. Pray and Keith O. Fuglie, 56–75. Agricultural Economic Report AER-805. Washington, DC: U.S. Department of Agriculture, Economic Research Service. Anwar, Talat, Sarfraz K. Qureshi, and Hammad Ali. 2004. “Landlessness and Rural Poverty in Pakistan.� The Pakistan Development Review 43 (4): 855–74. ASTI (Agricultural Science and Technology Indicators. 2012. ASTI Data Tool. www.asti.cgiar.org/data/. Beintema, Nienke M., Waqar Malik, Muhammad Sharif, Gert-Jan Stads, and Usman Mustafa. 2007. Agricultural Research and Development in Pakistan: Policy, Investments, and Institutional Profile. ASTI Country Report. Washington, DC: International Food Policy Research Institute. Debowicz, Dario, Paul Dorosh, Sherman Robinson, and Syed Hamza Haider. 2012. “Implications of Productivity Growth in Pakistan: An Economy Wide Analysis.� Pakistan Strategy Support Program Working Paper 2. International Food Policy Research Institute, Washington, DC. 119 FAO (Food and Agriculture Organization of the United Nations). 2010. “AQUASTAT Pakistan Country Profile.� Rome. Fuglie, Keith O. 2012. “Productivity Growth and Technology Capital in the Global Agricultural Economy.� In Productivity Growth in Agriculture: An Inter- national Perspective, ed. Keith O. Fuglie, Sun Ling Wang, and V. Eldon Ball. Oxon, UK: CAB International. Hazell, Peter B., Derek Headey, Alejandro N. Pratt, and Derek Byerlee. 2011. “Structural Imbalances and Farm Non-Farm Employment Prospects in Rural South Asia.� Background study for More and Better Jobs in South Asia, World Bank, Washington, DC. Inchauste, Gabriela, and Hernan Winkler. 2012. “Decomposing Distributional Changes in Pakistan.� Background paper to Fostering Employment with Inclusive Growth Report, World Bank, Washington, DC. Lopez-Calix, Jose, Raj T.G. Srinivasan, and Muhammed Waheed. 2012. “What Do We Know about Growth Patterns in Pakistan?� Pakistan Policy Notes Series, World Bank, Washington, DC. Lopez-Calix, Jose, and Irum Touqeer. 2013. “Mobilizing Revenue.� Pakistan Pol- icy Note 16, World Bank, Washington, DC. Planning Commission. 2009. Task Force on Food Security—Final Report. Islamabad: Government of Pakistan. ———. 2012. Canal Water Pricing for Irrigation in Pakistan—Assessment, Issues, and Options. Islamabad: Government of Pakistan. Prikhodko, D., and O. Zrilyi. 2012. “Review of the Wheat Sector and Grain Storage Issues.� Agriculture Policy Priorities Work in the Islamic Republic of Pakistan, World Bank, Washington, DC. Safavian, Mehnaz, Sarwat Aftab, and Sarmad Shaikh. 2013. “Enhancing the Business Environment.� Pakistan Policy Note 3, World Bank, Washington, DC. United Nations Statistics Division. n.d. Commodity Trade Statistics database. http://comtrade.un.org/db. Valdes, Alberto, Mehwish Ashraf, Ashraff Khan, and Muhammad Waheed. 2012. “Agricultural Trade and Price Policy in Pakistan.� Agricultural chapter of Country Economic Memorandum, World Bank, Washington, DC. World Bank. 2004. Rural Factor Markets Policy Reforms for Growth and Equity. Report 30381-PK. Washington, DC. ———. 2007. Promoting Rural Growth and Poverty Reduction. Report 30381-PK. Washington, DC. ———. 2011a. World Development Indicators 2011. Washington, DC. Increasing agricultural productivity ———. 2011b. “Pakistan Water Resource Notes.� Internal discussion note, South Asia Agriculture and Rural Development, Washington, DC. ———. 2011c. “Pakistan Agriculture Research and Extension Notes.� Internal dis- 120 cussion note, South Asia Agriculture and Rural Development, Washington, DC. ———. Forthcoming. Pakistan Poverty Assessment. Washington, DC. Yu, Winston, Yi-Chen Yang, Andre Savitsky, Donald Alford, Casey Brown, James Wescoat, Dario Debowicz, and Sherman Robinson. 2012. The Indus Basin of Pakistan: The Impacts of Climate Risks on Water and Agriculture. Washington, DC: World Bank. Cem Mete and Rehan R. Jamil Enhancing Labor 7 Market Conditions for Vulnerable Groups For the current generation of working-age adults, public interventions to increase female employment will require policies that address the social, cultural, and security constraints facing urban and rural women in entering labor markets. In the long term, targeted invest- ments to improve female education and health can enable more women to secure higher paying jobs. Child labor, however, will require a more strategic approach, going beyond legis- lation to eliminating the worst forms of child labor and increasing school enrollment for poor and vulnerable households through targeted cash transfers. Long term, child labor can be eliminated only by investments in human capital; short term, by incentives and penalties for private and informal employers. Because Pakistan has so many informal workers, skills-enhancing policies are needed to allow them to graduate into better paid formal jobs and to ensure that they receive basic social protection. Programs providing market-oriented skills hold promise, though they need to be evaluated carefully before being implemented on a large scale. Active labor market programs such as short-term public works programs that hire seasonal laborers, might also be considered for vulnerable workers, primarily postcrisis. As Pakistan seeks high and sustainable economic growth, increasing employ- ment and productivity will be crucial. The labor force is poorly educated, highly segmented by gender, and concentrated in low value-added informal sector activities. It thus presents a serious constraint on economic growth and poverty reduction. Although unemployment is relatively low at around 5  percent, underemployment is high. Many workers work part-time or contractu- ally in the informal sector, which makes up 56–73 percent of nonagricultural employment (Cnobloch and Salam 2012).1 The country’s labor force participa- tion rates for women are among the world’s lowest, at 28 percent, compared with 82 percent for men (World Bank 2013). Household survey evidence reveals that many women and youths from low-income households find temporary and low-paying employment in the informal sector, with minimal job security. Pakistan’s population is young—and will remain so in the short term. Total fertility rate, at 4.1, remained well above the South Asia average of 2.8 in 2006–07 (Figure 7.1). Although the total fertility rate declined from 7 in 1970 to 3.9 in Enhancing labor market conditions for vulnerable groups Figure Selected South Asia fertility rates, 2006 – 07 7.1 Fertility rate, total (births per woman) 5 122 4 3 2 1 0 Pakistan Bangladesh India South Asia average Source: Data for Pakistan are from MEASURE DHS (2007); all other data are from the World Development Indicators database. 2009, it is still high. The youth cohort (ages 15–24) and adolescent cohort (ages 10–14) are the most populous segments of the population (World Bank 2011a). Investing in youth education and health will thus be crucial for the future competitiveness of labor markets and other social outcomes. Without these investments in health and education, especially for girls, the economy’s competi- tiveness will be seriously compromised. And instead of a “demographic dividend,� Pakistan will face a “demographic bust� (World Bank 2011a). Pakistan’s low human development indicators undermine its labor force productivity and economic growth. In 2009, the country’s public spending on education as a share of GDP was 2.7 percent and on health 2.2 percent— placing it among the world’s lowest spenders on these two sectors (World Bank 2012b). This needs to change. Improvements in education and health enhance labor outcomes, such as employment rates and productivity, which stimulate economic growth. Cnobloch, Jamil, and Chowdhury (2012) use cross-country data to document the links between selected human development outcomes and GDP over the last two decades. They draw two main conclusions: the projections of how human development indicators might evolve under differ- ent economic growth scenarios in the short term indicate that even under the best-case scenario Pakistan would not catch up with the human development outcomes of most developing countries; and if the short-term human devel- opment projections were to be realized, long-term economic growth would suffer. In other words, without major government reforms and investments in health and education (thus deviating from the trends of the last two decades), Pakistan will not only miss its Millennium Development Goal targets but also Sheikhupura Sialkot Hyderabad Karachi Sukkur Quetta Peshawar Islamabad FaisalabadGujranwala Lahore Multan Rawalpindi hinder its long-term economic growth potential. Main Issues The key policy issues fall under three headings. First is low formal female employment and high underemployment. Second is lack of basic social 123 protection and limited access to education and training among informal sector workers. And third is high levels of child labor and uneasy transition of youths from school to work. Low formal female employment and high underemployment Most women entering the labor force work in agriculture, often without pay. Of the more than 3 million jobs for women created between 2001/02 and 2005/06, 72 percent were in unpaid family work in agriculture, and only 19 percent were paid nonfarm jobs (World Bank 2011a). Female unpaid family workers in agriculture typically live in rural areas, work on family-owned farms, or work as sharecroppers. They also have the traditional responsibility of taking care of household chores and childrearing, beyond their agricultural work. There is a wide urban–rural gap among women entering the workforce. In rural areas, nearly 1 in 3 women work, against only 1 in 12 in urban areas (Figure 7.2). Although women in urban areas are better educated than their rural counterparts, their employment rates are much lower. Urban women’s labor participation was just 9 percent in 2001/02, and by 2007/08 had declined to 7.6 percent. The low rate of high-skilled women entering the workforce is reflected in their minimal representation in senior or manage- rial positions (only 6 percent; Nikitin 2011). Figure Labor force participation, ages 18 –59, 2001– 07 7.2 Rural men Urban men Rural women Urban women 100 75 Percent 50 25 0 2001 2003 2005 2006 2007 Source: Nikitin 2011. Enhancing labor market conditions for vulnerable groups Although female labor participation rates are low across South Asia, Pakistan’s rates are among the lowest (IFC 2007). Bangladesh, India, and Pakistan, which together account for 95 percent of the region’s working-age population, also 124 have the lowest rates of female participation: 31 percent for Bangladesh, 30  percent for India, and 22  percent for Pakistan (World Bank 2011b). By province, working-age female employment rates are much higher in rural areas than in urban areas (Table 7.1). Unemployment is low overall but far higher for urban women. Despite the decline from 12  percent to barely 9  percent over 2001–07, unemployment among urban women remained three times that of rural women and more than four times that of urban men (Figure 7.3). Similarly, unemployment among women ages 15–24, at 10.5 percent in 2008, was far higher than among young men, at 7.0 percent (United Nations Statistics Division n.d.). Low schooling levels for women limit their earnings potential. Because women’s educational opportunities are lower than men’s (in both urban and Table Working-age female employment, by province, 2007/08 7.1 (percent) Punjab Sindh Khyber Pakhtunkhwa Balochistan Urban, all 10 5 6 5 Urban, paid 8 4 5 3 Rural, all 31 27 18 12 Rural, paid 11 1 6 1 Source: World Bank staff estimates based on Pakistan Bureau of Statistics (2008). Figure Unemployment, ages 18 –59, 2001– 07 7.3 Rural men Urban men Rural women Urban women 15 10 Percent 5 0 2001 2003 2005 2006 2007 Source: Nikitin 2011. rural areas), women entering the labor force tend to be less educated and lack specialized skills. Women are much less likely than men to have secondary or higher education. For Pakistani men, education is positively associated with entering the labor force; for women, however, education at the primary and 125 lower secondary levels is associated with lower labor force participation. Only higher education (at the college level or beyond) is positively correlated with female labor force participation (Nikitin 2011). With minimal education, the majority of rural women entering the workforce typically work in low-paying or unpaid jobs in agriculture, while urban women work in low-paying or part-time jobs in manufacturing or services. A small but growing number of educated urban women are entering the rapidly growing services sector. Overall, however, the number of educated professional women entering the workforce remains very small. High fertility rates often mean heavy child care duties for females. Pakistan presents much evidence that higher fertility is associated with lower investments in human capital per child. Rural areas have far higher fertility rates and lower use of contraception than urban areas. The availability of labor- intensive jobs in agriculture makes large families economically viable. In urban areas, however, the higher cost of child rearing and the demand for education in labor markets is slowly influencing parental choices on fertility rates, as is greater female literacy (World Bank 2011a). In urban and rural areas, the burden of household work and child rearing falls on women, especially in poor households where domestic help is unavailable. Extended family networks sometimes act as informal social safety nets for working women. In the absence of facilities for these women, such as day care, extended family support systems often fill the void, though not all working women have access to them. In addition to house work, limited formal arrange- ments for child day care and elderly care have detrimental effects on female employment opportunities. Married women are expected to look after both their children and their parents-in-law. The uneven division of household labor makes the opportunity for full-time employment very difficult for many women, particularly poor women who cannot afford the costs of day care even when it is available. Working women are often expected to maintain their household responsibilities, which has a negative impact on their well-being (Kabeer 2008). Traveling long distances for work is not always feasible for women. Despite wide regional and class variation, the social tradition of veiling and segregating women from nonfamily men—pardah —restricts certain groups of women’s access to public spaces in socially conservative areas.2 In rural areas, Enhancing labor market conditions for vulnerable groups the closeness of homes to agricultural areas where women work allows a greater degree of freedom to work, while maintaining the social norms of adhering to pardah. For many urban women, however, the lack of easily accessible female- 126 friendly workspaces severely reduces employment options. The sectoral composition of employment and high rates of unpaid female labor hurt female employment outcomes. Most unpaid female workers are in the agricultural sector and live in rural areas: 78  percent of female workers are unpaid. Ranking a distant second is the wholesale and retail sector, where 44 percent of female workers are unpaid (World Bank 2011a). Female unpaid family workers tend to have little or no education (lower than grade 1): in 2005, two-thirds of working women in this education group were unpaid family workers, compared with only 7 percent of working women with grade 11 education or higher. Only 20 percent of urban workers were unpaid, compared with 68 percent in rural areas (World Bank 2011a). Working men and women display a wide earnings gap. A woman with similar qualifications to a man earns far less. The services sector, often a large source of employment for skilled women in developing countries—though ­ rowing—has been unable to attract high numbers of skilled female rapidly g professionals. A lack of human capital investment in women has resulted in many women from poor and low-income households taking jobs in the informal sector, which pay little and do not require high levels of skills or education (Nikitin 2011). Having said that, Pakistanis—both women and men—enjoy respectable returns to education, with a 7–12 percent increase in earnings for an additional year of schooling (Figures 7.4 and 7.5). Figure Inflation-adjusted wage levels, 2001– 07 7.4 Men (nominal) Men (in ation adjusted) Women (nominal) Women (in ation adjusted) 30 Percent (base year, 2001) 20 10 0 2001 2003 2005 2006 2007 Source: Nikitin 2011. Returns to education, change in earnings for an additional year of Figure schooling, 2001– 07 7.5 Rural men Urban men Rural women Urban women 15 127 10 Percent 5 0 2001 2003 2005 2006 2007 Source: Nikitin 2011. Lack of basic social protection and limited access to education and training among informal sector workers The top five constraints reported by urban formal sector firms3 in Pakistan are tax administration, electricity, political instability, government policy uncertainty, and courts (in order of importance; World Bank 2012a). These constraints are important both for the formal firms surveyed and because they suggest reasons why 63–72 percent of workers are employed by informal enterprises (Box 7.1). Annual growth in Pakistan’s total factor productivity lags well behind South Asia’s better-performing countries (Figure 7.6). This outcome is unsurpris- ing given how this highly informal economy relies on workers with little education and not always in good health who are vulnerable to various shocks, constrained by poor infrastructure, and heavily concentrated in agriculture. Many of the large numbers of young people entering the workforce lack the entrepreneurial and technical skills that formal firms need, particularly the larger, more specialized firms. The average worker has only three or four years of schooling. The problem is especially acute in manufacturing, where two-thirds of employees have fewer than three years of schooling. In services, half the workers are educated above grade 6 (Nikitin 2011). IslamabadFinance An International Corporation FaisalabadGujranwala survey Lahore Multan Rawalpindi of Hyderabad Sheikhupura Sialkot small and Karachi medium-size Sukkur Quetta Peshawar firms in Pakistan found that less than 6  percent of the youths surveyed had any technical or vocational skills. Only 2.5 percent reported on-the-job-training, and most of the new low-skilled labor is absorbed by the informal sector (IFC 2007). Enhancing labor market conditions for vulnerable groups Box The informal sector and labor markets 7.1 To understand Pakistan’s labor markets, it is important to look beyond unemployment rates and con- sider the quality of employment. Many employed individuals find it difficult to find sufficient or secure 128 employment. Cnobloch and Salam (2012) estimate that 72 percent of all workers were in informal enterprises in 2007/08. Using an alternative definition based on type of employment, in which casual, piecemeal, own-account, and unpaid family workers in the nonagricultural sector are in informal employ- ment, informality is an estimated 63 percent, but because informal enterprises tend to be small and concentrated in low value-added activities, their share in national output is only 35–40 percent. Individuals in the informal sector come from the full spectrum of economic backgrounds, but most are uneducated and live in rural areas. Punjab, the most populous province, has the highest number of informal workers. Some workers switch back and forth between the formal and informal sectors. As individuals age, they can transition to the formal sector, so some aspects of informality are compatible with economic mobility. The informal labor market is segmented, especially along the urban–rural split. In urban areas, nearly 40 percent of informal workers come from the top 20 percent of the welfare distribution; in rural areas, only 8 percent do. With fewer opportunities in the formal sector, informality is the next best option for younger people, especially those in urban areas and with less schooling, despite the lower wages. The informal sector also attracts a greater share of poorer women, while men come from a variety of income levels. Source: Nikitin 2011; Cnobloch and Salam 2012. Annual growth in total factor productivity in Pakistan, Thailand, Figure India, and China, various periods 7.6 Reallocation Services Industry Agriculture 4 Total factor productivity 3 2 1 0 Pakistan Thailand India China China (1980–2008) (1977–96) (1980–2008) (1978–93) (1993–2004) Source: World Bank 2012a. Formal firms often prefer short-term contractual arrangements with workers. Short-term hiring is a common trend in Pakistani labor markets, with an average temporary contract of around five months. Of manufacturing employees, 30 per- cent have short-term employment arrangements, or more than double India’s rate (13 percent) and about Sri Lanka’s (28 percent; Nikitin 2011). High levels of child labor and uneasy transition of youths from school to work Child labor remains widespread in Pakistan, despite some success in reducing it in certain industries. In 1998, 16.3 percent of children ages 10–14 129 were working (ILO 1998). In this age group, 73 percent (2.4 million) of child laborers were boys while just 27 percent (0.8 million) were girls. Children in rural areas were eight times more likely to be economically active than those in urban areas. About 71 percent of working children were involved in unskilled occupations in agriculture and manufacturing, and about 70  percent of working children were unpaid family helpers (75 percent in rural areas and less than 33  percent in urban areas). More recent work confirms the high prevalence of child labor, estimated at 13.7 percent for ages 5–14 in 2004 (ILO 2006). U.S. Department of Labor (2002) statistics further show that 13 percent of the children worked 56 or more hours a week, and about 7 percent suffered from frequent illnesses or injuries. Despite government commitment to outlaw child labor, implementation lags far behind. In its national strategy to combat child labor, the govern- ment has committed to outlaw child labor by focusing on eliminating the worst forms of child labor prevalent in certain industries and by expanding education opportunities at the primary and secondary levels, to keep children out of the workforce. Yet a large informal economy enables employers to easily employ children for minimal pay and without record of their employment or need to adhere to government regulations. From a demand perspective, children remain a cheap supply of labor, working for paltry amounts with few benefits. From a supply perspective, poverty remains the primary reason why poor households have children work instead of attending school. Deficien- cies in enforcing regulations and high informal employment thus limit the effectiveness of policies to reduce child labor.4 Credit constraints and an uncertain future economic environment serve as disincentives for poor households to invest in their children’s schooling, which can make child labor a more attractive option. In 2008–10, households experi- enced several severe external shocks, such as spikes in food prices in 2008 and floods in 2010 (World Bank 2010). The Pakistan Social and Living Standards Measurement Survey data from this period show that poor households responded by cutting spending and, where possible, expanding adult and child labor participation (Nikitin 2011). Poor households that reported shocks in 2009/10 saw higher student dropout rates—either to enter labor markets or to reduce household expenditures. The drop in spending on education was far higher for girls than boys. Thus, even when returns to education are high, poor households may not invest enough in schooling their children (World Bank 2011c). Enhancing labor market conditions for vulnerable groups A girl is much less likely to enroll in school if her mother had not attended school or if her father works in agriculture (Lloyd, Mete, and Sattar 2005). For both sexes, the levels of community development and opportunities have a 130 positive impact on primary-school enrollment: in communities that have many schools, parents are more likely to invest in education. This makes investment in schools in remote areas with few public or private schools even more critical. Pakistan’s youth employment challenge would already be addressed to a large extent if the country made progress in tackling the female employ- ment and informal sector employment. Women between ages 15 and 24 also explain a large share of those who are not in education employment or training (ILO 2012). The share of women not in education employ- ment or training in Pakistan was 66.4 percent in 2005, the second highest among developing countries, compared with 10.9 percent of men, better than other countries. The quality of employment also matters. For informal sector employment, large numbers of youths entering the workforce lack the entrepreneurial and technical skills that formal firms need. The main reasons for not working for youths also differ significantly between men and women. For men, even if one focuses only those between ages 20 to 24, 70 percent do not work because they are students and not willing to work. But for women, only 10 percent do not work for these reasons. Instead, more than 88 percent of women indicated housekeeping as the main reason for not working. Policy Options The policy options for improving the labor market outcomes of the most vul- nerable groups are summarized under three subheadings. Many of them are complementary and will require an integrated approach that specifically targets the employment and social protection needs of poor and vulnerable workers, a large portion of them women and children. Improve conditions for women’s employment Public provision of, or subsidies for, child care can reduce the costs that poor women incur at home when they enter the labor market, particularly in urban areas. Urban and rural women report household responsibilities as one the primary reasons for being unable to seek employment. Although extended families play an important role in providing child care for working women, the absence of high-quality public or private day care facilities in major cities is a large barrier for women’s access to labor markets. Publicly provided or subsidized day care can be found in many Latin American countries, including Argentina, Brazil, and Colombia. Colombia’s Hogares Comunitarios program provides subsidies to designated homes turned into community child care centers, and its provision of affordable child care in low-income neighbor- hoods has greatly increased mothers’ participation in the labor market as well 131 as their hours worked (World Bank 2012b). Legal reform to reduce gender discrimination in the workplace is necessary to increase female employment rates—but not sufficient. Affirmative action programs to increase the entry of women into wage employment and their advancement on the job once they are employed is one way to increase the hiring of women. Women’s presence in public sector jobs remains low, and the public sector can take the lead in affirmative action—with incentives such as tax breaks for private firms that follow this example. Skills development and microfinance focusing on urban female work- ers and urban youths to increase their assets might be scaled up after due documentation of their impacts on employment outcomes. The Benazir Income Support Programme (BISP), Pakistan’s flagship social safety net program, has taken the lead by targeting cash transfers toward poor women. The program is piloting poverty graduation programs, including skills building and access to microfinance programs for beneficiary households. Evidence from other countries suggests that skills programs that include an employment-matching component can help graduates find jobs. In Latin America, for example, a number of youth employment programs, such as Peru Proven (Peru) and Jóvenes en Acción (Colombia), contract with decentralized training entities to offer vocational courses in which beneficiaries can enroll. The courses, offering class- room and on-the-job training, match local firms’ needs with the content of the training curricula. These two programs are successful because they are well targeted, demand driven, and linked to private labor demand. Not all skills programs work as intended, however, and such interventions should first be piloted and then evaluated in Pakistan. Investments in infrastructure help connect women and vulnerable youths to markets and reduce the time spent doing household chores. Evidence from the developing world suggests that improvements in basic infrastructure services—especially water and electricity—can help free women’s and children’s time and increase female labor force participation (World Bank 2012b). Access to water closer to households frees up women’s time, especially in rural areas where women often spend much time collecting water. A study from Pakistan showed that water sources closer to home were associated with increased time allocated to labor market work (Ilahi and Grimard 2000). Similarly in Bangla- desh, upgrading and expansion of rural roads led to increased labor supply Enhancing labor market conditions for vulnerable groups for men and women, while raising household incomes (Khandker, Bakht, and Koolwal 2006). Finally, transport facilities targeted at working women, such as buses to take them directly to and from their workplaces, can foster a safe and 132 convenient environment to encourage women to find employment. Implement measures to support informal workers Pilot skills programs that target informal workers need to be evaluated. Public skills training programs in Pakistan, as across South Asia, generally favor formal workers. Such programs have had limited success, in part because they reach very few workers and cannot match the skills demanded by job markets. Success- ful skills training programs are integrated programs that include both on-the- job and classroom components (World Bank 2013). As many poor and informal workers have little education, courses will need to be tailored to their needs. In addition, skills or vocational training providers need incentives to ensure that the training they offer is relevant for employers. Publicly subsidized training agencies are often out of touch with the changing demands and preferences of both firms and job seekers. An alternative approach is for public training funds to be directed to private and nonprofit providers on a competitive basis. Performance-based tendering can create incentives for more relevant training courses, while contracting can be designed so that the tough- est-to-reach groups do not lose out (World Bank 2013). Active labor market programs, such as public works programs, can help provide employment for poor informal workers, such as laborers, particularly in the offseason. The cost effectiveness of existing large-scale public works programs in South Asia is debatable, but such programs that are well designed and implemented can mitigate the negative impacts of economic downturns and fill the gap when employers or workers themselves underinvest in training. Public works programs can also promote skills building, if vocational training is built into their design. Such programs can also facilitate community partici- pation, particularly in postcrisis areas with high unemployment but requiring labor for reconstruction and infrastructure. Geographically targeted districts in the Federally Administered Tribal Areas, Khyber Pakhtunkhwa, and Baloch- istan may be suitable for well-designed small public works programs to help with reconstruction. Pakistan needs laws to recognize the rights and assets of informal workers. Laws governing property rights and the titling of assets of the poor are a cause of informality in many developing countries. In some cases, the poor hold assets, but these assets represent “dead capital� because they are held in forms not recognized by most legal systems and cannot be used to generate productive capital that the poor could then use for formal economic activities (de Soto 2003). Legal reforms to grant property rights to informal property owners and business vendors must be backed up by other services, such as 133 access to microfinance and streamlined and accessible government services for registering property and obtaining national identity cards. A targeted social insurance scheme should be considered for informal workers. Most workers in Pakistan, particularly in the informal sector, do not participate in a social insurance program. Pensions are largely restricted to a small number of government employees and full-time employees in the formal private sector. Given the fiscal constraints, piloting a program targeting poor and vulnerable workers who receive no social insurance would be an important step toward an integrated social protection system. And with high costs of health care and medicine, along with the detrimental impacts of health shocks on poor house- holds’ incomes, health coverage should be an important part of any comprehen- sive social insurance program. The poverty database developed in 2011 by the National Database and Registration Authority and BISP provides a readily avail- able information source for the design of a well-targeted program, which could be an extension of the current programs being piloted by BISP. Finally, although publicly financed noncontributory programs are the most common socia insur- ance programs in developing countries, subsidized social insurance programs (including government matching programs that double workers’ contributions) can help ensure the financial viability of the noncontributory ones (Box 7.2). Address high levels of child labor and facilitate youths’ transition from school to work An important first step toward reducing child labor is eliminating the worst forms of employment hazardous to children. The International Labour Organization The Self-Employed Women’s Association’s Integrated Insurance Box Scheme, India 7.2 The Self-Employed Women’s Association (SEWA) runs the largest comprehensive contributory social pro- tection scheme for informal workers in India. One-third of the premium is financed through interest paid on a grant provided by the Deutsche Gesellschaft für Technische Zusammenarbeit, one-third through direct contributions by women workers, and one-third through a subsidized package provided by the Life Insurance Corporation of India and the United India Insurance Company. SEWA members can choose whether to participate in the scheme, which covers insurance for health (including a maternity benefits component), life (death and disability), and assets (loss or damage to housing unit or work equipment). SEWA has also designed the payment of premiums to suit different income groups among the very poor. Source: Lund and Srinivas 2000. Enhancing labor market conditions for vulnerable groups defines the “worst forms of child labor� as any work that jeopardizes the health, safety, or morals of a child. Such work is determined to be hazardous depending on its specific nature, the demands on children in particular industries, and the 134 general working conditions (ILO 2007). Industries such as bricklaying, recycling, and carpet weaving are the most dangerous for children. Enforcing strict regulation of these industries, includ- ing fines, is one way to ensure underage children are not hired. Partnerships with private employers to help provide safety nets and education facilities for laid-off children is critical to ensure that households can gain alternative incomes without sending their children to work (Box 7.3). Public policies that promote part-time and informal methods of teaching, such as evening classes, can help older teenagers from poor households engaged in part-time work complete their secondary education. Conditional cash transfers (CCTs) might in the long run be the most effective way to reduce child labor, by creating incentives and safety nets that allow former and would-be child workers to go to school. Impact evaluations of programs that offer CCTs to combat child labor, such as those in Brazil, Colombia, Ecuador, and Mexico, have led to substantial decreases in child labor. In Cambodia— an example of a large reduction—the average child receiving the transfer was 10 percentage points less likely to work for pay (Filmer and Schady 2006). CCT programs must be supported by rigorous monitoring and evaluation to ensure that children stay in school and do not return to work. A pilot CCT program implemented by Pakistan Bait-ul-Maal over 2008–10 and evaluated by the World Bank demonstrated promising results, leading to a 12 percentage point increase in primary school enrollment for girls and a 9 percentage point increase for boys (Scott and others 2012). BISP, having learned from the pilot, is now rolling out a national CCT program (at the primary school level) as a Sialkot soccer ball industry: a model for public–private Box partnerships to reduce child labor 7.3 The city of Sialkot in Punjab has traditionally been one of the world’s major suppliers of hand-stitched soccer balls, and for many years child labor was endemic. Under an agreement between the International Labour Organization and the soccer ball manufacturers in 1997, children younger than 14 years were withdrawn from workplaces and, with their younger siblings, directed toward educational alternatives. Working with local communities, partners with the International Programme on the Elimination of Child Labour set up nonformal primary schools and vocational training centers for older children while mainstreaming others into formal school systems where possible. They also provided the children with health and social protection services and supported their families through income-generating activities. Source: IPEC 2009. top-off to the main unconditional cash transfer. Once the CCT program has been successfully implemented at primary schools nationally, current second- ary-school provincial interventions can be supplemented by additional incen- tives to poor children (as defined by the BISP database) for attending school. 135 Pakistan’s youth employment challenge is much broader than the unem- ployment challenge of higher educated youths. Indeed, even though higher educated youths are most visible to policy makers and tend to voice their concerns more than other youths, high unemployment rates of educated youths are due partly to (high) reservation wages that they maintain before accepting jobs. Rates of returns to education remain respectable in Pakistan and thus the educated youth tend to benefit from schooling in the long run. The policy interventions that will improve youth employment trends in Pakistan will be those that focus on women and on less educated youths who often have to work in the informal sector, simply because these groups overwhelmingly make up those who are either outside the labor force or occupied in low-productiv- ity and low-pay jobs. As a result, in addition to education policies covered in accompanying policy notes, the policies to increase women’s employment and to support informal sector workers emerge as key priorities to address Pakistan’s youth employment challenge. Notes 1. Sectors other than agriculture make up 58 percent of total employment. 2. Pardah —often translated as female s ­eclusion—is a contested term in Pakistan, as it can imply a variety of forms of veiling practiced in Pakistan— ranging from wearing a dupatta, chadar, or burqa to segregating women. 3. Firms that keep accounts, pay taxes, and use courts. 4. The National Assembly passed several important laws in the 1990s ban- ning or restricting child labor. The Employment of Children Act of 1991 prohibited employment of children younger than 14 years in factories and mines or in any hazardous environment (ILO 2009). The Bonded Labor Act of 1992 in theory abolished indentured servitude and the traditional peshgi system. Peshgi is a form of cash advances to employ- ees, which in its worst form is bonded labor, as it allows employers to retain workers if they cannot repay a cash advance. Despite being out- lawed, it is still common in the informal sector. Peshgi is practiced rou- tinely in the brick-kiln, bangle-making, and leather industries, in which labor violations and child labor are widespread and which attracts some of the country’s poorest communities (Collective for Social Sciences Research 2004). Enhancing labor market conditions for vulnerable groups References Cnobloch, Rodica, Rehan R. Jamil, and Afra R. Chowdhury. 2012. “How Are the 136 MDGs Likely to Evolve in the Short Term and Why Status Quo Is a Bad Out- come for Pakistan.� Unpublished manuscript. World Bank, Washington, DC. Cnobloch, Rodica, and Naveeda Salam. 2012. “Informal Employment: The Who, and How They Are Spending Their Time.� Unpublished manuscript. World Bank, Washington, DC. Collective for Social Science Research. 2004. “A Rapid Assessment of Bonded Labour in Hazardous Industries in Pakistan: Glass Bangle-Making, Tanneries and Construction.� Working Paper 21. Karachi. de Soto, Hernando. 2003. The Mystery of Capital. New York: Basic Books. Filmer, Deon, and Norbert Schady. 2006. “Getting Girls into School: Evidence from a Scholarship Program in Cambodia.� Policy Research Working Paper 3910, World Bank, Washington DC. IFC (International Finance Corporation). 2007. “Enterprise Survey: Pakistan Country Profile.� Washington DC. Ilahi, Nadeem, and Franqui Grimard. 2000. “Public Infrastructure and Private Costs: Water Supply and Time Allocation of Women in Rural Pakistan.� Economic Development and Cultural Change 49 (1): 45–75. ILO (International Labour Organization). 1998. Child Labour Survey in Pakistan. Islamabad. ———. 2006. “The End of Child Labour: Within Reach.� Global Report under the Follow-Up to the ILO Declaration on Fundamental Principles and Rights at Work. International Labour Conference, 95th Session, Geneva. ———. 2007. “Child Labor and Conditional Cash Transfer Programmes in Latin America.� Geneva. ———. 2009. “National Legislation and Policies against Child Labour in Pakistan.� Geneva. ———. 2012. Global Employment Trends for Youths. Geneva. IPEC (International Programme on the Elimination of Child Labour). 2009. “Football for Hope in Sialkot: A Football Lifeline for Children from Football Stitching Communities.� IPEC News Volume 1, March. Kabeer, Naila. 2008. “Gender, Labour Markets and Poverty: An Overview.� Poverty in Focus: Special Issue on Gender Equality (13): 3–5. Khandker, Shahidur R., Zaid Bakht, and Gayatri Koolwal. 2006. “The Poverty Impact of Rural Roads: Evidence from Bangladesh.� Policy Research Work- ing Paper 3875, World Bank, Washington, DC. Lloyd, Cynthia, Cem Mete, and Zeba Sattar. 2005. “The Effect of Gender Differ- ences in Primary School Access, Type, and Quality on the Decision to Enroll in Rural Pakistan.� Economic Development and Cultural Change 53 (3): 685–710. Lund, Francie Jane, and Smita Srinivas. 2000. Learning from Experience: A Gendered Approach to Social Protection for Workers in the Informal Economy. Geneva: International Labour Organization. MEASURE DHS. 2007. Pakistan Demographic and Health Survey 2006/07. ICF 137 International, Fairfax, VA. Nikitin, Denis. 2011. “Pakistan: Labor Market Trends.� Unpublished manuscript. World Bank, Washington, DC. Pakistan Bureau of Statistics. 2008. Labour Force Survey 2007/08. Government of Pakistan, Islamabad. Scott, Kinnon, Rodica Cnobloch, Xiaohui Hou, and Cem Mete. 2012. “Impact Evaluation Results of a Pilot CCT Program in Pakistan.� Unpublished manuscript. World Bank, Washington DC. United Nations Statistics Division. n.d. Millennium Development Goals data- base. Updated September 12, 2012, http://mdgs.un.org/unsd/mdg/ Default.aspx. U.S. Department of Labor. 2002. “2001 Findings on the Worst Forms of Child Labor—Pakistan.� Washington, DC. World Bank. 2010. “Pakistan Floods 2010: Preliminary Damage and Needs Assessment.� Washington, DC. ———. 2011a. “Pakistan: Poverty Assessment.� Washington, DC. ———. 2011b. South Asian Development Matters. Series. Washington, DC. ———. 2011c. “Social Safety Nets in Pakistan.� Washington, DC. ———. 2012a. “More and Better Jobs in South Asia.� Washington, DC. ———. 2012b. World Development Report 2012: Gender Equality and Development. Washington, DC. ———. 2013. World Development Report 2013: Jobs. Washington, DC. 8 John Newman Recovering Strong Positive Trends in Poverty and Opportunity Household survey data point to declining poverty over the past decade, despite some diffi- culties with data from three provinces. Indeed, Pakistan has turned economic growth into poverty reduction. In Punjab (the largest province, where the data appear more reliable and are thus analyzed in more detail), poverty has fallen considerably, from 33.5 percent in 2001/02 to 16.4 percent in 2007/08, after adjusting for higher food prices. This improvement was driven largely by increasing returns in the nonfarm sector, in both urban and rural areas. Over the period, the growth of per capita consumption of the bottom 40 percent of Punjab’s population exceeded GDP per capita growth. Subsequently, over 2007/08–2010/11, per capita real consumption growth in Punjab was stagnant, and the equality of opportunity for primary education completion rates seemed to improve but alongside a slowdown in the rate of improvement in indicators for water and sanita- tion and for primary enrollment. Nationwide, the rate of improvement for gas and elec- tricity deteriorated. Institutionally, there is a need for an independent expert committee to work with the Pakistan Bureau of Statistics to strengthen poverty monitoring. This note sheds some light on the last decade’s poverty trends in Pakistan. Based on the Household Income and Expenditure Surveys in 2001/02, 2004/05, 2007/08, and 2010/11, the broad trend appears to be one of substan- tial poverty reduction. However, given data concerns, it is premature to draw conclusions beyond this broad trend (see Box 8.1 below). Nevertheless, data from Punjab are mostly free of anomalous results, so this note provides more detailed analysis for that province. In addition to providing information on poverty headcount rates, the note introduces a possible measure of shared prosperity applied to the consumption growth of the bottom 40 percent of the population. This measure—an indicator the World Bank is testing for application worldwide—recognizes that consump- tion growth in the bottom 40 percent will matter most in building shared pros- perity. (It differs from the Gini coefficient, which pays attention to the equitable distribution of wealth but not growth at the bottom of the curve.) As changes in social indicators are also important, this note updates the Human Opportunity Index for 2010/11. It builds on earlier work that explored Time path of the relation between poverty and GDP per capita, Figure Pakistan and selected countries 8.1 Poverty rate at PPP $1.25 a day (percent) 100 China, 1980–2004 Vietnam, 1992–2007 139 75 Indonesia, 1983–2009 India, 1993/94–2009/10 50 Pakistan, 2001/02–2007/08 25 Sri Lanka, 1984–2006 0 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 GDP per capita (2005 international PPP $) Note: Each line represents the history of a country. For China, India, Indonesia, Sri Lanka, and Vietnam, the time paths are shown over a longer period; for other countries performance is shown only since 2000. Source: Author’s calculations. Poverty data are from PovcalNet, and GDP per capita data are from the World Development Indicators database. the trend in equality of human opportunity between 1998/99 and 2007/08. based poverty Historically, Pakistan has done better in reducing consumption-­ measures than in improving social indicators. Trends National Poverty reduction in Pakistan, as in most countries, is aligned closely with its growth in per capita income. While there are data concerns, the overall poverty rate appears to have fallen steeply over the past decade (Figure 8.1). China and Indonesia cut poverty substantially while their GDPs per capita increased sharply. Poverty reduction is strongest at the initial phase of growth, and as a country moves up the growth scale over time, it takes a greater increase in GDP per cap- Sheikhupura Sialkot Hyderabad Karachi Sukkur Quetta Peshawar Islamabad FaisalabadGujranwala Lahore Multan Rawalpindi ita to achieve a given percentage point reduction in poverty, with an even bigger impact in moving down at the lower levels of poverty. The line for Pakistan between 2001/02 and 2007/08 is quite steep, indicating a much stronger proportional reduction in poverty for the improvement in GDP per capita—an experience similar to those of other countries. Many other countries, either over their entire period or over a limited period, also experienced significant reductions in poverty relative to their gains in GDP per capita. Further progress will require not only sustained growth but also stronger growth. This sharp reduction in poverty over a short period also suggests that poverty in Pakistan is highly elastic to growth. Figure 8.2 compares the elasticity of Recovering Strong positive trends in poverty and opportunity poverty reduction with respect to growth across countries. From 2001/02 to 2007/08, Pakistan had high elasticity—well above the mean and toward the upper end of the distribution of developing countries. 140 However, while Pakistan is doing well converting growth into poverty reduction, it is struggling to sustain that growth. Since 1985, only once has Distribution of elasticities of poverty reduction with respect to Figure growth, selected countries 8.2 Fiji Nepal China Swaziland Mali Pakistan, 1987–2008 Guatemala Elasticity = –2.163 Niger Mozambique Namibia Malawi Cameroon Central African Rep. Azerbaijan Moldova India Panama Thailand Bangladesh Costa Rica Sri Lanka Nicaragua El Salvador Tunisia Guinea-Bissau Tanzania Brazil Jamaica Egypt, Arab Rep. Iran, Islamic Rep. Malaysia Algeria Turkey Belarus Rwanda Croatia Macedonia, FYR Serbia Slovenia Lithuania Czech Rep. Seychelles Mean = –0.716 Argentina Honduras Paraguay Belize Bolivia Yemen, Rep. –5 –4 –3 –2 –1 0 1 Elasticity Note: The elasticities are calculated over the entire range for which data are available (which varies for each country). Thus only one elasticity is presented per country. The figure only shows countries with at least six years between the first and last observation. The distribution is trimmed by dropping observations where the elasticities were less than –21 and greater than 20. Source: Author’s calculations. Poverty in the poverty elasticity calculations is the $1.25-a-day poverty headcount in international 2005 $ PPP from PovcalNet. Growth is calculated for GDP per capita (international 2005 $ PPP) from the World Development Indicators database. Pakistan’s GDP per capita in 2005 international $ PPP grown at least 3 percent a year for four consecutive years (2004–07)—and never has it for five consecutive. Unsurprisingly, countries that have been better at reducing poverty tend to be better at generating sustained growth (Figure 8.3). 141 This observation—that Pakistan is successful in reducing poverty when GDP grows but cannot sustain that growth—has two important policy implications. Number of growth episodes when GDP per capita grew at least Figure 3 percent a year for five consecutive years 8.3 Countries with average poverty reduction of more than 1 percent a year or end-period poverty of less than 5 percent 20 15 Growth episodes 10 5 0 Costa Rica Nepal Guatemala Fiji Namibia Senegal The Gambia Cameroon Pakistan Burkina Faso Uganda Timor-Leste Ghana Tajikistan Ethiopia Cambodia Bhutan Thailand Indonesia China Vietnam Lao PDR Countries with average poverty reduction of less than 1 percent a year 20 15 Growth episodes 10 5 0 Lesotho Kenya Guinea-Bissau Mauritania Philippines Zambia South A rca India Panama Nigeria Rwanda Tanzania Botswana Bangladesh Madagascar Note: The top panel shows the distribution of episodes of five consecutive years of growth above 3 percent for countries that started with poverty rates of more than 20 percent and that have averaged at least a 1 percentage point reduction a year or that had poverty rates below 5 percent at the end of the period. The bottom panel shows the distribution for countries that had initial poverty rates of more than 20 percent and averaged less than a 1 percentage point reduction a year. Source: Author’s calculations based on World Development Indicators database. Recovering Strong positive trends in poverty and opportunity First, with more growth interruptions, an adequate social protection system becomes more important. The poor are vulnerable to shocks—be they of natural disasters, health, or macro policy. An adequate system would ensure 142 that when shocks hit, the poor and vulnerable can still maintain the invest- ments they need to increase their incomes and their children’s welfare. With the advances in the Benazir Income Support Programme, Pakistan has made solid progress in developing such a system in recent years. The second policy implication is that a renewed effort to address the problems that work against sustained growth would be well justified for faster poverty reduction. This effort should lead to policy priorities for poverty reduction (such as more focus on the macro environment) different from those in countries better able to sustain growth but unable to convert that growth into rapid poverty reduction. In Punjab, overall Given other provinces’ data concerns, the focus is on Punjab, the only province free of anomalous results (Box  8.1). As the largest province, Punjab can provide some insight into the country as a whole. But even here it is necessary to adjust for increasing relative prices of food on the poverty line. The official poverty line uses the general consumer price index to adjust the consumption base, but starting in 2006/07 the index for food and other essen- tial items began to grow faster than the general consumer price index. Table 8.1 Box Concerns over poverty data in three provinces 8.1 Pakistan’s household survey data has the potential to provide useful evidence for guiding policy decisions on reducing poverty. Yet examining the poverty data from four recent Household Income and Expenditure Surveys at three-year intervals (2001/02, 2004/05, 2007/08, and 2010/11) reveals some data concerns that need to be resolved before a more detailed assessment can be made. These concerns include large reductions in rural poverty in Sindh combined with a decrease in food expenditure as a proportion of the total; large swings in poverty in Balochistan and Khyber Pakhtunkhwa (they are conflict-prone, so systematic data capture from all regions needs to be confirmed); the exclu- sion of many durable consumer goods from the estimate of total expenditure (which can understate improvements in poverty reduction as income grows, as well as the impact of an economic downturn); and the need to update the poverty line to account for changes in relative food and nonfood prices (as when food prices increase, higher expenditure will be required to generate the same minimum calorific intake). Until these problems are resolved, it will remain difficult to drill down to more detail using the national poverty estimates—hence the spotlight on Punjab, the only province free of anomalous results. Clearing up the concerns on national poverty data requires a review of the data collection practices in each province for each survey year. This can be done only with staff from the Pakistan Bureau of Statistics, who would have detailed information on field practices. illustrates how the poverty line, in both urban and rural areas, would change once such an adjustment is made. The overall trend remains strongly toward poverty reduction, from 33.5  percent to 14.6  percent rather than to 11.8  percent. And rural poverty remains much higher than urban poverty, but 143 some rates of poverty reduction would be somewhat slower, particularly in the most recent period. This adjustment is only illustrative, however. Ideally, the poverty line should be recalculated, using either the 2007/08 or 2010/11 Household Income and Expenditure Survey, at which time the government should also discuss the choice of price index to adjust the poverty line for inflation. It would be useful to have an expert committee pursue these proposed changes and resolving some of the anomalous results, work with the Pakistan Bureau of Statistics, and recommend ways that more reliable, regular, and timely poverty data can be produced and made available to the public. In Punjab, by income group Overall poverty reduction is important, but it is equally important to disaggregate the data to see the impacts at different incomes. Figure 8.4 presents the growth in consumption at each income group for the province as a whole and for urban and rural populations for the three growth periods. During the first growth period (2001/02–2004/05), poverty reduction was spurred by good consumption growth among all income groups, though the wealthy tended to benefit more. In the second (2004/05–2007/08), overall growth in consumption remained strong but was most pronounced in the lowest income groups. In the third (2007/08–2010/11), growth was far lower among all percentiles, as positive growth in urban areas was outweighed by negative growth across all percentiles in rural areas, hence the reduced rate of poverty reduction. These growth inci- dence curves should be viewed with an appreciation of population movement to Illustrative example of the effect of adjustment of poverty line to the increase in food prices relative to the general consumer Table price index, Punjab 8.1 Adjustment of original poverty line using Adjustment of original poverty line with adjustment general consumer price index for increased relative food prices in poverty line Nominal All Nominal All poverty line Punjab Urban Rural poverty line Punjab Urban Rural 2001/02 723.4 33.5 24.4 37.1 723.4 33.5 24.4 37.1 2004/05 878.6 24.3 17.3 27.5 888.2 25.0 18.2 28.1 2007/08 1,141.5 14.1 9.0 16.5 1,179.3 16.4 10.8 19.0 2010/11 1,671.7 11.8 5.3 14.9 1,771.4 14.6 6.5 18.4 Source: Author’s calculations based on Household Income and Expenditure Surveys 2001/02, 2004/05, 2007/08, and 2010/11. Recovering Strong positive trends in poverty and opportunity Figure Growth incidence curves (for three-year periods), Punjab 8.4 Growth incidence Growth at median Growth in mean Mean growth rate 99 percent con dence bounds 144 2001/02–2004/05 Total Urban Rural 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 0 25 50 75 100 0 25 50 75 100 0 25 50 75 100 Expenditure percentile Expenditure percentile Expenditure percentile 2004/05–2007/08 Total Urban Rural 30 30 30 20 20 20 10 10 10 0 0 0 –10 –10 –10 0 25 50 75 100 0 25 50 75 100 0 25 50 75 100 Expenditure percentile Expenditure percentile Expenditure percentile 2007/08–2010/11 Total Urban Rural 20 20 20 10 10 10 0 0 0 –10 –10 –10 –20 –20 –20 0 25 50 75 100 0 25 50 75 100 0 25 50 75 100 Expenditure percentile Expenditure percentile Expenditure percentile Note: The growth rates are over the entire three-year period (not annualized). The time periods are the same so that the growth can be compared directly. The scales of the various growth incidence curves are different to show more detail. Thus, in comparing the figures, please note the scale of the y axis. Source: Author’s calculations based on Household Income and Expenditure Surveys 2001/02, 2004/05, 2007/08, and 2010/11. urban areas. Indeed, although the share of the rural poor stayed roughly con- stant at 85 percent, overall poverty rates would have been higher without the migration from higher poverty rural areas to lower poverty urban areas. The bottom 40  percent can fully benefit from growth. Another indicator of whether the lower income population is benefiting fully from growth and poverty reduction is to look at the growth rate of the lower 40  percent of the population—a proxy measure of shared prosperity. Table 8.2 presents information on the growth of consumption of that 40 percent relative to the entire distribution and to the growth rate of real GDP per capita. During the first two periods, growth of the bottom 40 percent exceeded the growth of GDP per capita. Only in the last period did growth of that 40 percent fall below the growth of per capita GDP—yet it still stayed above that of the entire distribution. Thus for Punjab at least, it would appear that the bottom 40 percent can fully 145 benefit from growth. The end of growth tends to signify an end to poverty reduction. The problem seems to be more the history of overall inconsistent growth with short periods of positive (but not terribly high) growth not being sustained over long periods (see Table 8.1 and Figure 8.2). One final observation: the bottom 40 percent tends to be more rural than the top 60 percent. However, this does not mean that the bottom 40 percent are all in rural areas and the top 60 percent are all in urban areas (Table 8.3). Human Opportunity Index An opportunity can be defined as a good or service important enough to be made available to all, regardless of background. In most societies, basic Mean real consumption and growth in mean real consumption per adult equivalent of bottom 40 percent of population and total Table population 8.2 National Punjab Growth of GDP per capita Growth of mean real per capita consumption expenditure (constant local currency units; percent) All households (percent) Bottom 40 percent (percent) 2001/02–2004/05 3.2 5.0 3.9 2004/05–2007/08 4.5 4.8 5.2 2007/08–2010/11 1.1 –0.6 0.1 Source: GDP per capita from the World Development Indicators database; growth rate of mean real per capita consumption expenditure—author’s calculations based on Household Income and Expenditure Surveys 2001/02, 2004/05, 2007/08, and 2010/11. Distribution of population in bottom 40 percent and top Table 60 percent of real monthly consumption per capita 8.3 (percent) Bottom 40 percent Top 60 percent Urban Rural Urban Rural 2001/02 22 78 34 66 2004/05 23 77 37 63 2007/08 23 77 37 63 2010/11 20 80 40 60 Source: Author’s calculations based on Household Income and Expenditure Surveys 2001/02, 2004/05, 2007/08, and 2010/11. Recovering Strong positive trends in poverty and opportunity education, health, and infrastructure services are considered opportunities. An opportunity is said to be distributed according to a principle of equality of opportunity if circumstances exogenous to the individual—such as birth 146 place, gender, ethnicity, income, and education of parents—have no bearing on how the opportunity is distributed in the population. The Human Oppor- tunity Index (HOI) combines the overall coverage rate of the opportunity with a “penalty� for the share of access to opportunities distributed unequally.1 Opportunity is picking up. Figures 8.5 and 8.6 measure the education and sanitation opportunities for children in Pakistan. Opportunity is growing in both urban and rural areas, a very positive sign. Urban children have more absolute opportunity than rural children, but the rate of growth in rural areas is growing faster. The slowing rate of opportunity in sanitation is worrying, how- ever, as it can have implications for stunting. Figure Primary completion rates of children ages 15–19 8.5 1998/99 Human Opportunity Index National 2007/08 2010/11 1998/99 Urban 2007/08 2010/11 1998/99 Rural 2007/08 2010/11 0 25 50 75 100 Coverage rate (percent) 1998/99–2007/08 2007/08–2010/11 National Urban Rural 0 1 2 3 4 5 0 1 2 3 4 5 Annualized growth rate of Annualized growth rate of Human Opportunity Index (percent) Human Opportunity Index (percent) Note: The top panel shows the coverage rate for each year (shown by the bar) and the value of the inequality-adjusted coverage rate, the Human Opportunity Index (shown by the line in the bar). The closer the line to the end of the bar, the smaller the penalty. The two bottom graphs show the growth rate of the Human Opportunity Index, with the line in the bar showing the division between growth in coverage and growth in equality. Source: Pakistan Social and Living Standards Management Surveys 1998/99, 2007/08, and 2010/11. Figure Improved sanitation for children ages 0 –16 8.6 1998/99 Human Opportunity Index National 2007/08 147 2010/11 1998/99 Urban 2007/08 2010/11 1998/99 Rural 2007/08 2010/11 0 25 50 75 100 Coverage rate (percent) 1998/99–2007/08 2007/08–2010/11 National Urban Rural 0 2 4 6 8 10 0 2 4 6 8 10 Annualized growth rate of Annualized growth rate of Human Opportunity Index (percent) Human Opportunity Index (percent) Note: The top panel shows the coverage rate for each year (shown by the bar) and the value of the inequality-adjusted coverage rate, the Human Opportunity Index (shown by the line in the bar). The closer the line to the end of the bar, the smaller the penalty. The two bottom graphs show the growth rate of the Human Opportunity Index, with the line in the bar showing the division between growth in coverage and growth in equality. Source: Pakistan Social and Living Standards Management Surveys 1998/99, 2007/08, and 2010/11. There is a mixed picture of opportunity by province. Figures 8.7 and 8.8 pro- vide the same information but for the other provinces. It is important to look at both the level of coverage and HOI as well as their growth to judge whether there may be cause for concern. In primary completion, there appears to be a problem in Sindh. The level is still low, and the rate of growth is low and showed little improvement in 1998/99–2007/08 and 2007/08–2010/11. By contrast, while Balochistan and Khyber Pakhtunkhwa were behind Punjab and Sindh, the pace of progress in the two provinces has picked up. In sanitation, there are striking differences in the pace of improvement, with good recent progress in Baloch- istan and Sindh but decreases in HOI in Khyber Pakhtunkhwa and Punjab. Improvement is slowing some. Table 8.4 provides a summary of all the indicators that could be readily analyzed using the Social and Living Stan- dards Measurement Survey. In general, the annual pace of improvement in the indicators has slowed over the latest period relative to 1998/99–2007/08, Recovering Strong positive trends in poverty and opportunity Figure Primary completion rates of children ages 15–19, by province 8.7 Punjab 1998/99 2007/08 Human Opportunity Index 148 2010/11 1998/99 Sindh 2007/08 2010/11 Balochistan Pakhtunkhwa 1998/99 Khyber 2007/08 2010/11 1998/99 2007/08 2010/11 0 25 50 75 100 Coverage rate (percent) 1998/99–2007/08 2007/08–2010/11 Punjab Sindh Khyber Pakhtunkhwa Balochistan 0 2 4 6 0 2 4 6 Annualized growth rate of Annualized growth rate of Human Opportunity Index (percent) Human Opportunity Index (percent) Note: The top panel shows the coverage rate for each year (shown by the bar) and the value of the inequality-adjusted coverage rate, the Human Opportunity Index (shown by the line in the bar). The closer the line to the end of the bar, the smaller the penalty. The two bottom graphs show the growth rate of the Human Opportunity Index, with the line in the bar showing the division between growth in coverage and growth in equality. Source: Pakistan Social and Living Standards Management Surveys 1998/99, 2007/08, and 2010/11. apart from the primary completion rate. The slippage is seldom much, but in sanitation, gas, electricity, and primary enrollment the reduction ­ oteworthy—especially as Pakistan has not reached full coverage in is n these areas. Rather than repeating similar figures for all the indicators, the interested reader can go directly to a website and explore interactive “dashboards.�2 Conclusions Pakistan has a rich set of data that suggests a decline in poverty over the past decade. The data contain some concerns and need to be considered with caution, but the data for Punjab are free of some of the other provinces’ anom- alous results. The Household Income and Expenditure Survey data for Punjab tell us that between 2001/02 and 2007/08 poverty declined rapidly, especially for the rate of growth of real GDP per capita. Much of this decline was driven Figure Improved sanitation for children ages 0 –16, by province 8.8 1998/99 Punjab 2007/08 2010/11 149 1998/99 Sindh 2007/08 2010/11 1998/99 Human Opportunity Index Balochistan Pakhtunkhwa Khyber 2007/08 2010/11 1998/99 2007/08 2010/11 0 25 50 75 100 Coverage rate (percent) 1998/99–2007/08 2007/08–2010/11 Punjab Sindh Khyber Pakhtunkhwa Balochistan –5 0 5 10 15 –5 0 5 10 15 Annualized growth rate of Annualized growth rate of Human Opportunity Index (percent) Human Opportunity Index (percent) Note: The top panel shows the coverage rate for each year (shown by the bar) and the value of the inequality-adjusted coverage rate, the Human Opportunity Index (shown by the line in the bar). The closer the line to the end of the bar, the smaller the penalty. The two bottom graphs show the growth rate of the Human Opportunity Index, with the line in the bar showing the division between growth in coverage and growth in equality. Source: Pakistan Social and Living Standards Management Surveys 1998/99, 2007/08, and 2010/11. by increases in returns in the nonfarm sector in urban and rural areas, which reflects an increase in the relative price of labor. The rate of poverty reduction, however, appears to have slowed since, and without population movement from rural to urban areas, poverty would have been worse. Over 2001/02–2007/08, growth of mean per capita consumption of the lowest 40 percent of the population increased faster than real per capita GDP. When growth in per capita GDP declined, growth rates in per capita mean consumption expenditure and poverty reduction also declined. International comparisons suggest that Pakistan has been a good performer in turning growth into poverty reduction. Pakistan has demonstrated that it can reduce poverty even at the relatively low rates of growth of 3.2–4.5 per- cent—but not at growth of GDP per capita of 1 percent. Countries that are more successful in reducing poverty tend to be better at generating sustained growth. The issue for Pakistan will thus be sustaining growth. It has had only Recovering Strong positive trends in poverty and opportunity Table Trends and average annual change in Human Opportunity Index 8.4 Average annual change in Human Opportunity Index (percent) 150 1998/99 2007/08 2010/11 1998/99–2007/08 2007/08–2010/11 Education Primary enrollment 47.8 66.6 68.5 3.7 1.0 Secondary enrollment 41.8 55.0 59.0 3.1 2.3 Primary completion 45.6 55.0 59.7 2.1 2.7 Secondary completion 18.6 24.5 26.6 3.1 2.6 Health Percentage of births assisted by traditional and formal birth attendants 71.1 80.2 80.6 1.3 0.2 Percentage of births assisted by formal birth attendants 12.7 30.4 37.4 9.7 6.9 Institutional birth 10.8 27.7 34.9 10.5 7.7 Any postnatal care 6.0 15.7 21.1 10.7 9.8 Ever immunized 76.6 94.1 95.3 2.3 0.4 Full immunization (card and self-reported) 43.2 68.9 71.0 5.2 1.0 Full immunization (card) 25.4 45.1 25.1 6.4 –19.6 Free of diarrhea in last 30 days 86.8 88.6 87.6 0.2 –0.4 Infrastructure Gas 7.0 14.2 15.9 7.9 3.8 Electricity 61.7 84.7 87.6 3.5 1.1 Sanitation 27.3 46.0 48.0 5.8 1.5 Water 71.5 84.6 82.5 1.9 –0.9 Source: Author’s calculations based on Household Income and Expenditure Surveys 1998/99, 2007/08, and 2010/11. one period of four—and unlike many other developing countries, never five— consecutive years of growth of GDP per capita in international $ PPP above 3 percent. Because it is subject to stop-go growth and to many natural disasters, it has to ensure a strong safety net program as part of an overall poverty reduc- tion strategy—like the Benazir Income Support Programme—though this is no substitute for sustained growth. Results from the Household Income and Expenditure Surveys data on trends in social indicators can be used immediately to guide policy, though more work is required to increase confidence in the poverty estimates. The analysis for this note could be carried out readily outside Punjab, as soon as some poten- tial problems are resolved. Until then, some caution should be exercised in generating national figures on poverty. The problems would be best addressed by an independent task force of experts consisting of representatives from the Pakistan Bureau of Statistics and from the community of poverty researchers. The HOI can be useful in identifying whether there has been progress in coverage and equality of service provision. Analysis of level and growth together provides an indication of where increased efforts are needed. The pace of improvement in sanitation has declined. This is likely to have important implications for the 151 pace of improvement in nutrition outcomes, because there is increasing evidence that sanitation plays a very important role in affecting stunting—particularly in South Asia. There has been some good news with improvements in gender equal- ity—and the HOI analysis provides a useful metric to gauge whether inequal- ity in social indicators is narrowing. Finally, the last three years have seen some sizable differences in the improving social indicators. Sindh has been lagging in its primary completion rates, and Khyber Pakhtunkhwa has been lagging in coverage of improved sanitation. Policy makers and interested readers can explore the details of performance of the social indicators through interactive dashboards that have been prepared to complement this note (see References). Notes The author would like to thank Olivier Dupriez and Tefera Bekele for their help in preparing the aggregate total consumption expenditure estimates, Hernan Winkler and Gabriela Inchauste for useful conversations on the micro decompo- sitions, and Minh Cong Nguyen for his help in preparing this note. 1. Recently developed at the World Bank, the HOI has been estimated for more than 20 countries in Latin America and Africa. 2. Links to the dashboards can be found in the References section of this note. References Government of Pakistan. Various years. Household Income and Expenditure Survey. Islamabad. ———. Various years. Pakistan Social and Living Standards Management Survey. Islamabad. World Bank. n.d. PovcalNet. Washington, DC. http://iresearch.worldbank.org/ PovcalNet/index.htm. ———. n.d. Contribution National. Washington, DC. http://data- viz.worldbank.org/t/SASEP/views/PakistanContributionNational/ DshContributionNational?:embed=y. ———. n.d. Contribution Provinces. Washington, DC. http://data- viz.worldbank.org/t/SASEP/views/PakistanContributionProvinces/ DshContributionProvinces?:embed=y. ———. n.d. Dashboard HOI National. Washington, DC. http:// dataviz.worldbank.org/t/SASEP/views/PakistanHOINational/ DshHOINational?:embed=y; Recovering Strong positive trends in poverty and opportunity ———. n.d. Dashboard HOI Provinces. Washington, DC. http:// dataviz.worldbank.org/t/SASEP/views/PakistanHOIProvince/ DshHOIProvince?:embed=y; 152 ———. n.d. World Development Indicators. Washington, DC. http://data.world- bank.org/data-catalog/world-development-indicators. Halil Dundar and Huma Ali Waheed Expanding Access to 9 Quality Education Although Pakistan has made some progress in improving access to education over the past decade, it still faces major challenges in providing access to quality education at all levels. It has the world’s second highest out-of-school population (7 million)—two-thirds of them girls (though enrollment rates vary appreciably between and within provinces). National surveys of student learning suggest that achievement is also very low—a sizable share of school leavers do not achieve even minimum mastery of mathematics, reading, and language, as defined by the government. The main contributing factors include poor teacher quality and accountability, inadequate and inefficient funding, and weak management and governance. Since the 18th Amendment was passed in 2010, the management and financing of education has been decentralized to the provinces, but national standards need to be set and their achievements monitored to address disparities in access to quality education between provinces. The federal government should play this role and coordinate and facilitate the provision of “education for all.� Pakistan ranks 113 of 120 countries in the United Nations Educational, Scientific and Cultural Organization’s Education for All Education Devel- opment Index. 1 It has the world’s second highest out-of-school popula- tion (7 million), two-thirds of them girls (UNESCO 2012). It has generally performed worse than other countries in South Asia and other developing countries (at its level of per capita income; Figure 9.1). Based on current trends, the United Nations Development Programme reports that the country is unlikely to meet the Millennium Development Goal of universal primary education by 2015. Background Education service delivery up to grade 12 has primarily been the concern of provincial and area governments. They are responsible for policy formulation, sector financing, and implementation through their education departments. The 18th Amendment, passed in 2010, formalized this responsibility. When the amendment was approved, Article 25-A—which guarantees the right to free and compulsory education for all children ages 5–16 years—was also added. Article 25-A requires further legislation by the provincial assemblies to enforce free and compulsory education; the process is under way. Expanding access to quality education Figure Primary net enrollment rate, selected developing countries, 2009 9.1 100 90 154 80 Percent 70 60 50 Pakistan Bangladesh India Indonesia Sri Lanka Mozambique Philippines Thailand Source: UNESCO Institute for Statistics n.d. As in other South Asian countries, Pakistan’s private sector is emerging as an alternative to public schools, even for poor households. Today, nearly a third of primary and secondary students in Pakistan attend private schools: 31 percent of boys and 33 percent of girls ages 6–10 years and 26 percent of boys and 30 percent of girls ages 11–15 years (Figure 9.2). Over 2004/2005– 2010/11, the overall share of children ages 6 –10 and 11–15 enrolled in private schools increased 7 and 6 percentage points, respectively. In net terms, virtually all the gain in school participation in the country during the period, especially at the primary level, was a result of increased private school participation. Private school enrollment by gender, selected South Asian Figure countries 9.2 Male Female 60 40 Islamabad FaisalabadGujranwala Lahore Sheikhupura Sialkot Hyderabad Karachi Sukkur Multan Rawalpindi Quetta Peshawar Percent 20 0 Afghanistan Bangladesh Bhutan India Nepal Pakistan Afghanistan Bangladesh Bhutan India Nepal Pakistan 2007/08 2010 2007 2009/10 2009/10 2010/11 2007/08 2010 2007 2009/10 2009/10 2010/11 6–10 years old 11–15 years old Note: Data for Afghanistan are from 2007/08; Bangladesh, 2010; Bhutan, 2007; India, 2009/10; Nepal, 2009/10; and Pakistan 2010/11. Source: World Bank 2013. This increased enrollment was triggered by low public financing and poor public service delivery. In responding to the widespread demand for greater access to and better education, a sizeable and rapidly expanding low-cost private schooling system now serves as a viable alternative to the government 155 school system for even low-income and rural households. The private sector holds real potential for increasing access to quality education. To further increase access, the governments of all four provinces have set up “education foundations,� which channel public financing for low-cost private provision. Evidence from Sindh and Punjab confirms that the growth of such a dynamic school system is reaching more and more low-in- come and rural households. It also suggests that student achievement in private schools tends to be much higher than in government schools—and that the higher achievement is obtained far more cost effectively than in government schools, possibly arising from a combination of private schools’ lower labor costs and market stimulation. To address the major issues in education, all four provinces have embarked on sector reform programs that aim to increase equitable access to quality education. Education reform initiatives address issues ranging from merit-based teacher recruitment and mobilization and financing of school management committees to the provision of basic facilities and free textbooks to schools. Punjab initiated the reform process nearly a decade ago and is the most advanced in implementing its reforms (Box 9.1). Balochistan, which lags behind the most, is still developing its sector reform program. Main Issues Despite recent achievements in improving access to education, Pakistan’s education sector performs poorly compared with other South Asian countries. This undermines its efforts in economic growth and poverty alleviation. Low enrollment and major disparities in access Although Pakistan has made some progress in expanding enrollment since 2000, access to education remains low at all levels. Major shortfalls persist in school participation. 2 In 2010, the estimated 7 million children still not in primary school accounted for about 10 percent of all out-of-school children worldwide. Primary, middle, and matric net enrollment rates in 2010/11 were 66 percent, 35 percent, and 23 percent, 3 or close to those of low-income countries in Sub-Saharan Africa (Figure 9.3). There are also large inter- and Expanding access to quality education Box Punjab Education Sector Reform Program 9.1 In 2003, Punjab launched its Education Sector Reform Program to improve access, quality, and gover- nance in education. Key reforms include: 156 • External monitoring system. This relies on independent collection of data on school performance (such as teacher presence) and ensures that policy makers have reliable data to make informed, transparent, and needs-based decisions. • New teacher recruitment and placement policy. This was introduced to hire new teachers on merit, with preference to local candidates and women. All teacher recruitment is now transparent and has had a major impact on ensuring that teachers with the right qualifications are hired. This has, in turn, helped reduce chronic teacher absenteeism. The government is working on a teacher certification and licensing mechanism to ensure that teachers gain the required competencies. • New textbook policy. The development, printing, and distribution of textbooks are now managed through an open competitive process. • Public-private partnership program. This was established to provide public funding to low-cost pri- vate schools. Under this program, the Punjab Education Foundation was restructured as an auton- omous but publicly funded institution to support low-cost private schools. In 2012, more than a million students benefited from the program. • Measurement of student learning outcomes. Under this reform, an independent Punjab Examination Commission was established to carry out universal examinations for grades 5 and 8. The program also has tight links with content and quality of teaching and is a strong tool for holding teachers accountable for performance. According to Pakistan Social and Living Standards Measurement Surveys, the primary net enrollment rate increased from 45 percent in 2001/02 to 62 percent in 2006/07 (for girls, from 43 percent to 59 percent; for rural girls, from 38 percent to 55 percent). These gains were helped by improved sector governance, including the establishment of independent bodies and stronger monitoring systems. intraprovincial variations—for instance, district primary net enrollment rates ranged from 14 percent to 90 percent. Pakistan’s completion rate for primary education is among the lowest in the world. Less than half the country’s population has completed at least primary education, making Pakistan unlikely to meet the education Millennium Development Goal by 2015. Education participation is inequitable, even at the primary level. Girls, disadvantaged children from poor families, rural children, and children from some social groups have very low enrollment rates. The child’s age and the household’s socioeconomic status and location (urban or rural) appear to matter greatly for participation. The chances of participation at primary and middle levels increase with age, suggesting late entry into school. Children from poor households appear to suffer a large participation disad- vantage at all levels: only 43 percent of children ages 6–10 years belonging Figure Pakistan net enrollment rates, 2010/11 9.3 Federal Punjab Sindh Khyber Pakhtunkhwa Balochistan 75 157 50 Percent 25 0 Primary Middle Matric Source: Pakistan Bureau of Statistics 2011. to the poorest wealth quintiles are enrolled. The participation disadvantage rises sharply for rural households at middle and high education levels: only 36 percent of rural children ages 11–15 years are enrolled in grades 6–10, compared with 56 percent of urban children. Although some provinces, notably Punjab, have made progress in reducing the gender gap, female participation in primary and secondary education remains low. At 61 per- cent, female participation at the primary level lags 10 percentage points behind male participation. Poor learning outcomes International comparisons of student achievement are unavailable because Pakistan has not taken part in any international assessments. However, national surveys of student learning outcomes suggest low achievement. A large proportion of school leavers do not achieve minimum mastery of mathematics, reading, and language (as defined by the national govern- ment). The ASER4 2011 assessment found that, although arithmetic compe- tence was somewhat better in rural Pakistan than in rural India, it was very low in absolute terms (ASER-Pakistan 2011). For instance, only 37 percent of grade 5 students in rural Pakistan could divide a three-digit number by a single-digit number. By grade 8, only 72 percent could perform simple division. Unlike in rural India, however, recognition of two-digit numbers in rural Pakistan was widespread by grade 3. A survey of children in rural villages in selected Punjab districts found that, in 2003, students in grade 3 scored 30 percent on English tests, 29 percent on Urdu tests, and 38 percent on math tests. Four years later, students in grade 3 in the same villages scored similarly or slightly lower, at 31 percent, Expanding access to quality education 27 percent, and 34 percent in the same tests (Andrabi and others 2007). Data from universal testing exercises in Punjab in 2010 and 2011 suggest that the child’s gender (a general disadvantage for boys), urban or rural 158 status (a disadvantage for rural children), and district affect the mean test scores for government school students in grades 5 and 8. 5 The above low enrollment and poor learning outcomes have three main con- tributing factors, now discussed. Poor teacher quality and accountability Teachers are consistently found to be the most important factor in student learning (Hanushek and Rivkin 2010; Rivkin, Hanushek, and Kain 2005; Nye, Konstantopoulos, and Hedges 2004; Rockoff 2004; Park and Hannum 2001; Sanders and Rivers 1996). While the quality of education is affected by many factors (teachers, facilities, textbooks, and so forth), improving teach- ing is probably the most effective way to raise school quality. In Pakistan, both teacher effectiveness and accountability are weak. Key issues include: • Poor quality of teachers . Teachers across the country have low content knowledge and weak pedagogical skills. A recent survey showed that only 36 percent of primary school teachers in Punjab could explain two-digit addition, suggesting that they are not competent to teach the curriculum (Word Bank 2013). This is generally attributable to the country’s weak recruitment policies and practices. Two of the larger provinces—Punjab and Sindh—have introduced merit-based teacher recruitment, which could dramatically improve teacher quality at entry (Box 9.2). Policies and practices in teacher deployment are also a serious concern. • Weak teacher accountability . Measured by high absence rates and minimal effort, this is the main cause of poor learning outcomes even when teachers are sufficiently qualified. A recent survey of primary schools in rural Punjab found that 11 percent of the teachers were absent on any given day. Most nonattendance was unexplained, and illness accounted for most explained absences (World Bank 2013). In most parts Box Merit-based teacher recruitment in Punjab and Sindh 9.2 The recruitment of new teachers in Punjab follows objective, merit-based criteria based largely on academic and professional qualifications of candidates. However, evidence from the province suggests that credentials do not seem to be associated with improved teacher competency and performance. In addition, the authen- ticity of professional credentials has been brought into question by preservice diplomas and degrees through nonaccredited programs. Sindh has introduced test-based teacher recruitment, which potentially is a far better way of judging whether a candidate has the skills to be a good teacher. of the country, policy dictates that teachers can be dismissed for mis- conduct, child abuse, absenteeism, and poor performance, while teacher absenteeism can theoretically result in penalties such as salary deduc- tions—but there is no evidence that these strictures are enforced. 159 • Few incentives for teacher performance . Government school teacher salaries are determined principally by teacher grade and tenure, along with allow- ances and benefits that relate to the characteristics of the location and school where the teacher works. Importantly, government school teacher salaries are not linked to the levels of teacher absence, teacher subject knowledge, and student achievement. • Discrepancies in teacher placement . The placement of teachers shows wide vari- ations: some schools have fewer than five students per teacher, and others more than 100. Some schools are thus understaffed, and others have more teachers than they need. Political interference in teacher placement and transfers is the main contributor to the imbalance. Particularly in rural schools, teachers may be appointed to vacant positions, but they are known to use political pressure to transfer their posts within a few months, leav- ing remote schools understaffed or sometimes without a teacher altogether. Government policies dictating student–teacher ratios and teaching loads exist, but enforcement is weak and made harder by external influences. Inadequate and inefficient financing of public education Although Pakistan is increasing its attention to education, financing has not followed suit. The country has one of the lowest rates of public expendi- ture on education as a share of GDP (1.9  percent; Figure 9.4). Adequate financing is a major challenge for the provinces and the country as a whole. More than 90 percent of the education sector’s recurrent expenditure is related to remuneration. This leaves a very limited budget for spending on critical nonsalary inputs, such as teaching and learning materials, repair and maintenance of school infrastructure, and teacher professional develop- ment. The quality of public education suffers as a result. The education funding mechanism does not induce behavioral changes through incentive structures and accountability mechanisms. These could ensure that school inputs are used more efficiently and lead to gains in learn- ing. Establishing a link between financing and outcomes is essential for this. Weak management and governance The country’s capacity to plan, manage, and monitor the education sector is weak. The management and financing of primary and secondary education Expanding access to quality education Public expenditure on education, Pakistan, its provinces, and Figure selected countries 9.4 Punjab Sindh 160 Balochistan Khyber Pakhtunkhwa Pakistan Bangladesh Philippines Turkey India Indonesia Egypt, Arab Rep. Argentina Thailand Malaysia Mexico Brazil Iran, Islamic Rep. South Africa 0 1 2 3 4 5 6 Percent of GDP Note: Provincial expenditures are shown as a percentage of provincial GDPs. These numbers are not strictly comparable because of the different definitions, and it is imprudent to compare provinces with countries. These comparisons should therefore be considered as indicative only. Source: World Bank data. Ministry of Finance data for Pakistan and provinces. are the responsibility of provincial governments, which are below par in their capacity to gather and report on key performance indicators on time to effectively manage and monitor implementation of education reforms. The key issues in policy making, management, and monitoring capacity include: • Inadequate institutional capacity for primary and secondary education delivery, hampering the implementation of a consistent and effective education policy. Local capacity to deliver education services is especially minimal at district and village levels. • Weak monitoring and evaluation systems at all levels of government to guide reforms . Data, not always easily accessible, are rarely used either by the system to improve education decision making or by parents to demand school accountability. Student assessment systems are inadequate for providing systematic and reliable information on student learning outcomes. • Deficiency in coordination and support at the federal level for achieving national education goals . Provincial governments are responsible for delivering pri- mary and secondary education, including policy making, management, Sheikhupura Sialkot Hyderabad Karachi Sukkur Quetta Peshawar Islamabad FaisalabadGujranwala Lahore Multan Rawalpindi financing, planning, and monitoring. However, in principle, the federal government has a pivotal role to play in promoting national cohesion of the education system, whether through setting universal teacher stan- dards or in curricula development. Yet the federal Ministry of Education and Training, established in 2012 following the 18th Amendment, is not 161 involved in either activity. Policy Options Making learning outcomes an explicit goal of education policy. Although Pakistan has made some progress in improving access to education, it faces major challenges in providing access to quality education at all levels. How can it address the twin challenges of access and quality, given competing demands on a very limited national budget? Whereas continuing efforts are needed to increase participation and make education more inclusive, policy should highlight the importance of improved learning outcomes for all. In fact, learning outcomes need to be the central goal of any education pol- icy—an “umbrella priority� under which the other priorities fall. Five are identified below. Improve teacher effectiveness and accountability Developing and implementing policies that focus on teacher performance and management should be the cornerstone of an education system that is effective and efficient in providing quality education to all children. Punjab and Sindh have already launched reforms to improve teacher quality and accountability, focusing on recruitment and management policies. These policies should be implemented throughout the country and entail four steps. First, clear standards need to be set for recruitment, deployment, transfers, and postings, with strong safeguards against decisions not based on merit. Allocation of teaching posts to schools should be based solely on student– teacher ratio norms and minimum staffing criteria (for instance, a minimum of two teachers per school to ensure a functional school). Second, preservice and in-service training needs to operate on three levels: r aise teacher subject knowledge, equip teachers with up-to-date approaches  to teaching, and help teachers adopt effective pedagogical methods to enhance student learning. Third, programs to provide intensive support to teachers in low-performance schools need to be designed. These programs should regularly provide on-site Expanding access to quality education support to teachers to address the challenges they face in their particular environments (a multigrade classroom, for instance). A good example is the Punjab program to provide onsite advisory support to teachers through a 162 network of field-based district teacher educators and teacher educators based in high schools or government elementary teacher education colleges. Man- aged by the Directorate of Staff Development under the Punjab School Educa- tion Department, this field-based support system serves as a new and promising conduit for providing regular, customized teacher support and improving teaching performance. Fourth, teachers need to know that if they acquire new skills and perform well, they will be rewarded—and vice versa. To improve teacher performance, incentives that link teacher pay to performance need to be built into the system. A good example is the Improver’s Program of the Punjab Education Sector Reform Program. Launching strong accountability mechanisms (such as penalties for absenteeism) is also important. Improve governance, efficiency, and quality, using financing as a tool School finance systems must provide the necessary resources so that all students, regardless of background, can learn. All provinces and areas need mechanisms that ensure adequate resources for all schools. The mechanisms should be aimed at meeting the minimum requirements for creating and maintaining acceptable learning conditions. Minimum funding ­ mplemented—and be objective, standards for schools should be developed and i criteria-based, and linked closely with the needs of schools (most specifi- cally, the number of students in the schools). Other factors could include needs relating to school levels, school buildings, availability of basic ameni- ties, and so on. More funding is required to improve resource-use efficiency and learning out- comes. Pakistan could consider financing tools that have shown promise in other countries, such as changes in the incentives structure for both teach- ers and schools. Four promising approaches stand out. First, accountability systems (such as performance-related pay and promotion) based on student learning achievements could help modify teacher behavior and stimulate more effort in the classroom. Second, modifying school funding formulas has the potential to create incentives for quality improvement. Block grants that do not demand accountability from schools could be replaced by grants that carry a range of incentives for efficiency and equity. (Box 9.3 presents the case of Armenia, where this move has been effective.) Third, financial powers need Box Ensuring adequacy with per capita funding, Armenia 9.3 Armenia has undertaken a series of major (but incremental) education finance reforms—including decentralization and the use of per capita funding—to respond to demographic and political changes. 163 After the Soviet Union dissolved in the early 1990s, Armenia’s education expenditures and enrollment declined but its number of teachers rose. The inefficiency in spending was the result of previous school funding methods, based on a simple adjustment of historic levels—bargaining, and discretionary decisions. To increase efficiency, Armenia shifted the responsibility for managing education resources to admin- istrators at the subnational level through decentralization and to schools through school-based manage- ment reforms. A per capita formula made funding more transparent. Resources are now allocated per student and adjusted for rural location, level of schooling, and special needs. These structural reforms, piloted in 1999 and implemented throughout the country by 2007, have improved the efficiency of education finance tremendously, increasing class size to appropriate levels and making surplus teachers redundant. to be decentralized closer to the school level. School financing is currently managed at the tehsil level, with several hundred schools (in the larger provinces) under the purview of the drawing and disbursing officer. Fourth, greater use of public–private partnerships—Punjab and Sindh already have them—could also increase resources for education and maximize efficiency, as long as efficiency and equity incentives are built into the agreements. Leverage the contribution of the private sector In the face of capacity and resource constraints, leveraging the contribution of the private sector is key to meeting the challenges of improved access and quality. Since the private sector has already demonstrated that it can offer access at lower cost and with similar or better outcomes than the public sector, countries will gain by helping it expand, easing barriers to entry, and carefully designing PPPs. Innovative and cost-effective programs in Sindh and Punjab could be scaled up (Box 9.4). Improve sector coordination and support The federal government has an important role to play in promoting cohesion of the education system nationwide,whether by setting universal teacher standards or by playing a role in curriculum development. Coordination among provinces—a key function of the Ministry of Education and Training— needs to be strengthened. The ministry should play a key role in improving communication between the provinces/areas to facilitate knowledge exchange for improving Pakistan’s education indicators. Expanding access to quality education The promise of public–private partnerships for improving Box education quality 9.4 Several public–private partnerships introduced in Pakistan have generated cost-effective gains in 164 participation and achievement. Introduced in 2005, the Foundation-Assisted School program administered by the Punjab Education Foundation provides conditional cash subsidies to low-cost private schools to open up private schooling opportunities for children from low-income households and to raise the level of learning. Per student cash subsidies are provided monthly, with essentially no conditions on how they are to be used. The amount is purposely set low (half the estimated per student cost in the public school system) to ensure that only low-cost private schools self-select into the program. In return for the subsidy, the program school has to waive tuition and other fees for all students and ensure that the school achieves a minimum student pass rate in the Quality Assurance Test (QAT). The QAT is a curriculum-based, multisubject test designed by subject specialists and admin- istered by independent testing agencies. Program schools are also eligible for group bonuses for teachers who achieve high QAT pass rates and for competitive bonuses for schools that rank highest in the QAT. Schools that do not achieve a minimum pass rate twice in succession are dropped from the program. As of June 2010, the Foundation-Assisted School program had proceeded through six phases of expansion and supported about 800,000 students in 1,800 schools in 29 of the 36 districts in Punjab. Rigorous evaluation (Barrera-Osorio and Raju 2010, 2011) found that within two years, the program generated major gains in enrollment and school inputs (roughly 40 percent) and student achievement (0.3–0.5 standard deviation). The Punjab Education Foundation runs a sister program, the New School Program, which sup- ports the opening of new schools in underserved communities. The program provides per student subsidies to new private schools in underserved areas, conditional on a school’s achievement on standardized, competency-based tests. The program (early 2013) covers more than 20,000 students in 230 schools in 16 districts. A similar program, Promoting Private Schooling in Rural Sindh, is run by the Sindh government. It, too, attempts to leverage the private sector to deliver schooling to underserved rural communi- ties. Program schools get grants for construction and other support, as well as per student subsi- dies conditional on maintaining minimum student achievement. A rigorous evaluation found that the program produced substantial gains in participation and achievement (Barrera-Osorio and others 2011). Enhance learning-assessment systems Enhanced learning-assessment systems are necessary to monitor progress in learning outcomes and in improvements in schooling quality over time. These comprehensive assessments need to cover students in both public and private schools. Yet Pakistan’s student assessment systems, which are at the provincial level, are weak and do not provide regular information about student performance. Pakistan needs to strengthen these systems to emphasize classroom testing. It also needs to include large assessments. Notes Given the differences in challenges, opportunities, and priorities by level of education, this note focuses on primary and secondary education. 165 1. This is a composite index that provides an assessment of a country’s education system on four Education for All goals: universal primary education, adult literacy, quality, and gender. 2. All statistics in this paragraph and the next one are staff estimates based on various rounds of the Pakistan Social and Living Standards Measurement Surveys. 3. The net enrollment rate figures are based on the officially recognized age of children for each education level. 4. Assessment, Survey, Evaluation, Research. 5. World Bank staff estimates using student test score data from the Punjab Examination Commission. References Andrabi, Tahir, Jishnu Das, Asim Ijaz Khwaja, Tara Vishwanath, and Tristan Zajonc. 2007. “Pakistan: Learning and Educational Achievements in Pun- jab Schools (LEAPS): Insights to Inform the Educational Policy Debate.� www.leapsproject.org/assets/publications/LEAPS_Report_ExecSummary.pdf. ASER-Pakistan. 2011. The Annual Status of Education Report—Pakistan. South Asia Forum for Education Development and ITA, Lahore, Pakistan. Barrera-Osorio, Felipe, and Dhushyanth Raju. 2010. “Short-Run Learning Dynamics under a Test-Based Accountability System: Evidence from Paki- stan.� Policy Research Working Paper 5465, World Bank, Washington, DC. ———. 2011. “Evaluating Public Per-student Subsidies to Low-Cost Private Schools: Regression-discontinuity Evidence from Pakistan.� Policy Research Working Paper 5638, World Bank, Washington, DC. Barrera-Osorio, Felipe, David S. Blakeslee, Matthew Hoover, Leigh L. Linden, and Dhushyanth Raju. 2011. “Expanding Educational Opportunities in Remote Parts of the World: Evidence from a RCT of a Public Private Partner- ship in Pakistan.� In Third Institute for the Study of Labor (IZA) Workshop, “Child Labor in Developing Countries,� Mexico City. www.iza. org/conference_files/ childl2011/blakeslee_d6783. Hanushek, Eric A., and Steven G. Rivkin. 2010. “Generalizations about Using Value-Added Measures of Teacher Quality.� American Economic Review 100 (2): 267–71. Nye, Barbara, Spyros Konstantopoulos, and Larry V. Hedges. 2004. “How Large Are Teacher Effects?� Educational Evaluation and Policy Analysis 26 (3): 237–57. Expanding access to quality education Pakistan Bureau of Statistics. 2011. Pakistan Social and Living Standards Measurement Survey 2010/11. Government of Pakistan, Islamabad. Park, Albert, and Emily Hannum. 2001. “Do Teachers Affect Learning in 166 Developing Countries? Evidence from Matched Student-Teacher Data from China.� Paper presented at the Social Science Research Council Conference “Rethinking Social Science Research on the Developing World in the 21st Century,� Park City, UT, June 7–11. Rivkin, Steven G., Eric A. Hanushek, and John F. Kain. 2005. “Teachers, Schools and Student Achievement.� Econometrica 73 (2): 417–58. Rockoff, Jonah E. 2004. “The Impact of Individual Teachers on Student Achieve- ment: Evidence from Panel Data.� American Economic Review 94 (2): 247–52. Sanders, William L., and June C. Rivers. 1996. “Cumulative and Residual Effects of Teachers on Future Student Academic Achievement.� Research Progress Report. University of Tennessee Value-Added Research and Assessment Center, Knoxville, TN. UNESCO (United Nations Educational, Scientific and Cultural Organization). 2012. Education for All Global Monitoring Report 2012—Youth and Skills: Putting Education to Work. Paris. UNESCO Institute for Statistics. n.d. Data Centre. http://stats.uis.unesco.org/ unesco/tableviewer/document.aspx?ReportId=143. World Bank. 2013. “Improving School Quality and Learning Outcomes in South Asia�. Draft Sector Report. Human Development Unit, South Asia Region. Washington, DC. Shahnaz Kazi, Inaam Ul Haq, Luc Laviolette, and Kees Kostermans Expanding Quality 10 Health, Population, and Nutrition Services Pakistan’s health and nutrition outcomes and service coverage lag behind those in most other South Asian countries, despite slowly improving over the past decade. Key issues include persistent inequities in health and nutrition outcomes and service use by economic status, gender, and region; poor governance and weak and centralized management; low public spending on health; and programmatic shortcomings in reducing fertility and improving nutrition—areas with cross-cutting impacts on human development and economic growth. The adoption of the 18th Amendment to the constitution in 2010 and the subsequent devolution of most federal responsibilities for health and population welfare to the provinces provide opportunities for more responsive and accountable governance, but they also pose severe challenges. Special attention will be required to ensure appropriate institutional arrangements to house federal functions, a clear delineation of responsibili- ties, and the building of capacities and structures at all levels. Actions to improve health sector performance include improving health services targeting the poor; increasing health spending; strengthening health sector management and accountability with a greater focus on monitoring and information (in a context of devolution and contracting); and expanding family planning and nutrition services. Health and population outcomes in Pakistan have improved over the last decade but at a slower rate than in most other South Asian countries. While the infant mortality rate and under-five mortality rate have fallen, the decline is far slower than in other South Asian countries, and both rates are the second highest in South Asia, as is the total fertility rate (Table 10.1). High fertility puts an enormous burden on women’s health, as reflected in high maternal mortality ratios (260 deaths per 100,000 births). Pakistani women and children also suffer from some of the highest rates of malnutrition in the world, and have seen little improvement in nutritional outcomes. The prevalence of nutritional stunting among children under age 5 (43.7 percent) has remained virtually unchanged since 1965.1 The country has the second-highest rate of severely wasted children, after India: 15 percent of children under age 5 suffer from acute malnutrition and 6 percent suffer from severe acute malnutrition. Malnutrition is also a significant problem among Expanding quality health, population, and nutrition services Health, nutrition, and population outcomes in selected South Table Asian countries 10.1 Malnutrition prevalence, 168 Under-five mortality Infant mortality height for age Maternal mortality rate (per 1,000 rate (per 1,000 (share of children ratio (per 100,000 Total fertility rate live births) live births) under age 5) live births) (births per woman) Country 2011 2011 2004–11 2010 2010–11 Bangladesh 46 37 43 240 2.2 India 61 47 48 200 2.6 Nepal 48 39 41 170 2.7 Pakistan 72 59 43 260 3.4 Sri Lanka 12 11 19 35 2.3 Source: World Development Indicators database. women of reproductive age, as 14  percent have chronic energy deficiency. Micronutrient deficiencies are widespread with high rates of iron-deficiency anemia, with 62 percent of children under age 5 and 51 percent of pregnant women suffering from anemia and persisting high deficiencies of zinc, iodine, and vitamin A. Coverage of most maternal care services has improved significantly over the last decade but is still far from adequate. During 2001–10, antenatal care attendance increased from 35  percent to 60  percent, postnatal care attendance from 9 percent to 28 percent, skilled birth attendance from 31  percent to 43 percent, and tetanus toxoid 2 coverage in women from 46 per- cent to 69  percent. Immunization coverage improved modestly, increasing from 53  percent to 62  percent. 2 Contraceptive prevalence (using modern methods) remained stagnant at 22 percent (or less than half the rate of most South Asian countries). By province, health coverage remains better in Punjab and Sindh, while Balochistan stays the most underserved (Table 10.2). The private sector is the main source of outpatient consultations and institu- tional deliveries. The public sector provides only a fifth of curative services, even for the poor and rural population. The use of the private sector for outpatient consultations has increased from 69 percent in the late 1990s to 78 percent in 2010/11. Nearly 70 percent of institutional deliveries took place in private facilities in 2011. The public sector remains the main source of preventive services, particularly in rural areas, and accounts for 90 percent of immunization coverage. Over the last decade, Pakistan has aimed to improve service delivery through management reforms and a stronger programmatic focus on strengthening Table Health service coverage, federal and by province 10.2 (percent) Proportion of 169 Proportion of pregnant women Treatment of pregnant women receiving at least diarrhea in who visited one tetanus children with oral health facility for toxoid injection Skilled birth Postnatal care by rehydration salts antenatal care during pregnancy attendancea skilled provider Province 2001 2010 2001 2010 2001 2010 2004 2010 2001 2010 Pakistan 54 74 35 60 46 69 31 43 9 28 Punjab 44 64 40 63 53 77 33 44 10 28 Sindh 70 89 38 61 43 60 38 49 10 34 Khyber Pakhtunkhwa 58 83 22 50 35 61 26 37 4 23 Balochistan 70 90 21 40 17 31 14 23 7 18 a. Includes assistance from doctor, nurse, lady health worker, or midwife. Source: Pakistan Bureau of Statistics 2001, 2011. maternal and child health services. Fundamental institutional changes enacted through the 18th Amendment have implications for health sector manage- ment. Key reforms include: • Expanding access to maternal neonatal and child health services. The government made substantial investment over 2001–10 in expanding community-based programs, including the Lady Health Workers (LHW) Program and the National Maternal Neonatal and Child Health Program. This expansion supports the provision of family planning, basic primary health care, community midwifery, and emergency obstetrical services. • The Devolution Plan of 2001 envisaged a shift of responsibility for basic health services from the provincial governments to the newly created district governments, to improve service delivery by bringing government closer to the people . The provin- cial and district departments were restructured, and efforts were made to build capacity of district health systems. Yet the initiative stopped far short of full devolution of administrative and fiscal powers. District governments have limited administrative authority to hire and fire staff or make senior appointments, which are within the purview of the provincial government. The share of district governments in public expenditure increased, but the bulk of spending comprised salaries that could not be altered. The impact on service delivery was limited. • The 18th Amendment enhances provincial autonomy through devolving federal powers and responsibilities to the provinces of subjects in the concurrent legisla- tive list, which includes health and population welfare . The federal ministries of health and population welfare have been abolished. All vertical health programs that accounted for about 60 percent of Ministry of Health spending have been transferred to the provinces but will continue to be Expanding quality health, population, and nutrition services financed by the federal government until the next National Finance Commission award, expected in 2015. Some health oversight functions have been retained at the federal level and assigned to various federal entities, 170 including health financing, formulation of norms and standards, human resource planning, and information collection and analysis. The reform has the potential to make the government more responsive and accountable and to develop a more cohesive public health system, yet it faces significant challenges that will require attention, particularly during the transition phase. These include appropriate institutional arrangements to house federal functions, clear delineation of responsibilities within provincial departments, and capacity building at all levels. The provinces are develop- ing their own health sector strategies outlining their reform programs. • Contracting out health services management . More than half the districts adopted this approach, under the People’s Primary Health Initiative (PPHI). The model was first adopted in 2002 when Rahim Yar Khan District in Punjab contracted the Punjab Rural Support Program to man- age all 104 basic health units in the district. The intervention demonstrated that effective management could quickly increase use of public facilities without additional cost to the government, thus increasing efficiency. The program was scaled up in 2006. Findings of an external evaluation of PPHI in Balochistan, Khyber Pakhtunkhwa, and Sindh confirm the rise in utiliza- tion. Outpatient attendance increased (threefold in Khyber Pakhtunkhwa, twice in Balochistan, and by 25 percent in Sindh; TRF 2010). Staffing and physical conditions of facilities improved. And there were greater levels of satisfaction with PPHI managed services. • Health insurance for the poor. The Benazir Income Support Programme (BISP) in 2012 launched a pilot health insurance scheme in the first government initiative aimed at protecting poor households from the costs of catastrophic illness that involves hospitalization. Coverage is limited to the beneficiaries of BISP, who are identified through a poverty score card. The State Life Insurance Company has been contracted for managing the inpatient package of services. The initiative is financed directly by the federal budget. Launched in Faisalabad District, BISP aims to expand the program to the entire country. The progress of the pilot will be carefully monitored and evaluated before scaling up. Main Issues Health outcomes are influenced by several factors, some outside the health sector such as poverty, education (particularly for girls), and environmental (sanitation and water supply) factors. With challenges in most of these, the sector faces a range of key issues. Inequities in health outcomes and access to services Wide inequities persist in nearly all health outcomes and access to services between rich and poor (the most pronounced) and between rural and urban 171 regions. Under-five mortality and fertility rates of the poor are twice as high as those for the wealthiest households, while malnutrition, particularly stunt- ing, is more prevalent in rural areas: 46 percent of rural children are stunted compared with 37 percent in urban households. Health service use also varies markedly by economic status and region, with the rural poor having substantially lower use than the urban poor. 3 Immunization rates for children from the poorest urban households are comparable with those of rural children from the richest. Contraceptive prevalence among urban households in the lowest quintile is higher than for rural households in the top quintile. Significant interdistrict variations in coverage of health services are observed in Balochistan, Khyber Pakhtunkhwa, and Sindh (Pakistan Bureau of Statistics 2011), which to some extent reflect differences in socioeconomic development and communities’ remoteness. A few efforts have been made to target services to poorer communities and to more remote districts, although the potential of the LHW Program and contracting out to the nongovernmental organization (NGO) under the PPHI has not been fully exploited. With more LHWs, some progress is being made to cover less advantaged areas, but the program has been unable to reach the poorest areas due to difficulties in recruiting LHWs given the limited supply of women in these areas who meet the program’s minimal educational require- ments (OPM 2009). To ensure a focus on the poor would require explicit mention of the objective in the service package to be delivered by NGOs and in the monitoring and evaluation (M&E) component of the PPHI contract. Gender disparities have diminished over time. Sex differentials in child mortality have narrowed but have not disappeared. The lower mortality of boys ages 1–5 is attributed to better health care (Table 10.3). Research on determinants of nutrition in Pakistan has not found any significant differences in nutritional outcomes among boys and girls (Arif and others 2012; World Bank 2005). Differences in immunization coverage have also nar rowed, although girls are still less likely to be fully immunized than boys. Improvements in women (even in rural settings) seeking health care during pregnancy and in skilled birth attendance partly reflect efforts to address gender constraints through expansion of the LHW Program and more deployment of trained community midwives in rural communities (see Table 10.2). Expanding quality health, population, and nutrition services Table Differences in health status and service coverage 10.3 Under-five Child Contraceptive Births attended Prenatal care Children ages Total mortality rate mortality prevalence by skilled by skilled 11–23 months 172 fertility (per 1,000 rate rate (modern personnel provider immunized Indicators rate live births) (ages 1–5) methods) (percent) (percent) (percent) By wealth quintile Poorest 5.8 121 30 12.4 15.0 36.9 25.9 Richest 3.0 60 8 31.6 77.3 91.9 63.7 By area of residence Urban 3.3 78 13 29.9 60.0 78.1 54.2 Rural 4.5 100 20 17.7 30.0 53.4 44.0 By gender Male 93 14 49.8 Female 93 22 44.3 Source: National Institute of Population Studies and Macro International 2008. Low public spending on health Total health spending in Pakistan is extremely low relative to other countries in the region as well as to countries at similar levels of development. About $22 per person was spent on health in 2009 (against an average of $41 in Southeast Asian countries). About 70 percent comes from private sources, mainly out of pocket by households at the point of care. Few households have access to financial risk protection against catastrophic diseases and accidents, even though such health shocks, attendant income losses, and associated out-of- pocket payments greatly increase the risks of impoverishment. Public funding on health was less than 0.86  percent of GDP in 2010, compared with at least 3–4 percent in other low-income countries, even though in 2008/09–2010/11 total public health and population welfare expenditure increased 36 percent in nominal terms to more than PRs 122 billion ($1.34 billion).4 The increase was largely attributable to higher current spending, which rose from 66 percent of the total to 72 percent—reflecting a 50 percent increase in salaries of provincial employees—while development expenditures fell, even in nominal terms, from 34 percent to 28 percent. Population welfare spending remained low, at 3.2 percent of the total, declining in real terms. At a more disaggregated level, in provinces in 2008/09–10/11 district nonsalary expenditures stayed low (TRF 2011). District budgets cover primary and second- ary care facilities, including basic health units, rural health centers, tehsils, and district headquarter hospitals. In Khyber Pakhtunkhwa and Punjab, the district share of overall expenditures remained at around 30 percent, comprising current expenditures only as the districts receive no allocation for development funds. Nonsalary expenditure accounted for only 14 percent of the total in Khyber Pakhtunkhwa and 25 percent in Punjab. 173 Provincial revenues have recently increased. Changes in the National Finance Commission award have considerably enhanced the provinces’ fiscal space by sharply raising their share in the divisible tax pool, and the 18th Amendment devolved sales tax on services, a buoyant source of revenue for the provinces. These decisions raised transfers to the provinces by 23 –28 percent, with greater benefits to Balochistan and Khyber Pakhtunkhwa. In addition, the federal government is commit- ted to financing the devolved vertical health programs until the next award in 2015. The amounts involved are substantial. In 2010/11, the federal government share of total expenditure was 19  percent, nearly two-thirds for development programs (including vertical national health and population welfare programs). A major concern since the 18th Amendment is whether the additional provincial resources will be used for national priorities. Such resources include preventive and primary health care services and nonsalary spending, all grossly underfunded and essential for facilities to function effectively. Past trends in provincial spending raise some concerns about lack of prioritization and political will to invest in these areas. Today, there is no instrument to incen- tivize provinces to focus on priority services. Weak management and governance Poor health performance is due mainly to weak management and gover- nance, including widespread staff absenteeism, centralized management, and weak stewardship. Staff absenteeism is the most serious manifestation of weak management and governance, including lack of accountability in public health services. According to facility-based surveys in Balochistan and Sindh, most doctors were absent from their assigned posts. The absentee rate was 58 percent in Balochistan while in Sindh 45 percent of doctors were absent from basic health units and 56 percent from rural health centers (World Bank 2010). They can get away with this because of political patronage and managers’ weak administrative authority, who are not in a position to fire public sector staff. Political interference is also prevalent in appointments and postings, result- ing in lack of merit-based recruitment and frequent transfer of managers. Expanding quality health, population, and nutrition services Poor staff performance reflects systemic problems of weak motivation due to rigidity of civil service rules, which provide few performance incentives or flexibility for managers to innovate. Contracting out management has 174 worked in Pakistan and elsewhere largely because it can circumvent civil ser- vice regulations and provide the necessary economic incentives and auton- omy to managers (Loevinsohn and Harding 2004). The success of the PPHI and similar contracting-out experience after the earthquake in Battagram District in 2005 is thanks to capable managers who were motivated, well paid, and granted wide autonomy. Such autonomy included flexibility in fund management through a single line budget, full administrative authority for hiring and firing staff, and scope for providing financial incentives to attract staff, particularly female providers who were in short supply. Overly centralized management further erodes accountability and efficiency. Efforts to decentralize service delivery under the Devolution Plan of 2001 were unsuccessful as increased responsibility for basic health services was not associated with the requisite authority at the district level to allocate resources or manage human resources, both of which remained largely with provin- cial governments. Thus provincial governments remained occupied with service provision, with little time for oversight. Centralized management of vertical programs and other service delivery institutions left little time for the Ministry of Health to focus on its stewardship role, causing neglect of key public health functions including policy formulation, oversight, M&E, sur- veillance, and regulation. Capacity for stewardship functions was lacking at provincial and federal levels. But the 18th Amendment provides an opportunity to delineate clear channels of accountability, effectively integrate services, and focus on stewardship roles, but also raises major challenges (WHO and others 2012; Nishtar 2010). The most pressing issue concerns inadequate federal institutional arrangements. After the Ministry of Health was dissolved, federal functions were assigned to federal entities with little experience or motivation for taking them on. The fragmented setup led to problems of coordination between these entities, between federal and provincial governments, and with donors. The lack of a central authority left the provinces with little technical support or guidance for their new responsibilities. Programmatic issues—unmet need for family planning and the burden of malnutrition Programmatic achievements have been few in reducing fertility and tackling malnutrition, with barely any change in outcomes or intermediate indicators. Family planning, in addition to reducing the mortality of mothers, infants, and children, also greatly boosts primary schooling rates and women’s empowerment. Fertility decline also contributes to growth by increasing the relative size of the workforce and reducing the dependency burden. The 175 long-term benefits of early childhood interventions in nutrition on adult health, economic productivity, and life-time earnings are well documented. Unmet need for family planning. Pakistan is the world’s sixth most populous country, whose population of 180 million is likely to double in about 39 years if current growth rates persist. It will face a huge challenge in meeting the basic needs of this rising population, which will undermine its ability to sustain solid economic growth. Some of the indicators are worry- ing. The fertility decline from 5.6 children per woman in 1990/91 to 4.1 in 2006/07 was much slower (and later) than that in any of its neighbors. The contraceptive prevalence rate has stagnated over the last decade at half the rate of other South Asian countries. And the unmet need for family planning is rising—as is unwanted fertility. The increased demand with- out any changes in contraceptive use can partly be explained by access constraints and poor quality of service provision. One in four women who want to avoid pregnancy is not using any form of contraception, while two-fifths of pregnancies are unwanted (National Institute of Population Studies and Macro International 2008). Women from poorest households have the highest unmet need. Although Pakistan was one of the first countries to establish a family plan- ning program in 1965, the decline in fertility that started in the early 1990s was much later than in most of South Asia. A comparison with Bangladesh is illustrative, as the two counties started with identical levels of fertil- ity and similar sociocultural contexts. Between 1971 and 2011, fertility in Bangladesh declined steeply from 6.3 children per woman to 2.3. Bangladesh virtually stabilized population growth as it made its program a top development priority with a broad coalition of support. In contrast, Pakistan’s program has suffered from wavering political commitment and has not been central to the development agenda. Having two separate ministries (of population welfare and of health), with vertical structures going down to service outlets, was not only inefficient but also marginalized population issues at the Ministry of Health, which failed to fully own the program. Government spending on population, which increased during the 1990s, also shows a declining trend in more recent years—for example, no new initiatives have been launched to increase contraceptive use since social marketing was introduced and expanded in the late 1990s and the LHW Program was expanded in the first half of the 2000s. Expanding quality health, population, and nutrition services Nutritional outcomes have not improved over the last two decades—worse, they have deteriorated for some indicators. Pakistan has made little attempt to systematically address the causes of m ­ alnutrition—it has not made them a 176 priority, as reflected in the lack of an institutional home or a clear strategy. It has carried out a few fragmented interventions, mainly to address micronu- trient deficiencies (primarily vitamin A supplementation and salt iodization). More recently, Pakistan has started to develop a broader nutrition strategy and program in line with the international consensus on an action framework for scaling up nutrition. The strategy needs to be based on a two-pronged approach: addressing the determinants of nutrition through a multisectoral approach, and implementing and scaling up well-proven direct nutrition interventions through the health sector. Policy Options Improve targeting of health services to the poor Interventions and resources need to be better targeted to the poor as well as to lagging districts and regions. Further expansion of the LHW Program is necessary because those regions not yet covered by it are the most disadvantaged. Yet a constraint to further expansion is the difficulty in recruiting workers with the minimum educational requirements as well as the lack of functional health facilities in the underserved areas. To increase coverage, condensed-education courses to motivate girls to become LHWs and mobilization of disadvantaged communities to support the LHWs should be implemented. Another promising option to expand coverage is to contract NGOs, as they have greater flexibility to reach out to populations in difficult areas (see below). Increase expenditure on health Efficiency of resource use can be greatly increased, but provincial governments also need to boost spending, given the very low public health expenditures. In particular, district budget spending on the nonsalary component is grossly inadequate and needs to be raised substantially. Stronger funding is also needed if the reforms and policy options in this note are to be carried out. Federal government support to vertical programs would be more effec- tive and would inculcate greater ownership if it were financed through development grants and were incorporated in the provincial annual develop- ment program. In the medium term, provincial allocations need to expand considerably to finance the vertical programs, particularly the LHW Program and other devolved institutions now supported by the federal government. The federal government may also need to provide conditional grants to ensure that provincial policies are aligned to national priorities in areas such as 177 fertility reduction and nutrition, as otherwise they may not receive the required attention and resources. This support could be in the form of results-based financing or tied grants. Strengthen health sector management and accountability The 18th Amendment will require a medium- to long-term implementation plan to fully realize its potential for responsive and accountable governance. Supportive measures to manage the transition should include: • Establishing a federal institution, possibly under the Ministry of Inter- provincial Coordination, responsible for national functions in health that are currently dispersed across different ministries. • Delineating and clarifying roles and responsibilities for structures within provincial departments. • Expanding merit-based recruitment of staff and building capacity in key areas at federal and provincial levels. • Undertaking an assessment of staff capacity requirements to prioritize needs. • Finalizing and approving provincial health strategies. Service delivery should ideally be devolved to district governments along with the necessary administrative and spending powers. The 18th Amendment recognizes the third tier of local government, but the district government’s role and responsibilities are defined by the provinces in the Provincial Local Government Ordinance Act. As all provinces have opted for greater provincial control, the best option would be a deconcentrated system, with administrative powers at the district level. A stronger focus on results and M&E is a central element of the health sector management and accountability reforms. The federal government has an important role in M&E of provincial performance to track progress toward priority areas and to serve as the basis for advocacy with the provinces. A central role is also necessary for ensuring consistency in methods and instruments in data collection and indicators, for collating evidence, and for reporting progress at international forums. The provincial departments require data and analysis for feeding into the planning and policy-making process, to monitor imple- mentation, and to improve service delivery. Given the wide variations in health sector performance, provincial governments should carefully track and then Expanding quality health, population, and nutrition services share information on district health performance (partly as an accountabil- ity mechanism but also to find out where special efforts are required). Poorly performing districts should receive special programmatic and technical support. 178 A basic prerequisite for these tasks is a well-functioning M&E system. This entails sound information systems and strengthened capabilities for data analysis, possibly through existing structures such as the National Health Information and Resource Centre or the Health Systems Strengthening Project at the federal level and the Health Sector Reform Units in provin- cial departments. Investment is required in improving quality, completeness, and timely availability of District Health Information System data to monitor performance of government health services. In addition, information from population-based surveys regularly conducted by the Federal Bureau of Statis- tics should be used. The PPHI model is a viable option for improving performance of first-level care facilities through greater managerial autonomy, including flexibility in funding and full administrative powers to hire and fire. The approach should be further strengthened through the following measures at the provincial level: • Adopting a package of primary health services covering preventive and pro- motive care. • Granting managers control of all aspects of the primary health system, including basic health units, rural health centers, vaccinators, and LHWs. • Selecting NGOs through a competitive process and bring in performance-based contracts with a greater focus on monitoring. • Contracting NGOs to improve coverage of remote rural areas where provision of public services is constrained by staffing difficulties with an explicitly stated objective of the service package and of the M&E component of the contract. • Building capacity of government counterparts in contract management and monitoring. Expand provision of family planning and nutrition services Pakistan needs to prioritize family planning and nutrition. It needs to invest in improving provision and quality of family planning services, focusing on rural areas where supply constraints are more severe, and should pursue the following measures: • Ensuring coverage of family planning services through all public health outlets. • Ensuring the provision of skilled staff and family planning products to make a broad range of contraceptive methods available. • Broadening social marketing of family planning by expanding services to rural areas, where possible using performance-based contracts with the private sector. • Promoting male involvement, including through information and services 179 by trained male paramedics or doctors in health facilities. Pakistan has to invest heavily and systematically in order to address malnutri- tion on a broad scale. It should do this by: • Prioritizing nutrition in the national development agenda and assuring resources to carry out multisectoral provincial nutritional plans through inter- ventions in education, agriculture, social protection, and water and sanitation. • Ensuring implementation of provincial plans for scaling up nutrition interventions through the health sector for vulnerable women and children. The plans include promoting exclusive breast-feeding, promoting adequate complementary feeding, addressing micronutrient deficiencies, treating severely malnourished children through community-based approaches, and controlling childhood infections and increasing immunization. • Building institutional capacity for nutrition in the health sector and in a coordinated multisectoral approach. Notes 1. All data are from the Pakistan National Nutrition Survey 2011 unless otherwise indicated. The 2011 survey is the first to provide representative data for each province. By comparison, India is 48 percent, Nepal 45 per- cent, Bangladesh 43 percent, the Democratic Republic of Congo 43 per- cent, and Sri Lanka 17 percent. 2. Indicators are based on the Pakistan Social and Living Standard Measure- ment Survey (PSLM), except for immunization (Masud and Navaratne 2012). PSLM data were not used for immunization because of the large divergence between immunization rates based on PSLM and other data sources, including the Pakistan Demographic and Health Survey and Multiple Indicator Cluster Survey. The high rates of immunization reported in PSLM are also not in line with the continued outbreaks of vaccine-pre- ventable diseases such as measles. In the case of other indicators for mater- nal care services, the figures are comparable across the different surveys. 3. A breakdown of data by expenditure quintile and by rural–urban status was available in PSLM only through 2005/06. 4. Data are from the Ministry of Finance’s Poverty Reduction Strategy Progress Reports 2003–08 and TRF (2011). Expenditure data for Azad Jammu and Kashmir, Gilgit–Baltistan, Federally Administered Tribal Areas (FATA), and autonomous organizations were not included. Expanding quality health, population, and nutrition services References Arif, Ghulam M. 2004. “Child Health and Poverty.� Pakistan Development Review 180 43 (3): 211–38. Arif, Ghulam M., Saman Nazir, Maryam Satti, and Shujaat Farooq. 2012. “Child Malnutrition in Pakistan: Trends and Determinants.� Discussion Paper. Pakistan Institute of Development Economics, Islamabad. Garcia, Marito, and Harold Alderman. 1991. “Patterns and Determinants of Malnutrition in Children in Pakistan.� Pakistan Development Review 28 (4): 891–902. Institute of Public Policy. 2012, “Economic Cost of Undernutrition in Pakistan.� Islamabad. Loevinsohn, Benjamin, and April Harding. 2004. “Contracting for the Delivery of Community Health Services: A Review of Global Experience.� Health, Nutrition and Population Discussion Paper. World Bank. Washing- ton, DC. Masud, Tayyeb, and Kumari Navaratne. 2012. “The Expanded Program on Immunization in Pakistan: Recommendations to Improve Performance.� Health, Nutrition and Population Discussion Paper. World Bank, Washing- ton, DC. Ministry of Finance. Various years. “Poverty Reduction Strategy Progress Reports.� Government of Pakistan: Islamabad. National Institute of Population Studies [Pakistan] and Macro International. 2008. Pakistan Demographic and Health Survey 2006–07. Islamabad. Nishtar, Sania. 2010. “Health and the 18th Amendment: Retaining National Functions in Devolution.� Heartfile, Islamabad. OPM (Oxford Policy Management). 2009. “Lady Health Workers’ Program: Third Party Evaluation of Performance.� Oxford, UK. Pakistan Bureau of Statistics. 2001. Pakistan Integrated Household Survey 2001. Government of Pakistan, Islamabad. ———. 2011. Pakistan Social and Living Standards Measurement Survey 2010/11. Government of Pakistan, Islamabad. Social Policy Development Centre. 2012. “Devolution and Social Development, Annual Review 2011–12.� Karachi. TRF (Technical Resource Facility). 2010. “Third Party Evaluation of the PPHI in Pakistan.� Islamabad. ———. 2011. “Health Budget & Expenditure Analysis (2008–09 to 2010–11).� Islamabad. UNDESA (United Nations Department of Economic and Social Affairs). 2011. Sex Differentials in Childhood Mortality. New York: United Nations. WHO (World Health Organization), World Bank, DFID (Department for International Development), USAID (U.S. Agency for International Development), and TAUH (Pakistan Technical Assistance Unit for Health). 2012. “Devolution of Health Sector, Post 18th Amendment to the Constitution of Pakistan: Opportunities and Challenges.� Joint Mission Report, Geneva. 181 World Bank. 2005. Pakistan: Country Gender Assessment, Bridging the Gender Gap, Opportunities and Challenges. Washington, DC. ———. 2010. Delivering Better Health Services to Pakistan’s Poor. Washington, DC. ———. n.d. World Development Indicators database. http://data.worldbank. org/data-catalog/world-development-indicators. 11 Consolidating Iftikhar Malik and Lucian Pop Social Protection Pakistan has taken steps to deliver social protection services, including unanimously passing the Benazir Income Support Programme Act in 2010, and setting up systems for targeted cash transfers to the poor. Such measures are critical in a country with endemic poverty, significant food insecurity, low education, and poor health outcomes. But more efforts are needed, including bringing the federal and provincial governments together as partners in social protection service delivery, especially in the new decentralized context since the 18th Amendment to the constitution was approved in 2010. A national social protection framework is to be established, and its overarching objective should be a technically sound and financially viable system that provides appropriate, adequate, and predictable benefits to the target population. To meet this objective, policy makers can ensure that the framework is coherent between federal and provincial authorities, harmonize systems for design and delivery of social protection within that framework, ensure financing for social protection based on affordability and efficiency principles, and disaster early recovery cash transfers as part of the future disaster- enable effective post­ response action plan. Poverty in Pakistan is endemic, alongside significant food insecurity, low education, and poor health outcomes. The economy has struggled in recent years during the global economic downturn, with difficulties exacerbated by internal security problems and natural disasters. Steep price inflation for basic commodities has sharply lifted the cost of living. The poor suffer most from these shocks, and many of the poorest cannot meet their immediate needs for food, basic education, or health. The 2011 National Nutrition Survey shows food insecurity to be a daily reality for millions: nearly 30 percent of households experience food insecurity with either moderate or severe hunger (Aga Khan University, Pakistan Medical Research Council, and Ministry of Health 2011). Moreover, any income shock further increases their poverty, undermining their human development.1 Most have no savings or back-up resources to stop them falling into destitution. To address these challenges, the government has changed the profile of social protection markedly in the past five years. In June 2007, the Planning Com- mission launched a National Social Protection Strategy (NSPS), approved by the Federal Cabinet (Government of Pakistan 2007). The strategy reviewed social protection initiatives, concluding that Pakistan’s safety net developed largely as a series of ad hoc responses to problems thrown up by particular circumstances, and thus it contains duplicating and overlapping 183 programs (Government of Pakistan 2007). The NSPS identified gaps and shortcomings and set out a more comprehensive approach to social protec- tion for the country, with particular focus on safety nets. In 2008, with the immediate objective to cushion the adverse impact of the food, fuel, and finan- cial crises on the poor, the federal government launched the Benazir Income Support Programme (BISP), a national cash transfer program, by committing PRs 34 billion. The program was supported by legal, institutional, and adminis- trative reforms to respond to the challenges identified by the NSPS. This paradigm shift resulted in a large increase in pro-poor social protection spending. From 2006/07 to 2011/12, pro-poor social protection expenditure (excluding universal subsidies)2 rose almost fivefold from 0.16 percent of GDP to 0.79 percent, approaching the Poverty Reduction Strategy Paper II target of 0.9 percent (Government of Pakistan 2009). This increase included heavy spend- ing on social assistance cash transfers targeted to the poorest and disaster relief. In 2011/12, BISP expenditures represented 0.25 percent of GDP, while expendi- tures for disaster relief were 0.40 percent of GDP. It represents strong evidence of government intention to mitigate risks of shocks to the poorest and support pro-poor social protection to reduce income inequality. The country’s safety net has been modernized. The national cash transfer program under the BISP now serves as the country’s main safety net mechanism to provide minimum income support to female representatives of extremely poor families (Box 11.1). The program aimed to contribute to the redistributive goals of the federal government. It recognized that regular and predictable cash assistance for the poorest households would provide a reliable source of supplementary income, and it aimed to build the foundation for moving from the essential basic safety net to coherent social protection services that can help them to graduate out of poverty. The program contributed significantly to the country’s safety net modernization by introducing an objective targeting method and improving the administration and governance of cash transfer programs. BISP beneficiaries are selected through an objective targeting mechanism. From 2009 to 2012, the BISP undertook a national household census to establish an objectively verifiable welfare ranking of households to target its safety nets and other pro-poor interventions. This independently administered census, using a poverty scorecard3 and supported by various control and accountability processes, covered more than 27 million households and created consolidating social protection Box Impact of paying cash to women 11.1 Giving cash to women directly increases their bargaining power within the family, and when women have greater power to make decisions, the household increases its investment in human development. 184 Studies endorse this gender-differentiated impact in Pakistan. Data from the Pakistan Social and Living Standards Measurement survey show that when Pakistani women have more decision-making power, households spend significantly more on things that women value, including nutrition and child education (particularly for girls), and that women make greater use of reproductive health services. In addition, when a woman acquires a Computerized National Identity Card, she gets other rights of a citizen (voting, the option to open a bank account, and the like). From 2009 to 2012, more than 15 million female citizens obtained a computerized identity card, largely due to collaboration with the Benazir Income Support Programme. Source: World Bank 2011a,b; National Database and Registration Authority. a socioeconomic national poverty registry. The program is better targeted than other social assistance programs in Pakistan, such as parliamentarian selection at the start of the BISP, Bait-ul-Mal, and Zakat (Figure 11.1). The national cash transfer program under the BISP compares well with interna- tional best practice for design and administration of social assistance cash trans- fer programs. Alongside more transparent and efficient targeting, there were process improvements including a computerized management information and grievance redress system, an online complaints system, and other third-party mechanisms for spot checks, performance audits and a multiyear impact evalua- tion. New technology is used for payments to reduce the potential for fraud and corruption. Moreover, the BISP Act 2010 embraced a number of good principles Distribution of benefits (targeting accuracy) by quintile of Figure household consumption per adult equivalent 11.1 50 Benazir Income Support Programme Bait-ul-Maal FSP Zakat 40 30 Percent 20 10 0 Poorest Second Middle Fourth Richest quintile quintile quintile quintile quintile Note: Quintiles are defined by household consumption per adult equivalent. Source: World Bank 2013; Benazir Income Support Programme statistics represent simulations of the poverty scorecard using the Pakistan Social and Living Standards Measurement Survey 2007/08. Bait-ul-Maal and Zakat statistics are from World Bank (2007). of governance and accountability in delivering social protection. For instance, the members of the board are appointed from the government and nongov- ernmental organizations with equal representation (Government of Pakistan 2010a). 185 Coverage of the poor has increased. The families eligible for the safety net cash transfers under the BISP make up the poorest 21 percent of the country’s population. They receive a monthly cash transfer of PRs 1,000 (around $10), which is paid quarterly. The benefit amount was set in 2008 to provide income support for basic needs (that is, consumption smoothing). This small benefit does not discourage family members from working. By the end of 2012, as the program rolled out across all provinces and regions, 4.3 million women of the poorest families were receiving such cash transfers. As eligible women obtain Computerized National Identity Cards (CNICs) and register for payments, the number of eligible beneficiaries is expected to rise to more than 7 million families (the most poor and vulnerable in the c­ ountry—Figure 11.2). Detailed evidence of the overall impact of cash transfers will be available later in 2013 through the impact evaluation, although initial assessments suggest that the cash is spent on basic and pressing needs (World Bank 2010b). A recent study showed that 78 percent of the beneficiaries used their cash for food, while the next most common expenditure was on health (15 percent; ICF GHK 2012). The government recently introduced measures on promoting better health and education for the poorest. Many children of the poorest families do not attend primary school (Figure  11.3). The BISP recently launched Waseela-e- Taleem to provide supplementary monthly cash payments of PRs 200 a month Anticipated and current beneficiaries of monthly cash assistance, Figure March 2013 11.2 Families identi ed as eligible Families currently receiving bene t 3 2 Millions 1 0 Azad Jammu Balochistan Federally Gilgit– Khyber Punjab Sindh and Kashmir Administered Baltistan Pakhtunkhwa Tribal Areas Source: BISP 2012b. consolidating social protection Gender-disaggregated percentage of children (ages 5–12) not Figure attending school (poorest 21 percent of population) 11.3 100 Girls Boys 186 75 Percent 50 25 0 Punjab Sindh Balochistan Khyber Gilgit– Azad Jammu Federally Islamabad Pakhtunkhwa Baltistan and Kashmir Administered Capital Tribal Areas Territory Source: BISP 2012a. per child, over and above the PRs 1,000 received by BISP beneficiary families, but only if the primary school–age child enrolls and attends school (an example of a coresponsibility cash transfer).4 This will help in further supporting house- hold welfare and in addressing the intergenerational cycle of poverty. Waseela- e-Taleem is being piloted in five districts in collaboration with the provincial education authorities who will ensure the availability of schools and monitor student attendance. The federal government has publicly committed to scale up this program to reach 3 million children by 2015/16. At the same time, the BISP is trialing a health insurance initiative, Waseela-e-Sehat, in Faisalabad District and plans to expand it by adjusting the design and implementation arrangements after evaluation of the trial. Main Issues Social protection system after the 18th Amendment The roles and responsibilities of the federal and provincial governments in social protection need to be clarified, reflecting the institutional and fiscal land- scape after the 18th Amendment.5 The federal and provincial authorities share no understanding of their Islamabad FaisalabadGujranwala joint roles Lahore Multan in Rawalpindi Sheikhupura policy, Sialkot design, Hyderabad Karachiimplementation, Sukkur Quetta Peshawar and financing of social protection. Steps in this direction are especially important as social protection is not explicitly addressed in the constitution. In particular, strategic federal responsibilities and functions need clarification, as does how they would be integrated with the provincial programs to achieve equity for the poor and vulnerable in provinces with very different capacities and financial resources. Provincial buy-in to a dynamic and diversified social protection system is essential. Provinces need to prioritize and invest in the various social protec- tion interventions (social insurance, active labor market programs, social welfare services) if they are to achieve the desired sector outcomes. 187 Fragmentation and inconsistent administration practices across social protection programs Social protection programs are still fragmented and uncoordinated. Since 2008, the federal government has taken some good steps to meet its constitutional responsibility to help the poorest with a redistribution mechanism. But there is currently no coherent social protection system, just numerous disparate and uncoordinated programs operated by different ministries and departments at the federal and provincial levels (Kardar 2013). Programs set up to address specific problems remain ad hoc, reactive, isolated, and fragmented. Such programs include the initiatives of Paki- stan Bait-ul-Maal; Zakat; labor welfare programs; Rozgar initiatives in Khyber Pakhtunkhwa, Sindh, and Punjab; school stipends in some provinces as part of education reform programs; stipends for skills development programs; and special purpose technical and vocational training centers.6 There is a need to avoid program overlap and promote synergies between interventions. Administration practices across social assistance programs are inconsis- tent. But many such programs suffer from lack of objective and transparent beneficiary targeting and have unreliable and irregular systems for delivering benefits (Kardar 2013). Weak administration wastes resources and generates a poor profile among the public, which has little appreciation of the benefits of social protection. Financing social protection The government still devotes much social protection spending on universal subsidies, which are regressive and thus inefficient as a social protection mechanism. In recent years, the federal government has tried to offset inflation with untargeted (universal) subsidies. Since 2006/07, untargeted subsidies have recorded inordinate increases relative to trend gains in health, education, and social protection programs other than subsidies (Figure 11.4). In the severely constrained fiscal environment, the “unproductive expenses� on such subsidies represent a direct cost to the country’s human development and a major constraint to poverty reduction (Government of Pakistan 2011c). In 2011/12, more than 70 percent of what are defined as social protection outlays were untar- geted subsidies, and just 0.79 percent of GDP went to social assistance programs targeted to the poor.7 Untargeted subsidies were equivalent to almost 3 percent consolidating social protection Figure Human development and subsidies, 2006/07–2011/12 11.4 Health Education Social assistance and social security Untargeted subsidies 600 188 400 PRs million 200 0 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 Source: Government of Pakistan various years; authors’ calculations. of GDP, despite the government’s stated policy to move from “hefty subsidies� such as those on electricity, petroleum products, fertilizers (agricultural input), wheat (agricultural output), and utility stores (especially for ghee and edible oil; Government of Pakistan 2011c,d). Although inclusion of spending on universal subsidies in the Poverty Reduction Strategy Paper may suggest that they are “pro-poor,� in practice the poorest benefit the least from them. The main beneficiaries of electricity subsidies, which constitute about 80 percent of total subsidies in Pakistan, are in the richest quintile of the population, while the poorest quintile benefits from only about 10 percent of the subsidy (Figure 11.5). Proportion of benefits received for electricity subsidy, March Figure 2008 and March 2011 11.5 40 2008 2011 30 Percent 20 Islamabad FaisalabadGujranwala Lahore Sheikhupura Sialkot Hyderabad Karachi Sukkur Multan Rawalpindi Quetta Peshawar 10 0 Poorest Second Middle Fourth Richest quintile quintile quintile quintile quintile Source: Trimble, Yoshida, and Saqib 2011. Weak institutional capacity for early recovery disaster response Institutional capacity is insufficient for effective social protection response to disasters. Pakistan is prone to disasters, such as earthquakes, floods, and 189 conflict-related violence. Despite recent good experience in providing social protection through cash transfers to those hit by disasters, institutional capacity needs to be built to provide timely early recovery support in future disasters (Box 11.2). After the response to the 2010 floods, the federal government prepared a National Future Disaster Response Action Plan (FDRAP) in consultation with provincial authorities to outline how cash transfers could be best operationalized in case of future disasters (Government of Pakistan 2011b). The plan consolidates international best practice and Pakistan’s own experience, and has been approved by the government. Adequate capacity is lacking, however, and the FDRAP notes that the key to implementing effective cash transfers for early recovery is to ensure adequate capacities at all levels, from the federal government down to the districts. Although endorsed and published, the FDRAP has not yet been operationalized; in particular, the institutional capacity building is still pending. Policy Options Establish a national social protection framework Establishing a national social protection framework, agreed on between federal and provincial authorities, is essential for an effective and efficient social protection system. The framework would classify social protection Box Citizens’ Damage Compensation Programme 11.2 The government’s flood-response program after the devastating 2010 floods provided immediate cash assis- tance of PRs 20,000 to around 1.7 million families with further support of PRs 40,000 to more than 1.2 million households for early recovery. The cash transfers were designed to expand choice and empower citizens to prioritize their own needs for relief and recovery. With the federal and provincial governments, multidonor support provided the requisite funds for the second phase (PRs 40,000 per affected household) through improved targeting, third-party verification of targeting, beneficiary communication, and enhanced grievance redress. The program primarily targeted heads of households, with additional targeting of female- and disabled-headed families.1 An initial assessment showed that the largest uses of the cash transfers were food, repair of housing, medicines, and debt repayment. An independent impact assessment is under way and will report with more evidence later in 2013. 1. In the recovery phase, enhanced targeting resulted in around 18 percent of beneficiaries being female, a significant increase compared with the 8 percent female beneficiaries of the emergency phase. Source: Government of Pakistan 2010b, 2011a. consolidating social protection interventions to distinguish the different types of interventions, objectives, and target groups. It would need to the approval of the competent authori- ties (the Council of Common Interest or the Executive Committee of the 190 National Economic Council) when finalized. The Planning Commission can map federal, provincial, and regional interventions to aid discussions on the framework. Such a framework would help reduce fragmentation and program overlap. The following outline format could be considered, drawn from international practices (Table 11.1). Consultations among federal and provincial authorities are necessary to establish a national social protection framework owned by all concerned. These would help clarify institutional roles, possible sharing of financing responsibilities, and a complementary program of interventions between federal and provincial governments. Under the framework, the provin- cial and regional governments may consider establishing their own social Table Outline format for a national social protection framework 11.1 Intervention type Description Examples from Pakistana Social assistance Social safety net: Cash or in-kind transfers Benazir Income Support Programme, (addressing poverty or Unconditional cash transfers: Usually target Pakistan Bait-ul-Maal, Zakat, Food Support vulnerability) the poor Program Coresponsibility cash transfers: Promote Waseela-e-Taleem, Punjab Secondary School human development. Payments depend Stipend Program, Zakat educational vouchers on recipients fulfilling certain conditions, such as children attending school, mothers and children receiving primary health care services. Not necessarily poverty targeted Bait-ul-Maal, “Sweet Homes,� many of the Social care services: Social Welfare Department programs Target particular groups such as disabled people and orphans. Usually poverty Citizens’ Damage Compensation Program, targeted cash transfers for internally displaced Disaster relief: Cash or in-kind transfers persons Target those affected by disaster Untargeted subsidies on petroleum products, Subsidies: Reduce costs of basic goods or fertilizers, wheat, utility stores, and bread services. Most efficient when targeted, usually the poorest 20–40 percent Social security Such contributory social insurance programs Waseela-e-Sehat, Punjab pension fund, mitigate risk from old age, unemployment, Khyber Pakhtunkhwa pension fund, pensions and work injuries. Examples are old-age for federal and provincial government pensions and health insurance employees Employment promotion Support access to productive work such Waseela-e-Rozgar, Waseela-e-Haq, labor and labor market as skills training and apprenticeships. May welfare programs programsb have a specific target group, like youth. Not necessarily poverty targeted a. Not an exhaustive list. b. More details of labor market interventions, including promoting access to the labor market through skills training, apprenticeships, minimum labor standards, and decent work, are covered in a separate policy paper. Source: Authors’ compilation. protection authorities, to liaise with a federal social protection authority. There are good international examples of how different levels of govern- ment can work effectively in a coordinated manner under a coherent social protection framework (Box 11.3). 191 To encourage a dynamic, diversified, sustainable, and equitable social protection system, the pragmatic way forward would be to establish cofinancing arrangements through federal–provincial dialogue with a partnership approach. For example, the federal government’s contribution to the basic national safety net could be a fixed percent of GDP, and the provincial governments could top up the contribution. Flexibility in select- ing the program mix will enable provincial governments to adopt the programs best suited to their realities. Cofinancing will allow both levels of government to receive public acknowledgment for their efforts and will establish shared program ownership. Encourage wider use of the national poverty registry The national poverty registry could be used more often for targeting pro-poor social protection and disaster response, as well as programs in health, education, and other social sectors. The government needs to encourage the use of the national registry as the unified database for pro- poor programs and to ensure that household poverty data are freely available to provincial government departments and others. The Planning Commis- sion could facilitate the dialogue between federal and provincial author- ities on identifying the most appropriate institution to be the permanent Box Decentralized operations in Brazil 11.3 Brazil’s Bolsa Familia Program is a health and education coresponsibility cash transfer program. It is the largest program of its type in a developing country, implemented in a decentralized manner. The Bolsa Familia Program covers more than 11.2 million households, with 94 percent of the transfers reaching the two poorest quintiles. The program is managed at the federal level by the Ministry of Social Development and imple- mented by a number of different agencies at all levels of government. The 5,564 constitutionally independent municipalities play various roles in executing the program, including cofinancing, registering potential beneficiaries, monitoring education and health conditions, and prioritizing program beneficia- ries for other complementary services. In a highly fiscally decentralized state structure, a performance based management mechanism has been developed under the program to incentivize better quality program execution at the municipal level. Source: Lindert and others 2007. consolidating social protection custodian of the registry. The institution should be granted a mandate to manage development data. It should have the capacity to ensure the integrity of the database and to update it as necessary. And it should be tasked to help 192 others use the database. To strengthen the analytical underpinning of the registry and recertification of existing beneficiaries, the government needs to encourage the Federal Bureau of Statistics to include a social protection module in the Pakistan Social and Living Standards Measurement (PSLM) survey.8 Strengthen administration and monitoring of social protection programs Minimum standards need to be set for all social protection programs at the federal, provincial, and regional levels. Consensus needs to be built with provincial governments on program governance and on performance standards for all aspects of social protection based on good experience in Pakistan and elsewhere. Outcomes would encompass protocols for targeting, data management, beneficiary communication, grievance redress, and the like, using the experience of the national safety net program. All social protection programs should be monitored regularly and evaluated. Within a shared national social protection framework, programs can be evaluated individually or as coordinated operations for their impact on poverty, human development, economic growth, and public welfare. Federal and provincial governments should disseminate evaluation findings to improve practice and inform policy, allowing programs from one part of the country to be properly understood, results shared, and scaled up by others where appropriate. A coherent national CNIC-based management information system is one tool to improve coordination and governance of programs, including monitoring, tracking, and impact assessment. The national poverty reg- istry is linked to the National Database and Registration Authority’s (NADRA) CNIC database and provides a good mechanism for monitoring beneficiaries to ensure essential synergies and avoid duplication among var- ious programs. The Planning Commission and provincial planning and development departments should encourage all agencies to link social protection benefits to the beneficiary’s CNIC, and together they should maintain a national beneficiary database of social protection interven- tions. The programs, with support from NADRA, could use smart identity cards and a digital or e-platform to deliver social protection services or pay- ments. The system, linked to the CNIC poverty registry, could swiftly identify eligible beneficiaries and on biometric verification could activate the service or payment.9 Potentially, any ministry or department, federal or provincial, could use the cards or platform for targeting and delivering their services (with control and accountability mechanisms). Phase out universal subsidies to sustain financing of pro-poor social protection 193 Continuing to finance social protection programs is critical for sustaining the reform momentum and achieving the desired outcomes. Figure 11.6 attempts to estimate the expenditure required to meet even the minimum commitments of the planned poverty-targeted cash transfers under the BISP. (It presents estimates only and aims to give an idea of rough scale; it is based on the planned BISP rollout strategy, with estimates of administrative costs.) In addition to the costs implied by the full roll-out of the national safety net, the government needs to consider maintaining the purchasing power of benefits, which was seriously eroded by inflation over the last four years. Moreover, other programs would also need financing, requiring federal and provincial governments to commit fiscal resources to achieve the social protection objectives. Such a budgetary effort will be unsustainable if fiscal room is not created. Cutting untargeted subsidies would be one option to arrange the neces- sary financing. By 2011/12, untargeted subsidies had risen to nearly PRs  600 billion, far in excess of current expenditure and predicted requirements of social protection (see Figure  11.6). Benefit incidence analysis of these untargeted subsidies could inform policy makers about the impact on different income groups; consequent losses to the different groups if subsidies were cut; and options for redirecting expenditure to improve impact. Similar analytical work has been carried out recently in Morocco and Tunisia as the first step to Benazir Income Support Programme expenditure to date and future Figure anticipated annual expenditures, 2008/09–2015/16 11.6 Monthly cash transfer (to date) Monthly cash transfer (anticipated, increasing to cover 7 million children) Coresponsibility cash transfer (anticipated, increasing to cover 3 million children) 100 75 PRs billion 50 25 0 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 Source: BISP various years; authors’ calculations. consolidating social protection reducing untargeted subsidies. Subsidy reform was undertaken to finance social protection in Indonesia in 2005 (Box 11.4). 194 The government could also consider establishing and operationalizing the Income Support Fund as provided in the BISP Act 2010, to attract funds from local and international sources. This move would require rules and procedures to capitalize the fund, governance and accountability systems, identification of areas and definition of processes for investments from the fund, and definition of arrangements for monitoring and evaluating the results through the fund’s investments. Improve postdisaster early recovery support mechanism The FDRAP needs to be operationalized so that victims of future disasters receive rapid and appropriate cash transfers to support their early recovery. This would require financial and human resources from federal and provincial authorities. Provinces and regions should consider the FDRAP when preparing their disaster response action plans. All the key players—Cabinet Division, NADRA, National Disaster Management Authority, provincial disaster management authorities, and district authorities—need to maintain at least a skeleton capacity for postdisaster early recovery support to affected families in line with FDRAP. Notes This note benefited greatly from the inputs provided by Alison Barrett, Uzma ­ uanita Ali Khan, and Shirin Gul Sadozai. Basin, Q 1. The Pakistan Social and Living Standards Measurement Survey 2008–10 panel revealed that around 25 percent of households reported at least one idiosyncratic shock over the previous two years. Indonesia switches finance from fuel subsidies to social Box protection 11.4 In 2005, Indonesia removed a regressive universal fuel subsidy and, with the savings, financed new targeted social safety net programs. An unconditional cash transfer of $10 a month, much more than the increase in energy costs, reached more than 19 million households. The cash transfer cost 0.8 per- cent of GDP versus the fuel subsidy’s cost of 3.5 percent. Before the transfers were made, each household was given a proxy means test. Recipients were issued with smart cards, and transfers delivered through the post office system. By covering the poorest 40 percent of the population, the program also helped gain acquiescence to the fuel price rise from the less poor. Source: Handayani and Buckley 2009; World Bank forthcoming. 2. Based on the Poverty Reduction Strategy Paper classification of “Social Assistance and Social Security.� According to the Poverty Reduction Strategy Paper, the following programs fall under the “pro-poor� social protection spending: subsidies, social security and welfare, Benazir Income Support 195 Programme, Pakistan Bait–ul–Maal, Food Support Programme, natural calamities and other disasters, and low-cost housing. 3. The poverty scorecard is a measure of household assets and composition. The score is correlated directly using PSLM data on household consump- tion. The use of such scorecards is globally accepted best practice, known as a proxy means test. 4. There will be no gender-based priority for inclusion of girls into the program. The requirement for families to enroll the youngest three children is intended to ensure the inclusion of girls just like boys. 5. The 18th Amendment has induced structural changes that guide and rede- fine the architecture of governance in the country through the devolution to the provinces of all shared and overlapping responsibilities, particularly those of economic and social services. See Kardar (2013) for more detailed discussion of how this could relate to social protection. 6. For more detailed information, see Government of Pakistan (2007), World Bank (2013), Kardar (2013), and work being carried out by the Planning Commission to map current social protection interventions. 7. This total includes disaster relief spending after the 2010 and 2011 floods (0.4 percent of GDP in 2011/12) and BISP spending (0.25 percent of GDP). 8. The inclusion of a social protection module in the PSLM survey is essential for monitoring social protection program performance regularly, including capturing the contribution to the well-being of the poor. 9. The platform would support online processing capability and real-time reconciliation of transactions linking the central processing unit to all partner agencies (franchises) and commercial banks. For cash disbursements, the smart cards would operate dial-up or GPRS-based POS machines and ATMs across the country. References Aga Khan University, Pakistan Medical Research Council, and Minis- try of Health. 2011. “Pakistan National Nutrition Survey Report 2011.� Government of Pakistan, Islamabad. BISP (Benazir Income Support Programme). 2012a. “Progress Report, December.� Islamabad. ———. 2012b. “Progress Report, June.� Islamabad. ———. Various years. “Progress Reports.� Islamabad. consolidating social protection Government of Pakistan. 2007. “National Social Protection Strategy.� Planning Commission, Islamabad. ———. 2009. “Pakistan: Poverty Reduction Strategy Paper II.� Finance Division, 196 Islamabad. ———. 2010a. “Benazir Income Support Programme Act 2010.� Islamabad. ———. 2010b. “Memorandum of Understanding on Cash Transfers.� Signed between Cabinet Division and World Bank, Islamabad, November 26, 2010. ———. 2011a. “Memorandum of Understanding by and among the Islamic Republic of Pakistan and the International Development Association and the United States of America and Additional International Development Partners for the Joint Support to the Second Phase of the Islamic Republic of Pakistan’s Citizen’s Damage Compensation Program as of 9 June 2011.� Islamabad. ———. 2011b. Action Plan for Early Recovery in Future Disasters (through Cash Transfers). Cabinet Division. Islamabad. ———. 2011c. “FY2011/12 Mid-Year Progress Report.� PRSP Secretariat, Finance Division. Islamabad. ———. 2011d. “PRSP–II Period Progress Report FY 2008/09–FY 2010/11.� Ministry of Finance, Islamabad. ———. Various years. “PRSP Progress Reports.� Islamabad. Handayani, Sri Wening, and Clifford Buckley, eds. 2009. Social Assistance and Conditional Cash Transfers: The Proceedings of the Regional Workshop. Tunis: African Development Bank. ICF GHK. 2012. “Request for Expression of Interest: Financial Inclusion and Literacy Outcomes of Cash Transfers through the Banking System in Pakistan: Survey and Beneficiary Assessment.� Final Report. London. Kardar, S. 2013. “Social Protection and Constitutional Reform in Pakistan: Implications of the 18th Amendment.� World Bank, Washington, DC. Lindert, Kathy, Anja Linder, Jason Hobbs, and Bénédicte de la Briere. 2007. “The Nuts and Bolts of Brazil’s Bolsa Familia Program: Implementing Conditional Cash Transfers in a Decentralized Context.� Social Protection Discussion Paper 0759, World Bank, Washington, DC. Trimble, Chris, Nobuo Yoshida, and Mohammad Saqib. 2011. “Rethinking Electricity Tariffs and Subsidies in Pakistan� Report 62971-PK. World Bank, Washington, DC. World Bank. 2007. “Social Protection in Pakistan: Managing Household Risks and Vulnerability.� Washington, DC. ———. 2010. “2009 BISP Beneficiary Survey Report.� Washington, DC. ———. 2011a. “Women’s Decision Making and Human Development in Pakistan—Applications for BISP.� South Asia Human Development, Pakistan Social Protection Policy Notes. Washington, DC. ———. 2011b. “Empowering Women through BISP: The Effect of Women’s Decision Making Power on Reproductive Health Services Uptake in Pakistan.� South Asia Human Development, Pakistan Social Protection Policy Notes. Washington, DC. 197 ———. 2013. “Pakistan: Towards an Integrated National Safety Net System: Assisting Poor and Vulnerable Households: An Analysis of Pakistan’s Main Cash Transfer Program.� Report 66421-PK. Washington, DC. ———. Forthcoming. “Implementing Energy Subsidy Reforms: Evidence from Case Studies.� Washington, DC. 12 Managing Natural Haris Khan and Marc Forni Disasters Pakistan is prone to geological and hydrometeorological hazard events—earthquakes, floods, droughts, cyclones. Combined with rapid population growth, urbanization, environmental degradation, and a high concentration of people and assets in exposed areas, this hazard profile has raised population and infrastructure vulnerability, especially in urban areas. Losses from disasters have increased over the past 40 years, a trend that can be changed only through targeted interventions that boost resilience. Fortunately, the country is well endowed with capacity in government, academia, the private sector, and nongovernmental institutions. Greater resilience can be achieved through a holistic approach to disaster risk management: strengthening capacity and clarifying responsibilities of institutions engaged in disaster risk management at all levels of government; improving understand- ing of disaster risk among policy makers and disaster risk management practitioners, including the physical, human, and financial elements; investing in structural and nonstructural risk reduction measures; developing robust financial tools to reduce the fiscal impact of disasters on the state; and increasing awareness of and coping mechanisms for disaster risk among vulnerable populations. Disasters result from the intersection of three key elements. First are natural hazards, such as earthquakes, cyclones, excess rainfall, floods, and tsunamis. Second is exposure of people, property, livelihoods, and infrastructure (also known as human and physical capital) to these hazards. And third is vulnera- bility of the exposed elements, which is created by physical, social, economic, and environmental factors that increase a community’s susceptibility to the impacts of a natural hazard. The natural hazard element is difficult to change, but well-planned and timely investment in preparedness and mitigation can reduce vulnerability and build resilience. The purpose of engaging in disaster risk management (DRM) activities is to reduce the vulnerability of current and future assets exposed to natural hazards. Once vulnerability is carefully assessed, structural and nonstruc- tural mitigation activities can be launched. Structural measures include retrofitting buildings and constructing protective infrastructure. Nonstruc- tural measures include land-use policies to manage construction in vulnerable environments, contingency and response planning, and fiscal protection mechanisms. Pakistan is highly susceptible to natural hazards, with its location on a seismically 199 active geological plate, a coastline frequented by cyclone events, and an active floodplain fed by snowmelt from the Himalayas, Karakoram, and Hindukush mountains. Additionally, 60 percent of the landmass is classified as semi-arid and prone to drought. This hazard profile intersects with a growing population, rapid urbanization, and sociopolitical factors contributing to increased vulnera- bility. The result is that the country has, and will continue to have, frequent large disasters that halt or reverse development. Pakistan’s disaster record for the last 40 years shows severe impacts on citizens and the state. Major events include: • Floods in 2010 and 2011 that caused damage amounting to $10 billion (ADB and World Bank 2010) and $3.7 billion (ADB and World Bank 2011), respec- tively. Other significant floods were recorded in 1950, 1992, and 1998. • An earthquake in 2005 resulting in 73,000 fatalities and $5 billion in losses (Naeem and others 2005). Other damaging seismic events took place in 1974 and 1990. • Droughts in 2000 and 2002 that severely affected livelihoods and forced thousands to migrate. Drought is prevalent two or three years every decade. • Cyclone Yemyin in 2007 that caused $537 million in damage (IFRC 2007)— one of 14 cyclones to have reached Pakistan’s coastal zones in the past 40 years (ADB and World Bank 2007). The Calamity Act of 1958 historically governed government actions in DRM, but a series of disaster events highlighted the need for a holistic DRM frame- work, which was established in 2007 (NDMA 2007). Under this (and subse- quent laws), the National Disaster Management Authority and the National Disaster Management Commission were created at the federal level, and Provincial Disaster Management Authorities (PDMAs) and District Disaster Management Authorities (DDMAs) were created subnationally. The government has also drafted a national Climate Change Action Plan.1 Key impacts of climate change highlighted in the plan include glacial melt, which is causing the Indus river system to swell above historical averages, and rainfall events, which have been especially intense and irregular. The abnor- mal rainfall in 2011 over parts of Sindh Province, including the Thar Desert, was attributed by the Pakistan Meteorological Department to cloudburst, 2 characteristic of erratic rainfall patterns. These impacts highlight the need for further research and investment in mitigation and adaptation interventions. managing natural disasters Main Issues Weak institutional capacity for DRM 200 Lack of preparedness, poorly executed emergency responses, and weak insti- tutional systems to manage disasters have undermined government credibil- ity among its citizens. Pakistan has, though, begun to institutionalize and mainstream DRM activities, but much work remains. The PDMAs have only recently started receiving support through the government budget and international donor community. Continued support is required to enhance the capacity of the Provincial Emergency Operations Centers, to prepare standard operating procedures for response, and to institute policies to mainstream DRM. At the district level, DDMAs are functioning as ad hoc interdepartmental committees, activated only in times of disaster and with- out dedicated staffing or budgets. 3 The country’s major urban centers also lack an effective, dedicated DRM structure. Lack of coordination and clarity of roles in DRM The roles and relationships among federal and provincial entities—includ- ing the Floods Commission, Earthquake Reconstruction and Rehabilitation Authority, National Disaster Management Authority, and PDMAs—lack clar- ity. These bodies do not agree on which of them are ultimately responsible for understanding disaster risk, which are responsible for integrating DRM into development planning, and which are responsible for leading prepared- ness and response activities. Especially problematic is the devolution of responsibilities from the National Disaster Management Authority to the affected PDMA during a disaster. Global experience has proven that such an approach is inefficient due to a lack of national leadership and a lack of contribution from agencies not affected. Further, as nascent institutions, the PDMAs lack the capacity and capability to implement adequate DRM practices.4 The PDMAs might exist on paper, but are severely challenged by even medium-size events. Limited understanding of disaster risk (physical, economic, and fiscal) Policy makers and DRM practitioners lack the ability to define the occur- rence and impacts of disaster events, including floods, droughts, and earth- quakes. A comprehensive multihazard risk assessment tied to risk mitigation efforts has yet to be undertaken for Pakistan—a wide variety of actors has followed only a piecemeal approach, generating assessments with varying standards, implementation arrangements, and outcomes, as well as limited scope and geographic span.5 The costs and benefits of risk reduction—financial and nonfinancial—thus 201 cannot be qualified or quantified. This information would provide an evi- dence base on which to prioritize mitigation decisions, as well as data on how much of the country’s budget should be appropriated. The government bears much of the costs related to disasters, including explicit and implicit obligations, but the total fiscal impact is not well understood. Risk transfer mechanisms are used in other developing and industrialized countries 6 to provide a measure of protection for the state. Pakistan’s government has endorsed developing such mechanisms, particularly risk insurance and risk financing for catastrophes, though it has no roadmap, policy, or underpin- ning evidence to make an informed choice on them. Weak integration of DRM in government planning and development The government’s focus needs to shift from response to preparedness and risk reduction. This is partly because the costs of predisaster investments and mitigation measures are significantly less than the postdisaster costs of rebuilding. But these investments are not currently prioritized due to the difficulty in quantifying the benefits of such interventions. These benefits include the lives saved and damage averted, as well as the livelihoods pro- tected, the infrastructure left intact, and the services delivered in the after- math of an event. Poor communication of risk information and awareness building A wide range of actors in DRM convey messages on disaster risk. Lack of a single, coordinated message has led to confusion and misinformation over the risks faced and the measures to be taken to increase resilience and response capacity. Policy Options Increase institutional DRM capacity Continue to support the PDMAs and DDMAs to ensure that they can ful- fill their mandated roles and responsibilities. These institutions are very weak, particularly for their wide-ranging mandates. Each PDMA needs better equipment for its Emergency Operations Center, a management information system, and stronger technical capacity. managing natural disasters Create city emergency operations centers in major urban centers and link them to the current disaster management structure. With the exception of Rescue 1122 in Lahore, the capacity of municipal governments is quite limited due to 202 lack of communication systems, equipment, and technical capacity. Putting fully functioning emergency operations centers in place will improve the ability to respond to disaster events. Improve coordination and clarity of roles Review the roles and responsibilities of entities in a postdisaster environment. This should be completed at each level of government to remove overlaps and avoid confusion in times of disaster. Analyze the roles of governments, nongovernmental organizations, and private companies engaged in DRM. This can be accomplished through local and regional workshops heightening stakeholder awareness of the threat of natural disasters, as well as educational and training activities that increase the understanding of DRM among policy makers and DRM practitioners. Ensure that local governments, community-based organizations, and nongovernmental organizations use the same interactive tools and informa- tion to highlight disaster risks, thus enabling communities to improve their resilience. In addition, work toward a systematic and standardized approach to community-based DRM led by provincial and district authorities, and closely link governmental activities and investments to those at the community level. Enhance understanding of disaster risk (physical, economic, and fiscal) Undertake a national rapid risk assessment using existing data and tools. This preliminary information will enable the government to begin considering policies and procedures for reducing disaster risk. More compre- hensive risk assessments should follow once adequate data are gathered and stakeholders are incorporated. Ensure that systematic methods are in place for collecting disaster risk information in the future. Historical disaster-loss data can then be popu- lated into this database, which will ensure that financial, policy, and disaster management authorities can quantify Pakistan’s risk. Lead the risk assessment process. Pakistan is fortunate to have very high scientific capacity to assess natural hazard risks, but it is spread across government, academia, and the private sector. Harnessing it to produce risk assessments within and for the country will not only help ensure sustainable skills and knowledge but also greatly raise the chances of this information being used by decision makers.7 203 Ensure that decision makers from multiple sectors are engaged in risk assessments. Risk assessments need to be undertaken by scientists, spatial analysts, and engineers, but this requires a collaborative approach between technical experts and decision makers to ensure that the information gener- ated is fit for its purpose and provides answers to the questions posed, such as “How many hospitals may be affected by a 1 in 100-year flood event in Punjab?� Undertake national, provincial, and city risk assessments. These assessments should be aimed at planning for contingencies and responses, identifying and prioritizing risk mitigation interventions, ensuring that future development is disaster resilient, and guaranteeing adequate financial protection. Integrate DRM in planning and development Reduce current physical risk of disasters Undertake cost-benefit analyses of mitigation measures informed by robust assessments. Data from probabilistic risk assessments can be coupled with information on the expenditures required to reduce risk and on the net social and economic benefits at the local, provincial, and national levels. Engage in mitigation interventions informed by the National Working Group for Risk Assessments. Invest in structural and nonstructural measures to reduce the impacts of future hazard events. Examples include structural measures (flood control, building reinforcement) and nonstructural measures (land-use policies, early-warning systems). Reduce future physical risk to disasters Ensure physical resilience of new infrastructure. This is one of the most import- ant risk mitigation steps the government can take to prevent “locking in� future vulnerability. It is impractical to move people from their homes and livelihoods, but it is possible to start planning processes that ensure resilient construction in high-risk areas. Increase compliance with building codes through improved supervision of new construction. The public sector can provide economic incentives to owners of managing natural disasters existing structures to retrofit their buildings for the benefit of tenants and the public at large. 204 Increase fiscal resilience to disasters Understand the disaster-related annual expected loss and probable maximum loss to the government budget. These should be understood at the municipal, provincial, and federal levels. The analysis requires research on the fiscal impact of medium-size and large events of the past 15 years. Improve the National Disaster Management Fund. This needs to be greatly strengthened through standard operating procedures to administer and execute funds efficiently in times of disaster. Such a mechanism, replenished annually, is the most economically efficient source of ex ante risk financing and can be used for small to large events. Develop a comprehensive risk financing strategy with a range of financial instruments. Cost, size, and timeliness should be the three guiding prin- ciples for the instruments. Risks should be distributed among a variety of capital bases that, in aggregate, can adequately withstand the costs of natural disasters (Figure 12.1). Increase communication of risk information and build awareness Once identified, communicate disaster risks such that both individuals and the government are motivated to increase their resilience. Community-based DRM can help communities understand disaster risk and carry out small, physical resiliency measures. An informed community may decide against Figure Combining financial instruments to address various layers of risk 12.1 High severity International donor assistance Insurance-linked securities Risk transfer Insurance and reinsurance Contingent credit Risk retention Reserves Low severity Low frequency High frequency Source: Ghesquiere and Mahul 2010. building residential structures in vulnerable areas and instead use the land for agriculture. Incorporate credible and robust hazard and risk information into communi- 205 ty-based early-warning systems that advise communities of incoming threats. Such information can also be used by the government to enhance its institu- tional preparedness, emergency communications, and contingency planning. Notes 1. The Climate Change Action Plan 2012–2030 is being finalized by the Min- istry of Climate Change. 2. A cloudburst is an extreme amount of precipitation, sometimes with hail and thunder, which normally lasts no longer than a few minutes but is capable of creating flood conditions. Rainfall of 100 millimeters or more per hour is a cloudburst. 3. In 2012, the government of Khyber Pakhtunkhwa created full-time posi- tions in the districts for DRM, a practice that should be followed elsewhere. 4. The Federally Administered Tribal Areas Disaster Management Authority has only three staff members, responsible for DRM and issues related to displaced peoples, who respond to complex emergencies in the province. The Provincial Emergency Operations Centers across all provinces need strengthening. All PDMAs lack integrated Geographic Information System– based Management Information Systems and incidence response systems with links to other departments. 5. Donors have made risk assessments (often with external consultants) in the Northern Areas of Pakistan, Muzaffarabad, Mansehra, and Murree, using a range of techniques and implementation methods. 6. In the industrialized world, risk transfer mechanisms, such as insurance, cover around 30 percent of economic losses. 7. Government decision makers are more likely to trust information produced by locals within the country than that produced in a “black box� environ- ment with external consultants. References ADB (Asian Development Bank) and World Bank. 2007. “Pakistan Cyclone and Floods 2007: Preliminary Damage and Needs Assessment: Balochistan and Sindh.� Islamabad. ———. 2010. “Pakistan Floods 2010: Preliminary Damage and Needs Assessment.� Islamabad. http://gfdrr.org/gfdrr/sites/gfdrr.org/files/ publication/Pakistan_DNA.pdf. managing natural disasters ———. 2011. “Pakistan Floods 2011: Preliminary Damage and Needs Assessment.� Islamabad. Ghesquiere, Francis, and Oliver Mahul 2010. “Financial Protection of the State 206 against Natural Disasters: A Primer.� Policy Research Working Paper 5429. World Bank, Washington, DC. IFRC (International Federation of Red Cross). 2007. “Pakistan: Cyclone Yem- yin/Floods.� Geneva. www.ifrc.org/docs/appeals/07/MDRPK001OU18.pdf Naeem, Akhtar. Qaisar Ali, Muhammad Javed, Zakir Hussain, Amjad Naseer, Syed Muhammad Ali, Irshad Ahmed, and Muhammad Ashraf. 2005. “Pakistan: A Summary Report on Muzaffarabad Earthquake.� Earthquake Engineering Center at the Department of Civil Engineering, University of Engineering and Technology, Peshawar, Pakistan. NDMA (National Disaster Management Authority). 2007. National Disaster Risk Management Framework. Islamabad. http://ndma.gov.pk/Docs/NDRMFP.doc. Naseer Ahmad Rana, Iffat Idris, and Irum Touqeer Revamping 13 Governance Pakistan faces serious governance challenges that affect the state’s ability to deliver services to its citizens. These challenges are a combination of weaknesses in the state’s effectiveness, efficiency, and accountability. Effectiveness is hampered by the patron- age-prone political system, frail administrative capacity, weak provincial and local governments, and shortcomings in the justice system. Efficiency is constrained by the state’s inability to mobilize revenue and manage resources well. And accountability is undermined by fragmented and feeble governance institutions and lack of transparency, allowing for corruption and manipulation of rules. Hence a broad and long agenda of reform is needed, focusing on improving management of service delivery, bringing in a broad, simple, and equitable tax system, improving budget management and public procurement, and ensuring right to information regimes at the federal and provincial levels. Further, given that corruption is often institutionalized, Pakistan needs to prepare a new broad-based anticorruption strategy and ensure that it is implemented well. The findings from numerous international and local studies point to serious failings in governance and to institutionalized corruption in Pakistan. Consider Legatum Institute’s in Prosperity Index, which defines the concept of national prosperity not just in terms of wealth but also well-being, covering some 90 indicators to benchmark 142 countries.1 On this index, Pakistan ranks 132 of 142 countries (Indonesia 63, India 101, Bangladesh 102, Cambodia 107, and Nepal 108). Among the eight categories of the index, Pakistan ranks 121 in governance (Legatum Institute 2012). Similarly, Global Integrity’s Global Integrity Report assigned Pakistan an over- all rating of 68 out of 100 in 2010, due mainly to weak implementation of laws (Global Integrity 2010). The Bertelsmann Stiftung’s Transformation Index ranks Pakistan 110 of 128 countries due to weak democracy, a poorly functioning market economy, and weak political management (Bertelsmann Stiftung 2012). The World Bank’s World Governance Indicators 2011 also ranks Paki- stan very low on quality of governance –1.14 (average), lower than the aver- age rankings of South Asia (–0.69) and Sub-Saharan Africa (–0.64; World Bank 2013). R E v a m p i n g g o v er n a n c e These findings are echoed by the governance indicators of the Global Competi- tiveness Report 2012–2013, which looks at 111 indicators in 144 countries across the world. On this, Pakistan ranks 109 in transparency of government, 116 208 in property rights, 119 in irregular payments and bribes, 127 in reliability of police, 129 in favoritism in decisions of public officials, 136 in organized crime, and 143 in business costs of terrorism (only Colombia is worse). Perhaps because of this last score, Pakistan ranks 132 in the business costs of crime. The most problematic factors for doing business were corruption, inefficient government bureaucracy, policy instability, and government instability and coups (World Economic Forum 2012). In a recent analysis of the binding constraints to Pakistan’s economy, bad governance and a poor civil service appear to be undermining economic growth. Without improving governance, other efforts in realizing the country’s growth potential are destined to be less effective than they would be otherwise (Lopez- Calix and Touqeer forthcoming). The government also shares these findings in its recent Framework for Economic Growth (Government of Pakistan 2011). And Pakistanis agree. A new large-scale survey asked Pakistanis, “What is the most important obstacle to economic progress?� The most prevalent answer was cor- ruption (The Herald and SDPI 2013). Running second and third were incompetent leadership and poor governance. Young entrepreneurs in Pakistan consider cor- ruption a major constraint to industrial and economic development� (ICCI 2012). While it is difficult to give precise figures for the level of corruption in public sectors, available data show that perceptions of corruption have increased. In 2012, Pakistan ranked 139 of 183 countries on this indicator, compared with 79 of 91 countries in 2001 (Transparency International 2001, 2012a). About 42 percent of Pakistani businessmen reported losing new business because a competitor paid a bribe to win the contract. They see corruption as a widely accepted fact of life, because prosecution for such crimes is lacking (Transpar- ency International 2011). Combined, these data and analyses paint a bleak picture of governance in Pakistan. But it is not all bad news. The Global Competitiveness Report 2012–2013 found that on some measures like judicial independence and strength of inves- tor protection Pakistan actually performs much better than some far more developed countries, like Australia, Brazil, Finland, the Republic of Korea, Spain, and Turkey (World Economic Forum 2012). The challenge is reversing the governance slippage and enabling Pakistan to realize its full potential. Main Issues A useful way to categorize and understand the governance challenges is through their impact on effectiveness, efficiency, and accountability of service delivery. 209 Effectiveness refers to performance: the ability of the state to deliver services and meet the needs of citizens. Efficiency refers to the manner in which the state func- tions, specifically to its ability to mobilize revenue and its ability to use that money in a way that minimizes waste and corruption—and that maximizes results. And accountability refers to the degree of openness and transparency with which government operates, and the ability of citizens to hold those in public office accountable. The factors that hamper these three elements are now discussed. What hampers state effectiveness? Four issues stand out. A patronage-prone political system Pakistan’s political life is characterized by the predominance of “patronage� politics over “issues-based� politics. While the latter differentiates parties and their support largely along policy lines (for example, on the economy, defense, and international relations), the former is characterized by client–patron relation- ships—politicians are elected based on pledges to specific groups. Once in office they have to deliver on those pledges, meaning that, rather than working for the interests of the population as a whole, they disproportionately favor the groups that supported them. At the same time, politicians must also “recoup� their own investment in campaigning, buying support, and distributing largesse. They do so through corruption, kickbacks, and rent seeking. This system makes it very hard for politicians to focus on policy issues, economic growth, and development, or even to have a long-term vision for the country­ —their focus is on short-term, personal gains. Weak administrative capacity An important determinant of weak administrative capacity is the quality of people in the civil service, yet Pakistan has seen a steep decline in the ser- vice’s ability to attract and retain qualified, talented personnel. Part of this relates to low incentives and generally opaque selection and promotion pro- cesses, which favor less qualified people with connections over trained, edu- cated, and hardworking men and women.2 The drop in overall service quality is reflected in rising inefficiency, poor performance, and corruption (Husain 2009; NCGR 2008). R E v a m p i n g g o v er n a n c e Weak provincial and local governments The 18th Constitutional Amendment transferred most state functions and 210 responsibilities from the federal to provincial governments. But the provincial governments do not seem to have complete control over the functions assigned to them for several reasons. First, they lack capacity to exercise all the functions assigned to them: some require skills that may not currently exist at the provin- cial level. Second, there is confusion over accountability roles. Although the 18th Amendment has brought greater clarity in government accountability (public service accountability now starts and ends with provincial governments, which is a strong incentive for them to improve the quality of services), a large number of functions of three main federal ministries—agriculture, health, and educa- tion—are retained by the federal government (Mukhtar 2013). After elections in 2008, the provincial governments embarked on rolling back 2001’s devolution, reassuming many of the functions and powers previously delegated to local governments. Despite pledges, new local government systems have still not been fully defined or established, resulting in overlapping responsibilities between provincial and local governments and diminishing ability of local governments to deliver services. The incomplete administra- tive decentralization has created dual accountabilities of local government officials (to both provincial and local governments) that undermine incentives to perform (Husain 2009). Conversely, there are asymmetric responsibilities between provincial and district governments. After establish- ment of the local government system in 2001, health and education employees continued to function as provincial employees, with the district nazim having little authority over the hiring, firing, evaluation, or placement of these employees. Similarly, the salary budget was transferred to the districts but nonsalary recurrent and development budgets were inadequately transferred, which affected the ability of district governments to improve service delivery (Mukhtar 2013). Shortcomings in the justice sector As part of good governance, the police and the courts are the two principal entities charged with upholding the rule of law. The police, respon- sible for investigating crimes and protecting citizens, is widely seen as cor- rupt, oppressive, and manipulative—an agency to be feared and avoided rather than sought out for help and protection (Figure 13.1). The failings of the police can, in turn, be attributed to lack of resources, training, and proper systems, as well as a tendency of governments and politicians to use the police to further their personal interests. Perception of corruption in Pakistan, India, and Indonesia, Figure 2010/11 13.1 Pakistan India Indonesia Police Public of cials 211 Political parties Parliament Businesses Nongovernmental organizations Judiciary institutions Media Education institutions Military Religious bodies 0 1 2 3 4 5 Global Corruption Barometer Note: 1 is not at all corrupt, 5 extremely corrupt. Source: Transparency International 2012b. The courts, too, lack resources, trained staff, and facilities. Weaknesses in the court system are manifested in huge caseloads and backlogs with long delays in hearing cases. It is common for property disputes, for example, to drag on for years and even decades. The problems in the court system create space for corruption—those with connections and money can influence judgments. For the ordinary citizen, the time, effort, and expense involved effectively mean that “access to justice� does not exist, whereas for firms it affects their invest- ment decisions. The Investment Climate Assessment finds that more than a third of Pakistani firms perceive the low quality of courts as an obstacle, compared with about a fifth of firms in comparator countries (such as Bangla- desh and Chile; Figure 13.2, left panel). The World Bank’s Enterprise Survey finds courts system as a major for about 38 percent of the firms in Pakistan (Figure 13.2, right panel). Figure Quality of the courts Islamabad FaisalabadGujranwala Lahore Sheikhupura Sialkot Hyderabad Karachi Sukkur Multan Rawalpindi Quetta Peshawar 13.2 Firms’ perceptions of quality Share of rms identifying the of the courts courts system as a major constraint Pakistan Comparator countries 60 Pakistan Percent of rms 40 All countries 20 South Asia 0 Quick Affordable Impartial Able to 0 10 20 30 40 enforce Percent decisions Source: World Bank 2007, 2009. R E v a m p i n g g o v er n a n c e What hampers state efficiency? Revenue mobilization is one of the major challenges. Pakistan has declining 212 tax revenues and one of the world’s lowest tax to GDP ratios, stemming from five main tax policy and administration weaknesses: complexity, a narrow tax base, low compliance, inefficient tax administration, and low and declin- ing provincial tax revenues. Complexity provides scope for discretion and corruption. A narrow tax base and low compliance are outcomes of inequitable exemptions and preferential treatments, low tax registration, and massive tax evasion by potential taxpayers who prefer to stay informal. Poor management, weak human resources, and a lack of effective key supporting information technology (IT) systems are constraining the tax administration. These issues are also pronounced among provinces, as provincial tax collection efforts fall short of their expenditure responsibilities under the 18th Amendment, affect- ing their ability to deliver services (Lopez-Calix and Touqeer 2013). Weak public financial management also hurts Pakistan, as available resources are not used as efficiently or effectively as they should be. There are underlying problems with planning and budgeting systems and procurement, all of which provide scope for corruption through weak monitoring and accountability.3 What hampers state accountability? Two main issues are apparent. Fragmented and weak institutions of accountability Pakistan possesses an extensive institutional setup for accountability (Box 13.1). However, the functioning of many of these institutions is hindered by the following constraints. • Lack of financial and functional autonomy . Direct government control or involvement in the day-to-day affairs of accountability institutions affects their autonomy. The National Accountability Bureau faces political inter- ference, which affects its authority to conduct unbiased investigations.4 The financial dependence of these institutions on other departments or ministries is another constraint. For example, expenditure by the Federal Services Tribunal depends on approvals from the Ministry of Law. The authority of some accountability institutions is limited, restricting imple- mentation of rules and regulations. For instance, the ability of the Auditor General of Pakistan to conduct the full range of audits of all government entities is restrained due to resistance from government entities. Similarly, provincial anticorruption establishments lack the authority to initiate Box The accountability framework 13.1 The accountability framework consists of different institutions that cater to the supply and demand sides of accountability and support different accountability mechanisms. 213 On the supply side are public institutions established under the constitution, laws, regulations, cabinet resolutions, notifications, and orders of the government. The main public institutions in Pakistan are the Public Accounts Committee, Judicial Commission, Election Commission of Pakistan, Ombudsman institutions, Federal Investigation Agency, National Accountability Bureau, provincial anticorruption establishments, Auditor General of Pakistan, Procurement Regulatory Agencies, and institutions of market governance (State Bank of Pakistan, Competition Commission of Pakistan, and Securities and Exchange Commission of Pakistan). Demand for accountability comes from the media, nongovernmental institutions, professional orga- nizations, and private sector bodies (such as Transparency International). Freedom of the press, prolifer- ation of electronic media, and increasing assertiveness and activism of the judiciary in recent years show strengthening demand and some supply response of accountability (enforcement) in Pakistan. Stories and investigative reports compiled by the media have been picked up in many cases by the judiciary. Suo moto cases have been initiated based on these reports against those alleged to have been corrupt or to have misused their public offices. The Annual Corruption Perception Index has also stirred up action on the part of the Public Accounts Committee and the executive branch. Source: Husain, Rana, and Touqeer forthcoming. cases. Moreover, some legal provisions, such as the provision for appeal against the decisions of the Federal Ombudsman to the President, and provisions for term appointment of senior and management positions (as in the case of the Competition Commission of Pakistan) provide scope for political capture.5 • Some overlapping and duplication of mandates, leading to turf issues . This is seen, for example, in the reversal of the earlier decision to merge the Economic Crimes Wing of the Federal Investigation Agency with the National Accountability Bureau to avoid overlap. This reversal has strained the relation- ship between the two institutions and weakened the process of holding those in public office accountable. • Low levels of implementation or enforcement. This may be explained by outdated laws, opaque rules, and cumbersome procedures. Some of the laws need to be updated and procedures streamlined. The general public (or even senior officers) are unaware of these rules as they are not widely disseminated, which works to the advantage of those who store this knowledge and interpret the rules (generally, patwari and thanedar 6) and provides scope for corruption. All of this has conferred enormous discretionary powers on officials. • Weak human and physical capacity and internal accountability mechanisms. The relevant institutions generally fail to carry out regular performance moni- toring, reporting, and auditing. In most cases, there are no set performance indicators and related data to assess institutional performance against those indicators. Further, the institutions lack physical (especially IT) infrastructure and human capacity. R E v a m p i n g g o v er n a n c e Lack of transparency The lack of transparency restricts external accountability—for example, in the 214 budget process. Transparency in the budget process serves as a mechanism for external accountability to ensure sound budget decisions—geared to deliver- ing services and growth—and proper implementation of them. Yet Pakistan scores just 58 on the Open Budget Index 2012, the same as Bangladesh and Columbia, but lower than Indonesia (62), India (68), and Brazil (73), pointing to a lack of fiscal transparency (International Budget Partnership 2012). Over the past two decades, the number of countries with right to informa- tion (RTI) or freedom of information (FOI) laws has expanded.7 Pakistan was the first country in South Asia to pass such legislation, in the form of the FOI Ordinance 2002. However, its introduction by a military ruler (President Musharraf) undermined its legitimacy and credibility. Various other factors have also impeded its effectiveness, including shortcomings in the law itself and bureaucratic resistance to RTI. As a result, public bodies have poorly imple- mented FOI laws, such as requirements for proactive disclosure, designation of FOI officers, awareness raising among the public, and so on. At the provincial level, Balochistan and Sindh have “copycat� FOI legislation to the federal law, which suffers from the same shortcomings and has been equally ineffective, while Punjab seems closer to having one than the Khyber Pakhtunkhwa. The demand for RTI has been equally weak due to insufficient support from civil society and lack of public awareness of RTI and its relevance to their needs. As a result, the total number of FOI requests in Pakistan is only a few hundred. Policy Options The governance reform agenda for Pakistan is substantial. The country needs to further strengthen its democratic governance by strengthening the Election Commission of Pakistan to help carry out its mandate, introducing more trans- parency in political-party functioning, and building capacity of new legislators. Pakistan also needs to enhance revenue mobilization and address key public financial management issues, nationally and subnationally, to enhance state efficiency.8 Strengthening the rule of law through legislative reforms that focus on police service and the justice system is also important.9 Other areas need to be addressed to improve service delivery, but everything cannot be done at once. So where should Pakistan start? Based on available opportunities and drawing on lessons from international experience, priority areas are identified in which reforms are feasible and results are achievable, thus strengthening Pakistan’s overall governance.10 Improve management of service delivery Four key reforms can improve the delivery of basic services. 215 Strengthen provincial administration To properly execute the assigned functions and responsibilities, subnational governments should have complete control over their functions. This can be achieved with the help of the following: • Adequate staffing, training, and technical resources, which would entail, for example, capacity building at the provincial level for public financial management functions. • Clearly defined accountability roles with complete control of subnational governments over their staff (and less or no interference from federal government). • Clearly defined roles and functions of each tier of government to avoid over- lapping functions. • Adequate financial resources based on sufficient subnational tax powers to raise budgetary flexibility, with fiscal transfers from higher levels of govern- ment mainly for fiscal equalization and national and provincial objectives. Establish a local government system It is important to define and establish a viable and effective system of local government in each province. While the ground realities will determine the extent and form of devolution to the third tier, to be effective, local governments should be given significant service delivery functions, commensurate finances, and a fair degree of autonomy. The decentralized system should also allow for citizen voice and participation in local planning and development and should build subnational and local capacity to ensure effective implementation. The passage of local government legislation by all provinces (based on consulta- tion with all stakeholders) will be an immediate measure for this. Again, the roles and functions of each government should be further defined with clear account- ability roles. Local governments should have complete financial autonomy for making and executing their budgets according to their priorities, irrespective of whether the funds are transferred by provincial government or raised by local government. At the same time, the provincial government should be responsible for building fiscal and financial management capacity at the local level. Further, to strengthen the financial autonomy of local government, some buoyant reve- nue bases should be considered for devolving to larger local governments (for example, urban immovable property tax, calling tax, and entertainment tax).11 R E v a m p i n g g o v er n a n c e Promote information and communications technology The third reform is to build on the innovative uses of information and com- 216 munications technology (ICT) being pioneered in Punjab to improve gover- nance. The rapid spread of mobile phone use and even Internet access across Pakistan, and the declining costs of this technology, means that ICT can be an extremely rapid and cost-effective mechanism to bring about tangible improve- ments in basic services for citizens (Box 13.2). Implement a multipronged strategy for providing citizens’ right to services Following other countries, Pakistan could establish citizens’ right to public services in law. Such legislation would raise awareness among citizens of their entitlements and—more important—put pressure on service provid- ers to deliver. Karnataka’s Right to Service Act 2011 guarantees time-bound delivery of public services to citizens and provides mechanisms to punish the public servants deficient in providing those services. This has greatly helped control corruption among government officials and increase transparency and accountability. But such legislation on its own will not work. What is needed is a transforma- tion of the entire service delivery chain for effectiveness. So Pakistan should implement a multipronged strategy, focusing not only on improving front- line access to service delivery but also on developing a monitoring mecha- nism, standardizing and simplifying back-end processes, addressing capacity constraints, and generating awareness among citizens. The innovative use of ICT-based performance management and grievance redressal systems will Advantages of information and communications technology in Box government 13.2 Information and communications technology (ICT) can be used for gathering data and mapping services and assets—all of which can inform better decision making, planning, and resource allocation. It can also be used to maximize operational efficiency, track the performance of service delivery providers, and enhance accountability—promoting better performance generally. ICT can be employed to get feedback from citizens on the quality of services they obtained and any problems they encountered­ —this too can act as a check on corrupt or poorly performing officials and identify areas where service provision needs to be improved. It can also be used to enhance transparency, such as in public procurement as piloted in Gujranwala. Finally, ICT can be used to promote e-services: enabling citizens to obtain application forms, submit applications for licenses and permits, make payments, obtain information, and so on. E-government, once established, reduces transaction costs to both service provider and citizen, and—critically—reduces opportunities for rent seeking and corruption. be beneficial. ICT-based management improves performance and service delivery with a focus on quality, productivity, and evidence-based planning. Enhance transparency 217 There is a need to prioritize RTI reform to promote transparency. There is already momentum for this in Pakistan, particularly with the approval of a strong Punjab FOI Bill by the last Punjab cabinet on March 12, 2013.12 The new federal and provincial governments should build on this momentum to establish effective RTI regimes. This will require measures on the supply side (reforming or passing RTI legislation, and properly implementing and enforcing such legislation) and on the demand side (raising awareness, making RTI relevant to the needs of ordinary citizens, and helping them use it; Box 13.3). Four key reforms are put forward. Reform or pass FOI and RTI legislation The federal FOI Ordinance 2002, Balochistan FOI Act 2005, and Sindh FOI Act 2006 need to be reformed to bring them into line with accepted best practice on RTI legislation. The Punjab FOI Act should be passed into law by the newly elected Punjab Assembly. The government of Khyber Pakhtunkhwa should introduce RTI legislation based on accepted best practice. Promote FOI and RTI implementation by public bodies Implementation of FOI and RTI legislation requires several measures. They include raising awareness among public officials, appointing and training designated information officers, developing systems and procedures to handle FOI and RTI requests, ensuring that public bodies disclose information, improv- ing records management (ideally, automating records management systems), and engaging citizens to help them access information. All of this requires heavy resource allocations as well as time and effort. Strengthen FOI and RTI enforcement Enforcement of FOI and RTI legislation is critical. Without strong enforce- ment mechanisms, it will be all too easy for public officials to deny information to citizens—a particular danger in Pakistan with its bureau- cratic culture of secrecy. Current FOI legislation in Pakistan designates the Ombudsman as responsible for hearing appeals and enforcing FOI provisions. The Ombudsman offices need to be supported to carry out this role. International best practice recommends a dedicated independent R E v a m p i n g g o v er n a n c e International examples of citizens using freedom of information Box and right to information reforms to secure rights and services 13.3 Improving service delivery in Mexico. In 2006, Maderas del Pueblo del Sureste—a nongovernmental 218 environmental organization supporting indigenous people and rural communities in Chiapas, Mexico— filed access to information requests using the federal transparency law, seeking information about a sewage project hurting a village that was receiving waste from a neighboring town and had no access to clean water. Information released through these requests showed that the water treatment system was not properly designed and needed a filter system that had not been installed. The Cintalapa sewage project was halted, and authorities publicly acknowledged that changes had to be made to ensure water was properly treated. Enhancing service delivery in South Africa. Villagers in Emkhandlwini had no water, whereas neighboring villages were receiving water deliveries from municipal tankers. With the help of a civil society organiza- tion, the villagers filed a right to information (RTI) request for minutes from the council meetings at which water programs had been discussed and agreed on for the council’s integrated development plan and the associated budget. This information showed that there were plans to deliver water throughout the region, but that somehow Emkhandlwini had been left out. Armed with this information, the villagers were able to assert their claims for water. Achieving rights in India. A 2008 report by the Indian nongovernmental organization, Participatory Research in Asia, highlighted a case where police refused to accept a “first information report� from an individual who had been assaulted in a land dispute, even after the individual had produced a medical certificate attesting to his injuries. When he returned a few days later with an RTI request asking for information about why his first information report had not been accepted, the police immediately registered the assault claim. Achieving rights in Thailand. In early 1998, shortly after the Thai RTI law was adopted, a parent, Sumalee Limpa-owart, used it to fight corruption in the education system. Her daughter had been refused entry to a prestigious state-funded school. Admission was supposed to be based on a competi- tive entrance examination. However, the student body was largely composed of children from elite families, leading to a widely held perception that some form of bribery was involved. Sumalee sent a letter to the school requesting the marks and answer sheets of her daughter and the 120 students who were admitted. When she received no reply, she filed a petition under the Official Information Act and eventually appealed the case to the Official Information Commission, which ordered disclosure of the information. Sumalee was eventually given access to the answer sheets and marks in March 1999. By that time, the school had already admitted to corrupt practices in processing admissions of 38 students. Source: Dokeniya 2013; Mendel 2011. information commission to deal with RTI appeals (and to support oversight and improvements in RTI implementation). Such a body should have the required specialist skills to handle RTI cases, as well as time and resources to focus on RTI (something the already overburdened Ombudsman offices cannot do). If independent information commissions are set up in Pakistan (through reform of federal, Balochistan, and Sindh laws and passage of the Punjab FOI Act, which already provides for such a commission), adequate resources, staff, and infrastructure need to be allocated to them. Strengthen demand for RTI 219 Supply-side measures alone will not bring about strong RTI regimes. These measures have to be complemented by strong demand for RTI, which is very weak in Pakistan. Raising awareness among civil society groups, the media, and the general public is needed to change this, with special focus on high- lighting how FOI and RTI legislation can empower ordinary citizens and help them secure their rights. Strengthen accountability Prioritize reforms of accountability institutions A selective approach is recommended. It should start with a few priority institu- tions—that is, those that not only have important individual roles but also have accountability links with other institutions, citizens, and local communities that, if functioning well, can affect service delivery and outcomes. Possibilities include the Auditor General of Pakistan, Public Procurement Agencies, Ombudsman offices, the three institutions of market governance, the National Accountabil- ity Bureau, and subnational anticorruption establishments. In the medium to long term, attention should be given to the Election Commission of Pakistan, Federal Public Service Commission, and some important sectoral authorities and commissions, such as the Oil and Gas Regulatory Authority and the National Electric Power Regulatory Authority. Strengthen autonomy, authority, and accountability of institutions The functioning and financing of accountability institutions should be freed from excessive government control or interference. For example, the Auditor General of Pakistan should be strengthened to extend its jurisdictions to all government controlled or owned entities. Institutions like the Securities and Exchange Commission of Pakistan, the Ombudsman, and the Competi- tion Commission of Pakistan should be empowered for decision making and implementation. Implement an appropriate system for the selection and appointment of institution heads Senior management positions should be filled by qualified professionals, and security of tenure should be ensured. One option could be for the selection R E v a m p i n g g o v er n a n c e process of heads of institutions such as the Auditor General of Pakistan, the National Accountability Bureau, and the Ombudsman to follow the recent example of the appointment of the caretaker Prime Minister of Pakistan in 220 March 2013, which entailed consultative decision making with a defined time- line and courses of action in the event of a decision not being reached. Build up human and physical capacity to improve the functioning of institutions The accountability institutions should hire staff professionally and trans- parently. There should be structured capacity-building programs, including focused training and learning opportunities from international experience, customized to the needs of each institution. The institutions should also invest in physical infrastructure, especially business process automation (Box 13.4). International examples of strengthened accountability Box institutions 13.4 Hong Kong SAR, China One of the top anticorruption institutions is the Independent Commission Against Corruption (ICAC) of Hong Kong SAR, China. Backed by the strict Prevention of Bribery Ordinance, the ICAC was established as an independent multidisciplinary body with the mandates of pursuing the corrupt through effective detection and investigation; eliminating opportunities for corruption by introducing corruption-resistant practices; and educating the public on the harms of corruption and fostering their support in fighting corruption. The ICAC reports directly to the head of the government. It has strong legal powers. And it employs professional staff to fulfill its mandate. Further, it provides a favorable working environment, including appropriate salaries for civil servants, a high degree of professionalism and pride among officials, and an environment in which people caught offering bribes or manipulating the system find it hard to obtain business. Thus the commission makes corruption a high-risk activity. Based on its multidisciplinary institutional setup, the ICAC has been pivotal to the success of transpar- ent public procurement and effective completion and implementation of public projects in Hong Kong SAR, China. One of the most noted ICAC projects was the construction of the new airport. The Airport Core Programme was designed involving substantial reclamation of land and construction of the airport itself, associated bridges and railway systems, a cross-harbor tunnel, expressways, and a new town. The ICAC adopted a proactive approach to preventing corruption in this mega-size public development project. ICAC involvement started early, at the legislation stage, to ensure that corruption prevention safeguards were incorporated in the systems. And ICAC staff maintained close liaison with the senior management of the implementing agencies to provide advice during the procedure-formulation stages as well as during project implementation. Malaysia The Malaysian Anti-Corruption Act 1997 introduces preventive measures against corruption in the civil service and statutory boards and authorizes investigation of complaints meriting disciplinary actions against civil servants. The chief public prosecutor, the Attorney General, is the final decision-making authority. The (continued) International examples of strengthened accountability Box institutions (continued) 13.4 Attorney General and the Director General of Anti-Corruption Agencies are appointed by the King on the advice of the Prime Minister. A parliamentary select committee represented by all major political parties 221 maintains a check on anticorruption agencies. The Attorney General enjoys certain tenure of service, and removal requires following a prescribed and open procedure. The Public Accounts Committee of Malaysia ensures that accounts of federation and states be examined by the Auditor General. The National Integrity Plan and Integrity Institute of Malaysia strives to stamp out corruption and patronize good governance and ethical values. The Malaysian government has established a 17-member Royal Commission to enhance the operations and management of the Royal Malaysia Police. The Public Complaints Bureau lodges complaints against government officials to hold them accountable. All this institutional setup is supported by a courageous Malaysian media that pinpoints corrupt practices. The Securities Commission in Malaysia is a statutory body established under the Securities Commission Act 1993 to regulate and develop the Malaysian capital market. The commission’s functions, powers, and various governance arrangements are set out in the Act. The Minister of Finance appoints commission members. The commission members comprise an Executive Chairman together with seven other members, of whom four represent the government and the remaining three represent the private sector. Source: World Bank forthcoming; ICAC 2013; OECD 2008; Abdullah 2008. Prepare a broad-based anticorruption strategy Pakistan should consider preparing a new broad-based and pragmatic anticorruption strategy led by the government and supported by other polit- ical parties, civil society, and the international community (Box  13.5).13 The main steps in this new anticorruption approach include undertaking a thorough diagnosis of corrupt systems, devising carefully deliberated Box A new approach to fighting corruption 13.5 Pakistan should consider preparing a new broad-based anticorruption strategy. The usual approaches to fighting corruption are insufficient, though they pass new laws, dictate codes of conduct, train public administrators, and build capacity. These approaches do not, however, guarantee the implementation of laws, which is Pakistan’s dilemma. The new approach should be based on right leadership from govern- ment, support from civil society, and the right help from the international community. It should recog- nize that: • Corruption is a problem of political cultures as well as bad laws and poor policies. • Corruption involves informal systems that work parallel to the ostensibly legal systems. • Reformers have to build credibility and momentum by “frying big fish� and achieving some quick, highly visible successes. • The business community and civil society must participate in diagnosing and healing corrupt systems. (continued) R E v a m p i n g g o v er n a n c e Box A new approach to fighting corruption (continued) 13.5 The main steps include undertaking a thorough diagnosis, devising carefully deliberated strategies, ensuring proper implementation of the strategies, and reaching out to the international community to 222 make the progress visible and widely appreciated. Diagnosis. This should use a variety of country-level indicators of governance and competitiveness to analyze corruption, see where the country stands against other nations, and estimate the impact of governance improvements on investment, job creation, and growth. The diagnosis should cover the pro- filing of anticorruption apparatus and corrupt systems (not people) and consider citizens’ views and experience at all levels (local, provincial, and national). The diagnosis will raise interest if it also iden- tifies “big leaks� through a frank assessment of the sources, amounts, and destinations of outflows and the chances of getting them back. Strategy for transformation. Using findings of the diagnosis, the next step is to convene leaders to devise practical ideas (including local solutions) for fighting corruption. This will require an understand- ing of the economics and weaknesses of corrupt systems. The economic calculations of the corrupt can be changed by using the formula: corruption = monopoly + discretion − accountability. Thus reducing scope for discretion, or strengthening accountability, will help curb corruption. Implementation. Step three is to create confidence and momentum by giving priority to highly visible examples of improvement and learning from international experience. The analysis of big leaks should be continuously followed up with programs of plugging leaks and retrieving the stolen assets. Besides other things, this will depend largely on improving coordination among government entities. Also, build- ing partnership across the public-private divide and enhancing the scope of social networks to add further infusions of transparency (and thus commitment) will be helpful. For effective implementation of strategies, there is a need to strengthen capabilities by providing hands-on training in ethics, leadership, and management and to build better systems in such vital areas as internal audit, procurement, tax bureaus, and public works. Success stories at implementation should be recorded in a checklist at the level of individual government offices, which should help spread effective practices. Outreach. The above steps will have a broader impact on pride and professionalism of the civil service, on citizens’ confidence in government, and on international perceptions of Pakistan. Donors can help the country share its success stories internationally with opinion leaders and investors. Donors and others can also help the government learn from international experience in teaching ethics and civics in schools. A promising example is Indonesia, where student-run “honesty stores� in middle and high schools give students a chance to practice what they learn in their civics curriculum, which explicitly includes anticorruption education. Unlike the usual approaches to anticorruption, the approach here takes seriously the economic and political forces that reinforce corrupt practices. It emphasizes implementation and outreach. And it recognizes the central importance of participatory processes that encourage local problem solving abetted by the best of international experience and knowledge, as most of the above steps involve participation of government, civil society (citizens at large, the business community), and donors. (continued) Box A new approach to fighting corruption (continued) 13.5 The four steps of a new anticorruption approach for Pakistan 223 4 Outreach 3 Implementation Strategy for 2 • Indices of progress transformation 1 Diagnosis • Reorganizing the • Motivating • Convenings fight against examples corruption • Where we stand • “Laboratories� • International • Credible signals opinion leaders • “Parallel systems� • Public-private-citizen of effective change collaboration • Platforms • “The anticorruption • Changing outcomes apparatus� • Priority actions—and and perception • Analysis of the “big what to leave for later • Strengthening leaks� capabilities Source: Klitgaard 2013. strategies, ensuring proper implementation of the strategies, and reaching out to the international community to make the progress visible and widely appreciated. Notes The paper benefited immensely from informal discussions with Robert Klitgaard as well as from the overall guidance of Rachid Benmessoud and Antonius Verheijen and individual contributions from Ishrat Husain, Zubair Bhatti, Sher Shah Khan, Grace Morgan, Ana Bellver Vazquez-Dodero, and Roland Lomme. 1. See “The 2012 Legatum Prosperity Index: Pakistan,� Legatum Institute, www. prosperity.com/CountryProfile.aspx?id=586, for a summary for Pakistan. 2. This also determines their attitudes toward new, efficient and open systems, such as transparency and business automation. 3. For details, see Rashid (2013), identifying key public financial manage- ment issues: excessive centralization of budget management, which limits the flexibility of provinces in executing their budget; weak cash manage- ment due to weak internal controls, which affects accuracy, completeness, Sheikhupura Sialkot Hyderabad Karachi Sukkur Quetta Peshawar Islamabad FaisalabadGujranwala Lahore Multan Rawalpindi and timeliness of the information; and lack of implementation capacity in procurement regulatory bodies. 4. A recent example of trying to restrict the functioning and authority of the National Accountability Bureau was the introduction of a new Account- ability Commission Bill (which did not pass). 5. The Recent Federal Ombudsman Institutional Reform Ordinance (passed on March 1, 2013) will help strengthen enforcement and other reforms. It has the potential to significantly improve the effectiveness of the R E v a m p i n g g o v er n a n c e Ombudsman institution in Pakistan. Key improvements include response standard of 15 days, improved organizational structure, better enforce- ment, and greater consistency. This ordinance has opened an opportunity 224 to establish an Ombudsman institution in Pakistan that can become an international best practice. 6. Patwari is an administrative government position of a village accountant. Thanedar is a police officer. 7. Two decades ago only 13 countries had RTI legislation; today 93 countries have some form of it. While the implementation record and effectiveness vary, in a number of countries RTI has indeed brought about open govern- ment and empowered citizens. Mexico and India are examples: in both, people have used it in poorer and rural communities as a tool to access basic services (Idris 2010, 2013). 8. See Lopez-Calix and Touqeer (2013), Rashid (2013), and Mukhtar (2013) for specific recommendations on these issues. 9. Some development partners (such as the United Nations Development Programme and the U.S. Agency for International Development) have or had engagements with the Election Commission of Pakistan, police department, and the justice sector. 10. For examples of governance reforms in Georgia, Korea and Vietnam, consider World Bank (2010, 2012), Hwang (2011), Cheema (2010), and Snyder and others (2010). 11. See Mukhtar (2013) for detailed discussion on decentralization reforms in Pakistan for accountability and service delivery. 12. For an overview of the bill, see Transparency Advisory Group (2013). 13. The current anticorruption strategy (see Government of Pakistan 2002) needs to be renewed, taking into consideration political ownership and using an innovative and problem-solving approach. References Abdullah, Nik R. W. 2008. “Eradicating Corruption: The Malaysian Experience.� Journal of Administration and Governance 3 (1): 42–53. Bertelsmann Stiftung. 2012. “BTI 2012—Pakistan Country Report.� Gütersloh, Germany. www.bti-project.org/country-reports/aso/pak. Dokeniya, Anupama. 2013. Implementing Right to Information: Lessons from Experience. Washington, DC: World Bank, Poverty Reeducation and Economic Management. Cheema, G. Shabbir. 2010. “Building Trust in Government: An Introduction.� In Building Trust in Government: Innovations in Governance Reform in Asia, ed. G. Shabbir Cheema and Vesselin Popovski. Tokyo: United Nations University Press. www.gsdrc.org/go/display&type=Document&id=4194. Global Integrity. 2010. “Global Integrity Report Pakistan—2010.� Washington, DC. www.globalintegrity.org/report/Pakistan/2010/. Government of Pakistan. 2002. National Anti­ corruption Strategy. Islamabad: National Accountability Bureau. 225 ———. 2011. Pakistan: Framework for Economic Growth. Islamabad: Planning Commission. Husain, Ishrat. 2009. “Pakistan: Key Governance Challenges.� Background paper for Pakistan: Country Partnership Strategy for FY2010–13, World Bank, Washington, DC. Husain, Ishrat, Naseer Rana, and Irum Touqeer. Forthcoming. “Policy Note: Strengthening Institutions of Accountability.� Islamabad: World Bank. Hwang, Yunwon. 2011. “Comparative Analysis of Indonesian and Korean Governance.� Asian Research Centre Working Paper 37, London School of Economics, London. ICAC (Independent Commission against Corruption). 2013. “About ICAC.� www.icac.org.hk/en/about_icac/bh/index.html. ICCI (Islamabad Chamber of Commerce and Industry). 2012. Unpacking Corruption. Islamabad. www.sdpi.org/mediagallery/showvideo.php?id=436. Idris, Iffat. 2010. “Pakistan Country Paper.� Prepared for Regional Workshop “Towards More Open and Transparent Governance in South Asia,� April 27–29, 2010, New Delhi. ———. 2013. “Right to Information (RTI): Pakistan Country Paper Update.� Background paper for this policy note. International Budget Partnership. 2012. Open Budget Survey 2012. Washington, DC. Klitgaard, Robert. 2013. “Ideas for Improving Governance in Pakistan.� Presentations made at Sustainable Development Policy Institute workshop, March 2013, Islamabad. Legatum Institute. 2012. “The 2012 Legatum Prosperity Index.� London. www.prosperity.com. Lopez-Calix, Jose R., and Irum Touqeer. 2013. “Mobilizing Revenue.� Pakistan Policy Note 16, World Bank, Washington, DC. ———. Forthcoming. “Revisiting the Constraints to Pakistan’s Growth.� Pakistan Policy Paper Series, World Bank, Washington, DC. Mendel, Toby. 2011. “The Right to Information and Citizenship.� Background Paper for a Conference on Freedom of Information in Zambia, March 2011. Mukhtar, Hanid. 2013. “Promoting Efficient Service Delivery with Decentraliza- tion.� Pakistan Policy Note 15, World Bank, Washington, DC. NCGR (National Commission for Governance Reforms). 2008. Reforming the Government. Islamabad: Government of Pakistan. OECD (Organisation for Economic Co-­ operation and Development). 2008. “Specialised Anti-Corruption Institutions: Review of Models.� OECD, Anti-Corruption Network for Eastern Europe and Central Asia, Paris. Available at www.oecd.org/corruption/acn/39971975.pdf. R E v a m p i n g g o v er n a n c e Rashid, Saeeda Saba. 2013. “Improving Financial Management.� Pakistan Policy Note 14, World Bank, Washington, DC. Snyder, Nicholas, Evelyn Hernandez, Lisa Maxwell, Sharinita Hester, and Naim 226 Kapucu. 2010. “Metropolitan Governance Reforms: The Case of Seoul Metropolitan Government —2010.� European Journal of Economic and Political Studies: 107–29. http://ejeps.fatih.edu.tr/docs/articles/161.pdf. The Herald and SDPI (Sustainable Development Policy Institute). 2013. Political Barometer: A Study of Socio-Political Preferences of People of Pakistan. Islamabad. Transparency Advisory Group. 2013. “Punjab Freedom of Information Bill Approved by Cabinet.� www.transparencyadvisorygroup.org/uploads/Pun- jab_FOI_Bill_Brief_for_TAG_-_15_March_13_rev.pdf. Transparency International. 2001. Corruption Perception Index 2001. Berlin. ———. 2011. Business Survey 2011: Putting Corruption out of Business. Berlin. ———. 2012a. Corruption Perception Index 2012. Berlin. ———. 2012b. Global Corruption Barometer 2010/11. Berlin. World Bank. 2007. Pakistan Country Profile 2007: Enterprise Surveys. Washington, DC. ———. 2009. Pakistan’s Investment Climate: Laying the Foundation for Renewed Growth. Volume I, Report 46435-PK. Washington, DC. ———. 2010. Vietnam Development Report—2010: Modern Institutions. Joint Donor Report to the Vietnam Consultative Group Meeting Hanoi (December 3–4, 2009). ———. 2012. Fighting Corruption in Public Services: Chronicling Georgia’s Reforms. Washington, DC. ———. 2013. World Governance Indicators. Washington, DC. http://info.world- bank.org/governance/wgi/index.asp. ———. Forthcoming. Pakistan’s Institutions of Accountability Report (Internal Report). Islamabad. World Economic Forum. 2012. The Global Competitiveness Report 2012–2013. Geneva. Saeeda Sabah Rashid and Uzma Sadaf Improving 14 Financial Management This note outlines Pakistan’s challenges in public financial management and suggests options for meeting them. The legal and institutional framework for public financial management is well structured. The government has taken steps in the right direction including automating budgeting, accounting, and financial reporting nationwide, modernizing auditing practices, and adopting medium-term budgetary frame- works. The need now is to consolidate these reforms and introduce a second generation, focused on decentralizing budget management by fully delegating budget preparation and implementation functions to the line departments (including accounting and finan- cial reporting responsibilities), strengthening cash management through business process reengineering (to prepare accurate and complete statements of cash positions as well as reliable and timely forecasts of cash requirements), and augmenting the capacity of procurement regulatory authorities (to enable them to fulfill their mandate effectively). Pakistan has a fairly developed infrastructure for public financial manage- ment (PFM). At the policy level, Parliament has a key role in authorizing revenues, expenditures, and debts. The Ministry of Finance plays a pivotal role in budget preparation and expenditure control. Line ministries, departments, and agencies (MDAs) have well-defined roles in implementing budgets and submit- ting accounts for expenditures incurred. The controller general of accounts (CGA), with an extensive network of offices, makes payments, maintains accounts, and prepares annual financial statements. The auditor general has a well-established organization for reviewing financial compliance and conducting performance audits. Over the years, the government has undertaken a range of measures for improving PFM, most notably: • Decentralization . The 18th Amendment to the constitution passed in 2010 abolished the Concurrent List (subjects on which both the National Par- liament and provincial assemblies can legislate) and transferred 17 service delivery–related functions to the provinces. Further, the share of provinces in federal receipts increased from 46.5 percent in 2010 to 57.5 percent in subsequent years under the seventh award of the National Finance Com- mission (Mustafa 2011). The Council of Common Interests, comprising the Improving finanacial management federal government and the provinces, formulates and regulates policies on matters in Part II of the Federal Legislative List and exercises supervi- sion and control over related institutions. The council has been strength- 228 ened, and the independence and powers of the auditor general have been enhanced (Burki 2010). • Separation of accounting and auditing functions . The promulgation of the Controller General of Accounts Ordinance and the Auditor General’s Ordinance in 2001 separated roles and responsibilities of the CGA (account- ing) and the auditor general (auditing). • Medium-term budgetary framework . The federal Ministry of Finance has rolled out medium-term budgetary frameworks to all line ministries. This framework has also been piloted with some line departments in the provinces of Khyber Pakhtunkhwa, Punjab, and Sindh. • Output-based budgets. Khyber Pakhtunkhwa has piloted these budgets using conditional grants in health and education, which are being scaled up in 2012/13. • Cash management. The government has largely abolished personal ledger accounts that allowed parallel cash balances, along with the consolidated cash balance. Special drawing accounts have been introduced with expendi- ture ceilings that do not transfer cash until it is actually spent. • Procurement regulatory authorities . The federal government and the provincial governments of Sindh and Punjab have established procure- ment regulatory authorities, which are custodians of the rules for all public procurement. Responsibilities include monitoring application of rules, recommending revisions to rules, making regulations, providing guid- ance to implementing agencies, and supporting these agencies’ capacity building. Audit responsibility rests with the Auditor General’s Office, and other laws are in place to address financial misappropriation. • Coordination and transparency. The government has established a sound legal framework for coordinating fiscal and debt management policies and improving fiscal transparency through the Fiscal Responsibility and Debt Limitation Act 2005, even though compliance has been less than stellar. • Project for Improving Financial Reporting and Auditing. This project has helped improve PFM systems (Box 14.1). These measures have achieved concrete results. In 2012, the government and its development partners (World Bank, Department for International Development, Asian Development Bank, and European Union) carried out the Public Expen- diture and Financial Accountability (PEFA) repeat assessments for the federal and Punjab governments. Limited updates of PEFA assessments were conducted for Khyber Pakhtunkhwa in 2010 and Sindh in 2011. A full update of the Sindh 2009 assessment is under way in 2013. In 2007, an assessment was undertaken for Box Project for Improvement of Financial Reporting and Auditing 14.1 In 1997, the Project for Improvement of Financial Reporting and Auditing set up a public financial management reform agenda. The project has supported enhanced internal control and financial account- 229 ability systems by: • Implementing a government-wide financial management information system at all levels using a state-of-the-art SAP application, thus enabling automated budget compilation, accounting, and financial and fiscal reporting nationally, including for district account and finance offices. • Introducing a new accounting model based on international principles and standards. • Adopting a uniform chart of accounts for budgeting and accounting by federal, provincial, and district governments since 2007/08, compliant with the International Monetary Fund’s Government Finance Statistics. • Separating accounting and auditing functions. However, the auditor general still has authority for prescribing the form of accounts of the government. • Supporting informed decision making for public sector management through the availability of timely and reliable in-year budget execution reports. Connectivity is in place for all line departments to monitor budget execution on a real-time basis. Further, within 15 days of the end of each month, civil accounts are prepared and presented to the Ministry of Finance. The reliability of these reports has also improved, with material reductions in irreconcilable differences and the elimination of sus- pense accounts. • Preparing annual financial statements using cash-basis International Public Sector Accounting Standards. Submission of annual financial statements for audit has improved to an average of two to four months after the end of the fiscal year, compared with more than 15 months before 2006/07. • Enhancing the effectiveness, independence, standards, and capacity of government audit and intro- ducing risk-based audit methodology compliant with international standards for certification of annual financial statements. • Improving the efficiency of external audit through computer-assisted audit techniques and a sys- tem-based audit methodology. The financial statements and audit reports are submitted to the president and governors within eight months of the close of the financial year. Balochistan. These assessments support findings presented below and confirm certain improvements in recent years, as well as shortcomings. Substantial progress on transparency through public availability of information stands out for Pakistan. It is evident in the good ratings for indicators in PEFA assessments and the marked improvement in the open budget score, from 38 in 2010 to 58 in 2012 (International Budget Partnership 2012). Annual financial statements and an audit report are laid before Parliament within nine months of the end of the fiscal year. Some of the provincial finance departments are uploading in-year reports to their websites. Pakistan’s overall scores on PEFA performance indicators are comparable with those in other South Asian coun- tries (Figure 14.1). Improving finanacial management PEFA assessment scores of federal government in 2012, compared Figure with the average of selected South Asian countries 14.1 Federal South Asia PI-1 230 PI-2 PI-3 PI-4 PI-5 PI-6 PI-7 PI-8 PI-9 PI-10 PI-11 PI-12 PI-13 PI-14 PI-15 PI-16 PI-17 PI-18 PI-19 PI-20 PI-21 PI-22 PI-23 PI-24 PI-25 PI-26 PI-27 PI-28 0 1 2 3 4 5 6 Average converted score Note: South Asia represents the simple average scores of Afghanistan (2008), Bangladesh (2006), Bhutan (2010), India (2010), Maldives (2009), and Nepal (2008). Source: Government of Pakistan and others 2012a. Weaknesses in the PFM framework can be traced to colonial heritage, resis- tance to change, rent seeking by power groups, and a lack of political will. Yet a popular surge among all segments of society demands change. The next section presents a diagnostic of key issues and selective policy options for more effective performance of the public financial management systems. Main Issues Pakistan faces several challenges in the PFM framework: • PFM is fragmented . The Ministry of Finance plays a dominant role in budgeting and expenditure control. The CGA makes payments and maintains accounts. The MDAs execute the budget within the residual space permitted by the Ministry of Finance and the CGA even though the secretary of each line department is the designated principal accounting officer (PAO) accountable for MDA performance. • Internal controls are excessive but weak. Extensively documented rules are issued by 231 the Ministry of Finance. The CGA applies controls on payments and accounts. The MDAs only exercise operational controls. Partial controls, exercised by three silos of authority, leave loopholes (as highlighted in audit reports). • The MDAs have yet to develop the capacity for professional financial management. This is largely because they have not been given full responsibility and authority to manage their budgets. • The MDAs do not link budgets to implementable procurement plans. Without this link, the MDAs do not have a full picture of a prioritized expenditure plan for development or for major recurrent activities, leading to fragmented budget utilization and poor cash planning. Moreover, the MDAs do not report the planned and actual expenditures regularly and do not adjust budget priorities in light of actual events. From these problems stem weaknesses in PFM systems, as in the accountabil- ity framework. With fragmented controls, no single authority can be held accountable for failing to deliver. Similarly, because the Ministry of Finance controls budgets and cash, the MDAs cannot always deliver services on time and economically, which is why a stronger public procurement function is key for enhanced PFM management. Centralized budget management Good budget management enables a government to be effective in the following areas: planning, executing, and monitoring resource manage- ment; ensuring targeted outcomes; and strengthening accountability. Pakistan needs to decentralize budget management. The roles of the Minis- try of Finance and the CGA need to be redefined for greater delegation. Over the years, the ministry has taken some steps in this direction, for example: • It oversees more transparent and fair processes for reappropriation of funds and supplementary grants. • It has delegated more financial powers to PAOs in several steps (the latest in 2006). • It has decentralized accounting and payment functions for some MDAs like the military, railways, foreign affairs, food, forests, and public works. • It has improved information on fund releases to the MDAs. Despite these steps, some shortcomings still hold. The PAOs, for example, still have limited flexibility in executing budgets, and they must revert to the Improving finanacial management Ministry of Finance at several stages. Independence of the PAOs is further infringed by a centralized system of payments and accounts. Account offices under the CGA process payments, maintain accounts, and prepare annual 232 financial statements. Although the PAOs sign off the final appropriation accounts, they do not assume ownership of them. The performance of the system in terms of budgetary outcomes is poor, as seen in the expenditure outturns for federal and Punjab governments in the June 2012 PEFA assess- ment (Figure 14.2). Similarly, the MDAs need approval from the Ministry of Finance for the following: allocating savings from one grant to another or even from one budget line to another within the same grant; transferring funds from current to development expenditure, or vice versa; and using approved lump- sum grants. And though in 2006 the federal government created 22 posts of chief finance and accounts officers in the MDAs, their role is still secondary to the Ministry of Finance, which maintains the primary role for stewardship of funds. The financial advisors assigned to the MDAs by the ministry exer- cise considerable influence over frequent in-year reappropriations. The centralized budget management influences the MDAs’ ability to prepare realistic budget estimates. Despite credible and accurate data available on a timely basis from the government financial management information system, decision makers have not started using this information. Equally, although the PAOs are responsible for internal controls, due to shared responsibility with the Ministry of Finance and the CGA they have not developed the systems and processes for implementing the controls, overseeing management overrides, Expenditure outturns for federal and Punjab governments (PI-2), Figure 2008/09–2010/11 14.2 Federal government Punjab 30 20 Percent 10 0 2008/09 2009/10 2010/11 2008/09 2009/10 2010/11 Note: PI-2 is a composite variation (budget and actual expenditures) taking all sectors by functional classification. Source: Government of Pakistan and others 2012a. or preventing waste and fraud. These weaknesses have led to poor ratings for internal audit and nonsalary internal controls, per the latest PEFA assess- ments, and increase the risk of over- or underspending. These weaknesses also contribute to Pakistan’s low rating in PEFA assessment scores (PI-1 to PI-4) 233 on budget credibility—lower than the average of similar lower middle-income countries (Figure 14.3).2 Other scores (PI-13 to PI-21) related to budget execu- tion demonstrate a similar pattern (Figure 14.4). Weak cash management The objectives of modern cash management are to ensure that cash is available to pay for expenditures when due, borrowing occurs only when needed at optimum costs, returns on idle cash are maximized, and risks are managed by productive investment of surpluses. The generally accepted good practice for every government is to have a single treasury account, which Pakistan has yet to achieve according to PEFA assessments. At all levels of government, some accounts still remain outside the consolidated funds, even though they represent a very small share of total outlays (largely related to donor-funded projects). The Ministry of Finance has taken several steps for improving cash manage- ment, although accuracy, completeness, and timeliness of information remain a challenge. The federal, Sindh, and Khyber Pakhtunkhwa governments Figure Budget credibility (PI-1 to PI-4) 14.3 Pakistan Lower middle-income countries PI-1 4 3 2 1 PI-4 PI-2 PI-3 Note: PI-1 is aggregate expenditure outturn compared with original approved budget; PI-2 is composition of expenditure outturn compared with original approved budget; PI-3 is aggregate revenue outturn compared with original approved budget; and PI-4 is stock and monitoring of expenditure payment arrears. Source: Government of Pakistan and others 2012a. Improving finanacial management Figure Predictability and control in budget execution (PI-13 to PI-21) 14.4 Pakistan Lower middle-income countries PI-13 234 PI-21 PI-14 4 3 2 PI-20 1 PI-15 PI-19 PI-16 PI-18 PI-17 Note: PI-13 is transparency of taxpayer obligations and liabilities; PI-14 is effectiveness of measures for taxpayer registration and tax assessment; PI-15 is effectiveness in collection of tax payments; PI-16 is predictability in the availability of funds for commitment of expenditures; PI-17 is recording and management of cash balances, debt, and guarantees; PI-18 is effectiveness of payroll controls; PI-19 is competition, value for money, and controls in procurement; PI-20 is effectiveness of internal controls for nonsalary expenditure; and PI-21 is effectiveness of internal audit. Source: Government of Pakistan and others 2012a. prepare cash forecasts quarterly while the Punjab government prepares them every month. However, there are significant gaps on cash forecasts and actual cash positions. The accuracy of the cash forecasts is undermined by missing information on the stock of arrears, commitments, and unresolved recon- ciliation differences. While the State Bank of Pakistan provides daily data on monetary balances, the National Bank of Pakistan submits data with varying lags. Weak cash management affects the Ministry of Finance’s ability to prepare monthly cash-flow forecasts, in turn influencing the MDAs’ capacity to imple- ment the budget effectively. Due to delayed information on cash positions and commitments, the Sindh government unknowingly exceeded its overdraft limit in June 2012. Lack of timely reconciliation of balances, unresolved reconcilia- tion differences, and substantially incomplete information on commitments and arrears undermine the Ministry of Finance’s other efforts to streamline cash management (Figure 14.5). PEFA assessments indicate some improvement in reconciling bank and book balances in the Khyber Pakhtunkhwa, Punjab, and Sindh governments but report an unsatisfactory rating for the federal and Balochistan governments. There remain long-outstanding differences between the Economic Affairs Division and the Accountant General Pakistan Revenue on the status of foreign debt, but the overall net unresolved differences on this account as of June 30, 2012, were marginal, at PRs 1.86  billion. Progress in implementing PEFA assessment scores for indicators related to cash Figure management 14.5 Federal Punjab Sindh Khyber Pakhtunkhwa Balochistan 4 235 3 Score 2 1 0 PI-4 PI-16(i) PI-17 PI-22 PI-24(i) Note: PI-4 is stock and monitoring of expenditure payment arrears; PI-16(i) is extent to which cash flows are forecasted and monitored; PI-17 is recording and management of cash balances, debt, and guarantees; PI-22 is timeliness and regularity of accounts reconciliation; and PI-24(i) is scope of reports in terms of coverage and compatibility with budget estimates. Source: Government of Pakistan and others 2007, 2011a,b, 2012a,b. commitment accounting has been quite slow. Data on commitments and expen- diture arrears are absent from the in-year reports—generating poor scores in the latest PEFA assessments—thus creating a risk of overspending or failure to make timely payments, which can lead to penalties. One contributing factor is the cen- tralized accounting and payments function. Lack of capacity in procurement regulatory authorities Procurement is a devolved subject under the constitution. Reform progress has been intermittent at the federal and provincial levels, though all levels of government have clear regulatory frameworks that apply to all public procurement for which open competitive bidding is the default method, for adequate disclosures of all procurement information, and for dispute mecha- nisms. The federal and provincial governments have enacted public procure- ment rules, generally aligned with good practice. The federal government, as well as Punjab and Sindh, have procurement reg- ulatory authorities (PRAs), but Balochistan and Khyber Pakhtunkhwa have yet to establish them. Regulatory frameworks at the federal and provincial uniform levels follow a Islamabad principle of keeping FaisalabadGujranwala Lahore Multan Rawalpindi PRAs in the role of custodian of Sheikhupura Sialkot Hyderabad Karachi Sukkur Quetta Peshawar rules. The PRAs develop adequate documentation and monitoring, assessing and enhancing procurement effectiveness. The audit function lies with the Auditor General’s Office, and parallel laws and systems are in place to take Improving finanacial management action if major issues arise in compliance, including financial misappropria- tion. Yet although these responsibilities for custodianship, audit, and penal actions seem to ensure sufficient checks and balances, effective implementa- 236 tion has yet to be established. Implementing agencies conduct the procurement transactional function. Procurement responsibilities are allocated at different operational levels, depending on the cost ceilings of contracts. Various implementing agencies have developed their own operating procedures for devolving financial pow- ers. Transactions with the highest costs require approval of the concerned secretary. Thus, understanding and compliance with procurement rules vary by agency. Some have a good appreciation of them, as well as the resources and capacity for sound procurement, while others do not. The latest PEFA assessments on procurement management show very low scores for Pakistan, largely because of a general lack of credible data among PRAs stemming from the absence of effective monitoring and evaluation (Table 14.1). The major challenge is to improve the implementation of the procurement framework. The PRAs are responsible for initiatives to better implement and further develop the framework, but they face many challenges, such as a lack of resources, undermining their capacity. Broader challenges include a general lack of procurement capacity at every level of public administration and an Audi- tor General’s Office historically focused on compliance rather than outcomes. More urgent issues include the following: • The PR As do not have a monitoring and evaluating system to collect and analyze data on procurement actions or evaluate the procurement performance of the MDAs. • A capacity-building strategy for procurement by the MDAs is lacking. • There is no clear capacity-building program for vendors, especially small and medium-size enterprises. Any healthy procurement system relies on Table PEFA assessment scores for the procurement-related indicator (PI-19) 14.1 Government Score Federal D+ Punjab D Sindh C Khyber Pakhtunkhwa C Balochistan D+ Note: PI-19 is transparency, competition, and complaints mechanism in procurement. Source: Government of Pakistan and others 2007, 2011a,b, 2012a,b. its vendors, who are fundamental for competition and thus in overall pro- curement, including its transparency. • The PRAs have no mechanism (formal or otherwise) for knowledge and experience sharing, coordination, or cooperation. 237 • The regulatory framework considers independent dispute resolution mechanisms for vendors’ complaints, but they have yet to be set up. Policy Options Decentralize budget management Experiences of countries like Norway, South Africa, and Sweden suggest that the best option is to move toward decentralizing budget management, through the following measures: • All MDAs should have dedicated units for financial management with qualified staff. The PAOs should have full responsibility and accountability for the entire budget cycle. Internal controls need to be strengthened through internal audits. • The Ministry of Finance’s role in budget management should be limited to broad policy guidelines, and the current regime of the financial advisor needs to end. • Payments and accounts functions, including accounting for revenues, should be transferred, with staff, to the MDAs. They would be responsible for ensur- ing commitment accounting, preparing annual financial statements, and getting these statements passed by the auditor general. The Ministry of Finance, CGA, auditor general, and Establishment Division will need to work in tandem in bringing to being the proposed decentralized system. It will require dedicated financial management units staffed by professional personnel, who can initially come from the existing accounting staff. The auto- mation of budgeting, accounting, and financial reporting, as well as connectivity to line ministries, is already in place and provide a basic platform for this reform. Improve cash management Informed cash management based on monthly cash-flow forecasts, a single treasury account, and regular reconciliation of accounts has become generally accepted best practice. Especially in times of fiscal constraint, a credible state- ment of the cash position and a robust basis for preparing a comprehensive cash forecast are critical. Given current gaps, business process reengineering will be required to meet these objectives. For institutionalizing top-rate cash manage- ment, the Ministry of Finance should set up a high-level committee to oversee monthly cash-flow forecasts, reconcile accounts, and borrow and invest surplus Improving finanacial management funds in the short term. Countries such as Argentina, Brazil, Chile, Costa Rica, Ecuador, and Peru have such committees in their treasury departments (van Eden and Weiner 2010). 238 Arrangements to be carried out include the following, benefiting the Ministry of Finance as well as the MDAs, which will get more reliable infor- mation on resource availability: • Procedures to ensure that all cash balances become part of the consolidated funds. Policies and procedures already issued for the establishment of a single treasury account need to be made operational. All existing exemptions for keeping some cash balances out of the consolidated funds should be withdrawn. • Timely and regular reconciliation of differences in accounts needs to be enforced. The present pace of reconciling outstanding balances must be accelerated. A cut-off date must be specified for historical items, after which these should be written off as a one-time adjustment. • Associated with timely information on commitments are the data on stock of arrears, which are unavailable at any level. • Better and more robust planning through preparation of procurement plans concurrently with the finalization of budget proposals. Ensure that PRAs run effectively Making federal and provincial governments’ PRAs operational and effective should be a priority, based partly on the following options: • Effectively implement a complaint and dispute resolution mechanism and strengthen information disclosure mechanisms, as well as civil society engagement in them. • Design a mechanism for monitoring and evaluating the MDAs’ procurement performance. This would be cost-effective and would help benchmark the MDAs’ performance, develop action plans for better functioning, and improve underlying documents and rules. E-procure- ment is crucial, as it will provide the raw data on which to build indicators and algorithms that will empower the PR As in their enforcement task. It will also provide natural integration with the country’s financial management systems. • Have PR As write consistent and sustainable capacity-building strate- gies for procurement professionals inside the MDAs and the vendor community, reflecting the vendors’ crucial role. • Set up a forum for sharing knowledge among all PRAs, and streamline the standard documents of the PRAs, to harmonize the national procurement regime. • Support the auditor general to develop a strategic plan for auditing the procurement performance of all major procurement actions to improve over- sight and accountability. 239 Notes This note was written with contributions from Uzma Sadaf, Muhammad Akram Khan, and Hammad Yunus. 1. “The government� in this note generally includes the federal government and provincial governments. 2. For more information on performance indicators (PI), see Government of Pakistan and others (2007, 2011a,b, 2012a,b). References Burki, S. Javed. 2010. “The 18th Amendment: Pakistan’s Constitution Rede- signed.� ISAS Working Paper 112. Institute of South Asian Studies, Singapore. Government of Pakistan, World Bank Group, European Commission, DFID (Department for International Development), and ADB (Asian Develop- ment Bank). 2007. Pakistan Balochistan Province Government Public Financial Management and Accountability Assessment. May. Islamabad: Government of Pakistan. ———. 2011a. Pakistan Sindh Province Government Public Financial Management and Accountability Assessment Update. October. Islamabad: Government of Pakistan. ———. 2011b. Pakistan Khyber Pakhtunkhwa Province Government Public Financial Management and Accountability Assessment Update. June. Islamabad: Govern- ment of Pakistan. ———. 2012a. Pakistan Federal Government Public Financial Management and Accountability Assessment. June. Islamabad: Government of Pakistan. ———. 2012b. Pakistan Punjab Government Public Financial Management and Accountability Assessment. June. Islamabad: Government of Pakistan. International Budget Partnership. 2012. Open Budget Survey 2012: Open Bud- gets, Transform Lives. Washington, DC. http://internationalbudget.org/ publications/obi-full-report-2012-english/. Mustafa, Usman. 2011. “Fiscal Federalism in Pakistan: The 7th National Finance Commission Award and Its Implications.� Working Paper 73. Pakistan Insti- tute of Development Economics, Islamabad. van Eden, Holger, and Fanny Weiner. 2010. “Treasury Management Practices in Latin America: Results of a Survey of State Treasuries.� PowerPoint pre- sentation at Seminario Latinoamericano sobre Gestion de Tesoreria, Lima, April 15–16. 15 Promoting Efficient Hanid Mukhtar Service Delivery with Decentralization Over the last few years, Pakistan’s intragovernment relations have undergone major changes. While they have created many opportunities for improving service delivery, they have also created serious risks. The biggest is state authority concentrated in the middle tier of government, which remains too isolated from the population to make any serious shift toward “bottom-up� accountability. This note argues that the disconnect between political goals and service delivery is the biggest reason decentralization has failed to improve social outcomes. It presents contours of the key actions for a more pragmatic decentralization model, which can bridge the gap between the normative and doable aspects of decentralization. The model sees a role for local governments in municipal and other local services, while devolving other key provincial services like health and education to autonomous provincial author- ities to be headed by elected local representatives. The financial sustainability of devolved functions needs to be ensured through improved revenue efforts by the federal and provincial governments. And debt management concerns can be allayed if the province maintains fiscal discipline by legislating and adhering to fiscal responsibility laws. A Historical Perspective Over the last few decades, more than 75 countries have attempted to transfer state responsibilities to lower tiers of government. Significantly, most of these governments were elected, making such decentralization not just administrative or fiscal but also political. The political motivation for these measures varied. In Eastern Europe and the former Soviet Union, it was part of the political and economic transition. In Latin America, it was to reinforce the move to deepen democracy. In Indonesia, South Africa, and Sri Lanka, it was in response to ethnic or regional conflict. And in Chile, Côte d’Ivoire, and Uganda, it was to improve the delivery of basic services (Shah, Thompson, and Zou 2004). Irrespective of the actual motivation, an explicit (or implicit) reason for decen- tralization was general dissatisfaction among the populace with centralized delivery of public services. Many people believed that decentralization could improve governance of services by improving allocative efficiency (better match- ing public services to local demands) and accountability (taking government closer to the people while enhancing productive efficiency of public services and infrastructure). Yet experience has been mixed—see, for example, Burki, Perry, and Dillinger (1999); World Bank (2001, 2003); and Ahmad and others (2005). Some common problems associated with decentralization’s impact on service 241 delivery have begun to emerge. It is thus critical to understand that even if a decentralization plan has political, administrative, financial, and service deliv- ery objectives, it could only be achieved, and sustained, if it can address all the above aspects—and they must fit well together. Pakistan has tried to decentralize government, but most of its initiatives, if not all, have failed to deliver the desired results—mainly because the primary goal of each was political, and little attempt was made to balance the other objectives, especially service delivery. Since its independence in 1947, Pakistan has been a federation. This is partly an outcome of the constitutional legacy of British India as reflected in the Pakistan Resolution, which contained many elements of federation (and indeed confederation); partly because of its geographic reality of consisting of two noncontiguous territorial units (West and East Pakistan); and partly because the provinces developed distinct ethnic and linguistic identities of their own as former sovereign states. However, the federal provisions of the resolution remained inactive at the national level even as it continued to serve as a Magna Carta for ethnonationalists. That the two units were considered vulnerable to Indian designs pushed the ruling elite toward centralizing political power in Karachi’s—and later Islamabad’s—hands. The center amassed enormous powers under the 1947 ­ Independence of India Act (Salamat 1992). Pakistan was thus born into a para- doxical political situation that made federalism at once the only viable form of government and as toothless as possible (Waseem 2010). It is no surprise, then, that Pakistan’s successive constitutions (see below) remained federal in form but unitary in spirit. Pakistan had an unbalanced federal structure because of the demographic weight of the province of East Pakistan (55 percent of the country’s popula- tion). The ruling elite based in the West Wing shunned the grim prospects of rendering a permanent majority position to the East Wing and pushed the agenda for interwing parity of equal representation in Parliament (Waseem 2010). This quest for parity was the driving force behind merging all four provinces in the West Wing into a megaprovince—West Pakistan. But this had two unintended consequences. First, it led to fierce backlash in the form of ethnonationalist movements in Sindh, Balochistan, and North West Frontier Province. Second, Lahore was made the capital of West Pakistan, and in 1960 the nation’s capital was shifted from Karachi to Rawalpindi and promoting efficient service delivery with decentralization then Islamabad, both in Punjab. These federalist arrangements were widely seen as a sign of the “Punjabization� of Pakistan. Punjab’s domination of the armed forces, which ruled the country for most of the 1960s, further 242 strengthened this view. In the 1956 Constitution, the distribution of legislative powers between the federation and the federating units was enumerated in three “lists.� The Fed- eral Legislative List had 30 items, the Provincial List 94, and the Concurrent List (which enumerated functions that could be performed by the federal, provincial government, or both) only 19. Although wide powers were given to provincial legislatures, the shadow of federal control remained as the federal government was entrusted with ensuring that each province was governed according to the constitution. A provincial government was obliged to exer- cise its executive authority so as to comply with Acts of Parliament and laws applying to that province. The central legislature, however, could legislate on any matter connected with a provincial subject on the grounds of national stability, planning, coordination, or achievement of uniformity on any matters in different parts of Pakistan. In a conf lict between central and provincial laws, the former prevailed. The 1956 Constitution did little to tilt the distribution of financial resources toward the provinces and away from the center. The main sources of the provinces’ income were taxes on mineral rights, electricity, vehicles, adver- tisements, animals, boats, professions, trades, and luxuries, as well as an excise on alcohol and drugs. The 1962 Constitution, framed by a military government, adopted a pres- idential form of government and enhanced the legislative role of the central government. This constitution had only a Federal Legislative List, allowing provinces legislative jurisdiction over residual functions only. On fiscal resources, it gave the president powers to establish a National Finance Commission (NFC), comprising the central finance minister, provincial finance ministers, and other persons the president might appoint after consulting with provincial governors. These centralizing moves created serious dissention in the East Wing. In 1966, the influential Awami League’s Six Points Program sought to radically redefine federalism by demanding six key areas for the center. The six areas are defense and foreign policy (with communications); two con- vertible separate currencies (or one currency to be handled by two separate reserve banks for the two wings); taxation for the provinces; the right of the provinces to handle foreign exchange and foreign trade; and paramilitary forces for East Pakistan. However, the political leaderships in the two wings could not agree on the amount of provincial autonomy. This, and the refusal by the West Wing’s political leadership to allow the Awami League to form a national government after the 1971 elections, led to the federation’s collapse in 1971, when East Pakistan became the independent state of Bangladesh. 243 The secession of East Pakistan in 1971 prompted new thinking on feder- alism. As Punjab—with 58  percent of the population in the remaining areas—held a dominant position, the smaller provinces moved to constrain the majority of one province in Parliament. Politicians from Sindh, North West Frontier Province, and Balochistan successfully lobbied for some kind of majority-constraining federalism. The 1973 Constitution therefore adopted a two-tiered legislative structure. It consisted of a National Assembly, whose membership was decided on majority basis and dominated by members from Punjab, and a Senate, where all four provinces enjoyed equal representation. The Senate was thus supposed to balance the majority dominance of the lower house. But it was given no control over the crucial Money Bill, such that the national budget could be sent for presidential approval after bypassing the Senate.1 The 1973 Consti- tution also brought back the Concurrent List.2 Pakistan has remained a federation of provinces, but none of the consti- tutions saw any role for subprovincial governments. Nonetheless, the 1973 Constitution recognized management of local governments as important, including it in the Concurrent List, but the function was performed by provincial governments. In that sense, the constitution acknowledged the need to have local governments but only as extended arms of provincial government. This constitutional ambiguity has remained Pakistan’s biggest impediment to developing local governments. 3 The first major move to institute local governments in Pakistan was in the 1960s under the military regime of General Ayub Khan, which established the system of Basic Democracies. Local governments ceased to exist in the early 1970s because the elected government of Zulfiqar Ali Bhutto did not hold local government elections, despite its stated intent. During General Zia ul Haq’s rule, local governments came back as a third tier of government under the 1979 Local Government Ordinance (LGO), but again with only limited authority, as provincial governments were given broad oversight and advisory authority over local governments. The first real attempt to fully empower local governments was made under General Musharraf. The 2001 LGO devolved some key functions to local promoting efficient service delivery with decentralization governments, including substantial concomitant financial powers. The 17th Amendment to the constitution devolved key functions like education and health to the local governments and provided them time-bound protection. 244 Still, the main objectives of this devolution were not different from those of the previous military governments. Since 1947, local governments have been used mainly to depoliticize governance, create a new political elite to challenge and undermine the political opposition, and demonstrate the democratic credentials of a regime to domestic and external audiences. Political objectives aside, the 2001 Devolution Plan aimed to change the gover- nance structure, with strong implications for delivery of basic public services.4 For example: • It expanded the scope and authority of elected local government . Local governments before this devolution had little role, as they were practically inactive and most state functions were carried out by the provincial bureaucracy. The plan devolved some key functions, including education, health, and agricul- ture, to local governments, transferring the vast majority of public services previously performed by provincial governments’ district administrations to elected local governments. Local governments’ responsibilities for services and how they allocated district expenditures rose substantially (Cheema, Khwaja, and Qadir 2005). • It engendered electoral accountability. The district administration head—the district coordination officer—now reported directly to the elected head of the government—the nazim. This was a significant departure from the previous systems, where the de facto head of the district administration, the deputy commissioner, reported to the provincial government. • It reduced bureaucratic power. Unlike the deputy commissioner, the district coordination officer was deprived of executive magistracy and revenue- collection powers. • It established a rule-based system of fiscal transfers. Distribution of fiscal resources between provincial and local governments was determined by a new formula to which the provincial finance commission agreed. This body comprised representatives of the provincial and local governments. From the outset, however, provincial political and administrative cadres were uneasy with the new local government system. This was partly because the devolution plan did not provide any viable framework for integrating provin- cial and local governments. These frictions led the provincial government to systematically obstruct local government by not implementing the plan as envisaged. The administrative authority of local governments was curtailed, as employees stayed in the provincial domain. Even the appointment, posting, and transfer authority given to the local government was frequently subverted on various pretexts. Similarly for fiscal transfers, the district share in the devel- opment budget was much smaller than in the recurrent budget (Table 15.1), and even for the recurrent budget to the districts, the bulk of development expenditure was retained at the provincial level. Moreover, a sizable share of the 245 recurrent budget became allocated for wages and pensions (Table 15.2), leaving a very small amount for operating expenditure.5 With the return to power of elected governments (in 2008), the 2001 devolution was largely reversed. The new governments showed no inclination to continue with the constitutional protection provided to local governments by the 17th Amendment, which lapsed in December 2009. No attempt was made to hold fresh local elections. As such, the local governments today are controlled by provincial government officers rather than elected local representatives. In addition, all four provincial governments moved to curtail the administrative and financial powers of local governments.6 There are two overlapping reasons for this lack of support for local govern- ments by elected regimes. First, nondemocratic regimes in Pakistan tend to centralize power at higher levels of government but must rely on local gov- ernments to lend political legitimacy to their rule and to seek a political con- stituency to discharge government functions.7 Second, as the constitution recognizes only federal and provincial governments, most political parties Table Share of districts in overall provincial expenditure, Punjab 15.1 (percent) Fiscal year Current Development Total 2008/09 29.5 20.3 27.4 2009/10 28.6 18.5 25.6 2010/11 31.0 19.7 28.5 Source: Office of the Accountant General, Government of Punjab. Table Share of wages and pensions, Punjab 15.2 (percent) Fiscal year Province Districts 2008/09 28.9 71.9 2009/10 28.0 68.3 2010/11 29.9 71.5 Source: Office of the Accountant General, Government of Punjab. promoting efficient service delivery with decentralization are structured along these lines. Moreover, leaders of political parties are much more comfortable working with a limited number of party workers if they have to be engaged with national and provincial politicians. Including a 246 large cadre of local politicians may create difficult managerial problems for the party leadership (Cheema, Khwaja, and Qadir 2005). Main Issues One lesson that emerges from the fiscal federalism of recent years is that the 18th Amendment does not have service delivery as its prime objective, and moves to bottom-up accountability are incomplete. The amendment was driven mainly by political considerations, and decentralization remains unfinished. Specific issues are now discussed. Increased provincial dependence on federal transfers. The federal and pro- vincial governments agreed on a new (Seventh) NFC Award (2010)8 that significantly raised the share of provinces in the divisible pool of federally collected taxes—from 46.3  percent to 57.5  percent. The federal government also cut that pool byreducing its retained collec- tion charges from an average of 5.2 percent to just 1.0  percent. For horizontal distribution of revenues among provinces, the award moved from using population share as the sole criterion and included other factors as well, such as poverty and backwardness, tax collection, and geographic size. The award thus made a direct attempt to cater to the needs of the two smaller provinces. Over the last few years, natural gas–related revenue transfers, especially to Balochistan, have declined sharply, leaving a huge resource gap in the provincial budget. To cover this gap, the new award greatly increased Balochistan’s share in the divisible pool. The shares of all other provinces were reduced. Nonetheless, to help Khyber Pakhtunkhwa cover the cost of the “war on terror,� 1 percent of gross revenue from the divisible pool is provided to the province as a direct transfer. Conversely, all federal grants and subventions have beendiscontinued and rolled into the divisible pool. More complex fiscal management. The rapid increase in revenue transfers to the provinces has strong implications—not only for federal and provincial government finances but also for overall fiscal management. Indeed, given the budget’s tightness and rigidity, the implication of the Seventh NFC Award is that the federal government has to either mobilize large levels of additional resources to meet all its expenditure needs or discontinue a large portion of its budget- ary activities. It faces this choice because the award is not predicated on new monies and in a sense plays a zero-sum game. The award can make one provincial government fiscally better off 9 only by making one or more other governments (clearly the federal government) worse off. The award has with- out a doubt made fiscal consolidation harder, as seen in the following: • Increased fiscal stress at the federal level . Under the Seventh NFC Award, the fed- eral government transfers additional revenue of more than 1.2 percent of 247 GDP to provinces each year. Under the 18th Amendment, annual expendi- ture devolved to the provinces is less than 0.4 percent of GDP. This is mainly because only a small portion of employees associated with devolved functions was accepted by provincial governments, and many remain on the federal payroll, implying that federal government fiscal deficit rose 0.8 percent of GDP a year solely because of new fiscal federalism arrangements. • Expenditure rigidities . With a sharp decline in its revenue share, the fed- eral government is expected to scale back its expenditures, preferably current spending. However, a very large portion of current spending of the federal government is extremely rigid—almost three-fourths is spent on interest payments, defense, and employees’ wages, allowances, and pensions.10 The federal government might thus not be positioned to substantially scale back its current expenditure. However, every PRs 100 increase in federal transfers leads to an increase of more than PRs 40 in provincial current expenditure. As such, current expenditure is expected to continue to increase. In addition, the development expenditure of the provincial governments also appears linked to the size of their revenue surplus (expected to be greater with larger transfers from the federal government). The provinces may be unwilling—or unable11—to reduce their expenditure, such that fiscal consolidation can be achieved only be scaling back the federal government’s development program, with strong implications for investment and growth. And even this may not be enough to achieve desired levels of fiscal consolidation. • Disincentives for revenue mobilization . Pakistan’s tax to GDP ratio is among South Asia’s lowest, mainly because some major sectors and areas of the economy are untaxed or taxed very lightly.12 Although a major effort from the federal and provincial governments is required to raise more revenue, the NFC Award has built-in disincentives against any such attempt. The majority of revenue collected at the federal level is now transferred to the provinces, and the federal government is left with insufficient revenue to adequately finance its functions. Thus, from the federal government’s perspective, the balance between the need for collecting more taxes and the associated political cost has tilted toward the latter. But the large inflow of free federal transfers helps reduce the already weak political will in the provinces to make any real revenue effort of their own.13 In addition, the assignment of the general sales tax on services to the provinces has forestalled all efforts to institute an integrated general sales tax and value added tax system in the country. promoting efficient service delivery with decentralization Increased social spending captured by wage and price increases. Despite its adverse implications for fiscal management, the main motivation behind the NFC Award—and perhaps for the optimism that followed—was that provinces would 248 have almost total jurisdiction over social sectors. Weak provincial finances and the consequent underfinancing of social sectors were considered major factors in the country’s poor social indicators. Data from post-NFC provincial accounts show that provinces made a sizable increase in fiscal resources to social sectors. Over 2009/10–2011/12, provinces’ overall expenditure increased 24 percent a year, with education and health expenditures each growing 25 percent a year. However, corresponding progress is not yet visible, mainly because much of the increase has been eroded by steep wage increases. Indeed, the salary expenditure of district governments rose 34 percent in 2010/11. Political motivation of decentralized governance. By eliminating the Concurrent List, the 18th Amendment abolished the federal government’s role in discharg- ing these functions.14 As these functions were transferred to the provinces, 17 federal ministries were abolished, making more than 38,000 federal employees redundant. Nonetheless, decentralization did not intend to reduce payroll. It instead targeted the broader objective of stabilizing democracy by reestablish- ing the supremacy of Parliament, balancing the power of the president, prime minister, and National Assembly, and reducing opportunities for military intervention in government.15 Laws on, and delivery of, services assigned to the federation. The 18th Amendment strengthened the Council of Common Interest (CCI) and made the National Economic Council more responsive to provincial interests. It also eliminated federal–provincial shared functions, reassigned some functions to the fed- eration (not the federal government) to be guided by the CCI, and devolved others to the provinces. This change has raised some pertinent questions about framing rules and formulating procedures for services. The CCI is solely a grievance-redress and decision-making body, so it has no mechanism to legislate the laws or frame the rules and procedures for delivering the functions that the 18th Amendment assigns to it. Any CCI decision requir- ing a change in any law thus requires legislation by the national and/or pro- vincial assemblies.16 Similarly, the procedures have to be formulated by the federal and/or provincial governments. Moreover, federal and/or provincial governments will be responsible for implementing all CCI decisions, which may cause significant delay or uneven implementation. Bringing the government closer to, or distancing it from, the people? By not provid- ing any constitutional basis for the local government, the 18th Amendment has paved the way for concentrating state authority at the middle tier of government. This move runs counter to worldwide trends, which reflect tech- nological developments that have made the middle tier of government largely redundant and the concentration of state power in the two tails—federal and local governments. 249 Debt management concerns. By opening the door for provinces to borrow directly from the domestic markets and international donors, the 18th Amendment has added potential risks and brings urgency to debt management. Provinces should adequately finance their development needs, and unless they do this prudently, their actions can create enormous fiscal risks.17 Issues with decentralized service delivery.18 This dominance of politics over service delivery has manifested itself in at least three factors that have prevented any significant improvements in service delivery. They are: • Capacity bottlenecks . The lack of subnational government capacity to exercise responsibility for public services is deep-rooted. Similarly, some the functions devolved to the provinces under the 18th Amendment require skills that may not exist at that level. Until provinces acquire or develop these capacities, the quality of associated services is likely to suffer. • Asymmetric responsibilities . All the employees associated with the devolved functions in 2001, including health and education, continued to function as provincial employees, with the district nazim having little authority over their hiring, firing, evaluation, or placement. Similarly, while oversight of the salary budget was transferred to districts, inadequate transfers were made for nonsalary recurrent and development budgets, depriving district governments of any opportunity to improve service delivery. • Misaligned priorities . When the 2001 LGO assigned much greater respon- sibility to elected district governments by devolving education and health to district cadres, these broader functions appeared a little out of sync with the nature and character of these governments. Their focus remained on small civil works and local functions, as very little of their development budget was spent of health and education (Hasnain 2010). Priorities are still skewed heavily toward infrastructure (Table 15.3). Confusing accountability roles. One argument in favor of the 18th Amendment is that it has brought greater clarity to government accountability. Account- ability for public services now starts and ends with provincial governments—a strong incentive for them to improve service outreach and quality. However, only a fraction of the functions of three main federal ­ministries—agriculture, health, and education, abolished under the 18th Amendment—were devolved (or abolished), and many were retained by the federal government, assigned to existing and new federal ministries (Table 15.4). promoting efficient service delivery with decentralization Table Sectoral shares in district development expenditures 15.3 (percent) 250 Sector 2009/10 2010/11 General public service 5.3 6.9 Public order and safety affairs 0.0 0.0 Economic affairs 40.5 38.9 General economic, commercial, and labor affairs 0.0 0.0 Agriculture, food, irrigation, forestry, and fishing 3.5 5.5 Mining and manufacturing 0.0 0.0 Construction and transport 36.4 32.8 Communications 0.0 0.0 Other industries 0.0 0.0 Research and development economic affairs 0.6 0.5 Environment protection 0.5 0.7 Housing and community amenities 2.2 3.2 Health 2.3 2.0 Recreation, culture, and religion 0.4 0.2 Education affairs and services 8.2 9.1 Social protection 0.1 0.1 Total 100.0 100.0 Source: Office of the Accountant General, Government of Punjab. Number of functions and departments devolved, abolished, or Table retained by the federal government 15.4 Before 18th Amendment After 18th Amendment Devolved Abolished or merged Retained Ministry Functions Departments Functions Departments Functions Departments Functions Departments Agriculture 38 13 9 1 2 3 27 9 Health 15 22 3 4 0 3 12 15 Educationa 14 19 4 4 0 0 10 15 a. Functions for standards in higher education and research, scientific, and technical institutions were taken to the Federal Legislative List, Part II. Source: Paul and others 2012. Policy Options For the decentralization introduced by the 18th Amendment to improve service delivery, the following actions are critical:19 Strengthen fiscal management and discipline • The CCI should establish a national fiscal framework before the beginning of each fiscal year, which would establish the key fiscal goals at the national level and assign each government (federal and provincial) targets (reve- nue, expenditure, fiscal deficit, and financing)20 to help achieve national fiscal goals. This framework would guide the federal and all four provincial budgets for the forthcoming fiscal year. The CCI would periodically review 251 the implementation of the fiscal framework during the year and make required changes (if any). • The federal and provincial governments should undertake tax policy and tax administration reforms to mobilize additional revenue and make the tax system more efficient and equitable. Bringing untaxed and lightly taxed sectors in line with other sectors should help achieve these objectives. • Provinces should legislate (and adhere to) their own fiscal responsibility laws to ensure overall fiscal discipline. • The public financial management system (budgeting protocols, accounting codes and classification, and auditing and procurement rules and proce- dures) should be extended to all local governments. This would require considerable strengthening of provincial government capacity, especially of the district accounts officers, provincial accountant generals, provincial procurement regulatory agencies, and Auditor General of Pakistan. Rationalize the assignment of functions • The CCI (or a similar body) should review the devolution of “national� functions (Shah 2012) and develop modalities and a timeline for assigning them to the federal government or federation (that is, jointly to a body having representation of the federal government and all four provincial governments). • Roles and functions of each level of government should be further clarified. Overlapping functions across government tiers cloud this clarity and diffuse accountability. Improve service delivery • Each level of government should have complete control over the functions assigned to it. This implies revising the rules and procedures at each level to weed out provisions that can enable intrusion from higher levels. • Institutions supporting the federation, such as the CCI, NFC, and National Economic Council, need to be strengthened with secretariats to do upfront work on consensus building and on monitoring implementation of these institutions’ decisions. • Adequate human and technical resources for performing the assigned func- tions should be available to the level of government responsible for those functions. The higher level of government keeping even an oblique control promoting efficient service delivery with decentralization over staff not only makes it difficult for the responsible level to perform the functions per its policies and priorities but also gives it an excuse and a shield against bottom-up accountability. 252 • Similarly, the responsible level should have adequate financial resources to discharge these functions. Irrespective of the strong economies-of-scale argument for centrally collecting taxes, each level of government should have enough tax powers to raise its budgetary flexibility, but more import- ant, to strengthen bottom-up accountability. Fiscal transfers from higher level government should be mainly for fiscal equalization and achieving national and provincial objectives.21 • To ensure top-down (and bottom-up) accountability, local governments should have complete financial autonomy. Each district government (metropolitan corporation, municipal corporation, municipal committee, district education and health authorities, and the like) should be given complete control of their budgets according to their priorities and ground conditions, irrespective of whether the funds are transferred by the provincial government or raised by the local government. Each provincial finance department, plan- ning and development department, accountant general office, and office of the Auditor General of Pakistan has a twin responsibility to build local fiscal and financial management capacity and to ensure that broad fiscal discipline is maintained—the latter through establishing a system of fiscal reporting that can raise red flags and timely alerts to avoid fiscal or financial disaster. • Financial autonomy of local governments can also be strengthened if some broad-based and buoyant revenue bases were devolved to them, especially to the larger ones. The main candidates are the Urban Immovable Property Tax, Profession and Callings Tax, and Entertainment Tax. Over time, local governments may be given the right to piggyback the provincial agricultural income tax, motor vehicle tax, and even general sales tax on services, where the local government would set the rate (of additional tax) and the provincial government would collect the tax on its behalf. • A “citizen’s report card� (or similar measure) should be launched to assess the performance of each local government and to promote healthy competition among them. Completing decentralization therefore remains critical to establishing an accountable and efficient system—whether decentralizing from the federal to provincial level or from provincial to local. The logical completion of decen- tralization defined by the 18th Amendment is a viable and effective system of local government in each province. While the extent and form of devolution to the third tier needs to be driven by the reality on the ground in each prov- ince, to be effective local governments should be granted wide service delivery functions, commensurate finances, and a fair degree of autonomy. It is important that a bottom-up accountability system be established and given a chance to function. Ideally, this calls for local government to be accepted as a genuine third tier of government and to be given a role in and protection under the constitution. One lesson from Pakistan’s moves to decentralize is 253 that the main reason for the lack of effective local government is the mutual discomfort between provincial and local governments. Further, irrespective of political affiliations of the local government, the provincial government has always remained wary of the authority (and ability) of local governments, making every effort to obstruct their functioning. This invariably stymies full decentralization, generating worse than desired outcomes. Toward a pragmatic model of decentralized service delivery To establish a reasonably well-functioning decentralized system of governance and service delivery, it is critical to adopt a more pragmatic approach to bot- tom-up accountability. As mentioned, this would require a system capable of integrating political, administrative, and financial aspects of decentralization. The first step is to better align political priorities of provincial and local tiers of government, by: • Unbundling functions assigned to district governments by the 2001 LGO depending on their nature. District governments should be assigned local functions. In other words, provincial governments should adopt LGOs similar to the 1979 LGO.22 • Improving management and quality of key social services like education and health by devolving them to autonomous provincial authorities. For this: • Provincial governments would create these authorities, preferably through actions of provincial assemblies. • These authorities would be created at the markaz (sub-tehsil) level. • The geographic boundaries of the markaz would be defined to coincide with the provincial assembly constituencies. • These authorities would be governed by autonomous boards comprising local members of district councils, local professionals in these fields, and senior officers of education and health. • The boards would have complete autonomy over hiring, firing, trans- fers, and postings and over funds received as grants from the provincial government or raised by the board (perhaps by levying a fee, cess, or surcharge on delivery of the service).23 • The Member of Provincial Assembly (MPA) elected from each constitu- ency or markaz should be the chairperson of (all) these authorities. • The function delivery budgets, as well as the staff associated with these functions, should be transferred to these authorities, giving authorities full control over their management. promoting efficient service delivery with decentralization • Among provincial governments, setting up mechanisms for effective but nonintrusive monitoring of these authorities, based on pre-agreed bench- marks and targets. The results of such monitoring are to be made public so 254 that people can see each authority’s performance. Advantages and risks of the new model The biggest advantage of the new model is that it aligns the incentives among key stakeholders for a more transparent and accountable service delivery system. Reflecting the prevailing trend, all municipal and local services are to be delivered by elected local governments, but some key provincial services will be delivered by autonomous authorities, which would be headed by the elected representative of the provincial government. This approach would not only devolve the provincial service to a level where it could be much better managed but also, by making the MPA responsible for delivering these services, enhance the transparency and accountability of delivery. Under the current system, the MPA makes many important decisions on the delivery of education, health, and other services within his or her constituency but without getting any credit or being held accountable. Under the new model, the MPAs would be responsible for delivering these key services in their constituencies (the markaz), boosting their reelection chances if services are provided satisfactorily. The model aims to bring about better coordination between provincial govern- ment policies and the devolved service delivery system. The MPAs, who would now head the autonomous authorities in their constituencies, are likely to advocate for social (and other) services within the provincial assemblies much more forcefully than before. Similarly, they would be in a much better position than the district nazims, for example, to bring about changes in provincial policies supporting services in their constituency. As with any new system, the proposed model has several risks: • It gives MPAs almost full control over delivering key public services, making them vulnerable to capture by political and local elites. • The MPAs are members of provincial assemblies, with a key responsibility for framing provincial legislation. Such a time split (between legislative and service delivery responsibilities) may be detrimental to both. • Provincial departments and service delivery cadres are likely to show fierce resistance, which would require careful management. Specifically, the model limits appointments, transfers, and postings only to the markaz, thereby “depriving� staff from being transferred to more preferred jurisdic- tions, such as district headquarters or the provincial capital. • The model intends to decentralize services to a level that has never man- aged such services, creating obvious capacity concerns, especially in health. • As the service authorities are to get financial autonomy, the provincial financial management mechanisms (procurement, accounting) are likely 255 to be bypassed. This can potentially take large sums of public expenditure outside the public financial management system. The accounting and auditing parameters might have to be redefined to ensure effective account- ability of these authorities (see above). • An additional cadre in service delivery may raise administrative costs. Most of the risks should be mitigated by the bottom-up accountability realized through the electoral system, and the nonintrusive oversight by the provincial government would enhance the systemic accountability of these authorities, helping them achieve the objectives of this decentralized model. Notes 1. In 2003, through the 17th Amendment based on General Musharraf’s 2002 Legal Framework Order, it became mandatory to present money bills to the Senate, but only for review. 2. According to Burki (2010), the Concurrent List included functions on which the federal government made a promise during the process of draft- ing the 1973 Constitution that provinces would get control after a “period of political maturation,� defined as 10 years. 3. In 2006, the government inserted Clause 140A into the constitution, which made it incumbent on every provincial government to establish a local government system and devolve political, administrative, and financial authority to the elected representatives of the local governments. This ensures the existence of a local government system, but only through an action of a provincial government. Moreover, the degree of authority and responsibility to be devolved to the local governments is left to provincial governments. 4. The government announced the Devolution Plan first, which was later implemented through the 2001 LGO. 5. Tables 1 and 2 refer only to Punjab. Given that fiscal decentralization was much more advanced there than in the other provinces, the aggregate situation is likely to be even starker than the tables show. 6. The Sindh 2012 LGO has created a much starker divide between rural and urban areas than before. The six most urbanized districts have been given metropolitan status and will be governed similarly to those under the 2001 LGO, while the remaining areas of the province will have a system closer to that set up by the 1979 LGO. The provincial cabinet of Punjab has approved a draft LGO, similar to the 1979 LGO. promoting efficient service delivery with decentralization 7. The Basic Democracies of Ayub Khan in the 1960s, the 1979 LGO of General Zia-ul-Haq, and the 2001 Devolution Plan of General Musharraf were all designed to centralize power while giving local governments a 256 greater role in government. 8. For its part, the 18th Amendment greatly altered the assignment of functions between federal and provincial governments. For a more detailed assessment of the motivation for and successes and failures of the amendment, especially in service delivery, see Hussain and Kokab (2012) and Paul and others (2012). 9. In this case, provincial governments in general and the governments of Balochistan and Khyber Pakhtunkhwa in particular. 10. Including 10–12 percent on grants to provincial governments and state-owned enterprises and another 8–12 percent on subsidies. 11. Part of the reason for provincial governments’ inability to limit their recurrent expenditure stems from large and frequent increases in wages and pensions of government employees. With provinces having much larger payrolls than the federal government, the budget impacts of such wage increases are much stronger. 12. The manufacturing sector is taxed much more heavily than the agricul- ture and services sectors. However, within the manufacturing sector some industries (such as textiles) are taxed much more lightly than the others. 13. Our preliminary estimates show that for every PRs 100 that provincial governments receive as transfers from the federal government, expenditure increases by PRs 87, PRs 9 go to reduce the provincial fiscal deficit, and provincial revenue declines by PRs 4. 14. Some of the functions of the Concurrent List were shifted to Part I or Part II of the Federal Legislative List. 15. The 18th Amendment was a vehicle to operationalize the Charter of Democracy, which was agreed on by the two mainstream parties—the Pakistan People’s Party and the Pakistan Muslim League (N)—and some other smaller parties in London in May 2006. The Charter demanded, among other things: ending the presidential powers of dissolution of the National Assembly and appointment of governors, armed services chiefs, and the Chief Justice; abolishing the Concurrent List: issuing a new NFC award; expanding the Senate to give representation to minorities; integrating the Federally Administered Tribal Areas with North West Fron- tier Province; empowering Gilgit–Baltistan; lifting the ban on assumption of the office of prime minister for the third time; establishing a Truth and Reconciliation Commission and a National Democracy Commission; making the Inter-Services Intelligence, Military Intelligence, and other security agencies accountable to the elected government; removing indem- nities introduced by military governments; appointing the higher judiciary through a commission chaired by a chief justice who had never taken oath under the Provisional Constitutional Order; and establishing a Federal Constitutional Court with equal representation for all the federating units in order to resolve the constitutional issues. 257 16. Possible delays may stem from CCI decision making, as it has to come to a consensus on the decision and its modalities. 17. The amendment has tried to establish some check on provincial bor- rowing by mandating that the National Economic Council (NEC) set ceilings on the borrowing of each province. But given the strong presence of provinces in the revised composition of the NEC, the amendment may not be able to impose effective checks on provincial borrowing. 18. For the conceptual basis, potential benefits, and risks of decentralization, see World Bank (2001) and Ahmad and others (2005). For evaluation of the 2001 devolution in Pakistan, see World Bank (2004) and Keefer, Narayan, and Vishwanath (2006). 19. Some critical prerequisites for this decentralization are highlighted in Shah (2012). 20. Under the 18th Amendment, determining the ceiling on provincial government borrowing is the responsibility of the NEC. This recommen- dation implies that NEC decisions of provincial borrowing have to align with the national fiscal framework defined by the CCI. 21. There will be a role for the federation (CCI) or comparable provincial institutions to harmonize taxes subnationally, thereby prohibiting tax wars between subnational governments. 22. The provinces share a strong consensus to adopt the 1979 (or similar) LGO, suggesting that they are quite amenable to devolving municipal and local services to the local government. However, some key services devolved to district governments by the 2001 LGO, such as education, health, and agriculture, would revert to the provinces. 23. These revenue mobilization measures, however, have to be approved by the provincial government to ensure their consistency with provincial revenue measures. References Ahmad, Junaid, Shantayanan Devarajan, Stuti Khemani, and Shekhar Shah. 2005. “Decentralization and Service Delivery.� Policy Research Working Paper 3603, World Bank, Washington, DC. Burki, Shahid Javed. 2010. “Provincial Rights and Responsibilities.� The Lahore Journal of Economics 15: SE (September): 1–14. Burki, Shahid Javed, Guillermo E. Perry, and William R. Dollinger, eds. 1999. Beyond the Center: Decentralizing the State. Washington, DC: World Bank. promoting efficient service delivery with decentralization Cheema, Ali, Asim Ijaz Khwaja, and Adnan Qadir. 2006. “Local Government Reform in Pakistan: Context, Content and Causes.� In Decentralization and Local Governance in Developing Countries: A Comparative Perspective, eds. Pranab 258 Bardhan and Dilip Mookherjee. Cambridge, MA: MIT Press. Hasnain, Zahid. 2010. “Devolution, Accountability, and Service Delivery in Pakistan.� The Pakistan Development Review 49 (2): 129–52. Hussain, Mahboob, and Rizwan Ullah Kokab. 2012. “Eighteenth Amendment in the Constitution of Pakistan: Success and Controversies.� Asian Social Science 8 (1): 81–88. Keefer, Philip E., Ambar Narayan, and Tara Vishwanath. 2006. “Decentral- ization in Pakistan: Are Local Governments Likely to Be More Account- able than Central Government?� In Decentralization and Local Governance in Developing Countries: A Comparative Perspective, eds. Pranab Bardhan and Dilip Mookherjee. Cambridge, MA: MIT Press. Paul, Jawad, Shahid Zaman, Naveed Saleh Siddiqui, Moshan Atta, and Qamar Mustafa Shah. 2012. “Strategic Appraisal of 18th Amendment Federal/ Provincial Roles and Impact on Service Delivery.� Prepared by Senior Management Wing of the National Management College, National School of Public Policy, Lahore, Pakistan. Salamat, Zarina. 1992. Pakistan 1947–58: An Historical Review. Islamabad: National Institute of Historical and Cultural Research. Shah, Anwar. 2012. “Making Federalism Work—The 18th Constitutional Amend- ment.� Policy Paper Series on Pakistan Pk 03/12, World Bank, Washington, DC. Shah, Anwar, Teresa Thompson and Heng-Fu Zou. 2004. “The Impact of Decentralization on Service Delivery, Corruption, Fiscal Management and Growth in Developing and Emerging Market Economies: A Synthesis of Empirical Evidence.� Journal for Institutional Comparisons 2 (1): 10–14. Waseem, Mohammad. 2010. “Federalism in Pakistan.� Lahore University of Management Sciences, Lahore, Pakistan. World Bank. 2001. �Decentralization and Governance: Does Decentralization Improve Public Service Delivery.� PREM Notes 55. Washington, DC. ———. 2003. World Development Report 2004: Making Services Work for Poor People. Washington, DC. ———. 2004. Devolution in Pakistan. Washington, DC. Jose R. Lopez-Calix and Irum Touqeer Mobilizing Revenue 16 This note reviews key shortcomings in Pakistan’s revenue mobilization system and provides directions for revitalizing it and raising collection by 3–4 percentage points of GDP over the next five years. Pakistan has one of the world’s lowest tax ratios, stemming from five main weaknesses: complexity, a narrow tax base, low compliance, inefficient tax administration, and low and declining provincial tax revenues. Complexity provides scope for discretion and corruption. A narrow tax base and low compliance are the outcomes of inequitable exemptions and preferential treatments, low tax registration or filing, and massive tax evasion by potential taxpayers that prefer to stay informal. Provincial taxation is low and declining. For its part, nontax revenue is also declining. Proposed reforms aim for a tax system that is broad, simple, and equitable, that facilitates taxpayer registration and compliance, and that promotes provincial revenue efforts commensurate with their new expenditure responsibilities. This implies implement- ing effective tax policy and administration, particularly eliminating exemptions and zero rates to broaden the base, adjusting income tax rates, simplifying tariffs, expand- ing user-friendly electronic registration and filing, enforcing a zero-tolerance policy for noncompliance and evasion, and overhauling the technical capacity and accountability of Federal Board of Revenue staff, especially in information technology systems, auditing, and enforcement. At the provincial level, this implies introducing incentives for collecting provincial taxes, enhancing the capacity of tax administration, and updating selected rates. In this regard, broadening the general sales tax on services, collecting the motor vehicle tax more efficiently, and revamping the property tax would be desirable. One of Pakistan’s major challenges is to expand government revenues. The economy is structurally weak on the revenue front, as evidenced by large and recurrent fiscal deficits, financed with loans from commercial banks or the State Bank of Pakistan, which until recently contributed to double-digit inflation. Rev- enue ­ collection—at 12.5 percent of GDP in 2010/11—is much lower than aver- ages for the world, South Asia, and emerging and developing countries (Figure 16.1). The tax to GDP ratio, declining from a 10.8 percent average in the 2000s to a bottom 9.6 percent in 2010/11, is now among the world’s lowest. The author- ities need to address this to raise public investment and secure stronger growth and development. Indeed, public investment also fell, from 5.6 percent of GDP in 1999/2000 to 2.6 percent in 2010/11. The public debt burden is above the upper limit allowed by the Fiscal Respon- sibility Bill, and the fiscal space is shrinking. With inadequate tax revenues, to mobilizing revenue Figure General government revenue, 2000/01–2011/12 16.1 Afghanistan Bangladesh Bhutan India Maldives Nepal Pakistan Sri Lanka World Emerging and developing economies 260 40 Percent of GDP 30 20 10 0 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 (estimated) 2011/12 (projected) Source: World Bank 2012a. create fiscal space the government has relied increasingly on high nontax reve- nues (Figure 16.2), but since 2009 this source has also declined—due to falling State Bank of Pakistan profits, defense receipts (coalition support funds from the U.S. government for military expenditures), and dividends from state- owned enterprises.1 The wide fiscal deficits reached 8.5  percent of GDP in 2011/12, the financing of which added to a public debt burden that exceeded the 60 percent of GDP ceiling allowed by the country’s Fiscal Responsibility Bill (Figure 16.3). More important, despite a benevolent negative differential between real interest rates and growth that points to a decline in the real value of debt given currently high inflation levels, the rising primary deficit endan- gers debt sustainability. The country’s medium-term fiscal space depends less on declining nontax revenue and more on mobilizing additional revenue (with a goal of 3–4 percentage points of GDP in the next five years) to finance the country’s development agenda. The buoyancy of the country’s tax system is low and declining. Tax buoyancy (for taxes collected by the Federal Board of Revenue, or FBR) decreased from above unity in the 1960s to 0.93 in the 2000s, indicating that tax revenues are growing more slowly than GDP (World Bank 2009, 2012c).2 Tax revenues rely mainly on federal taxes, and much less on provincial taxes, but the overall level of federal taxes is low, despite their changing mix. Federal taxes accounted for about 95 percent of total tax revenues in 2010/11, with minimal contributions from provincial taxes—0.4 percent of GDP in 2010/11 (Figure 16.4, panel a).3 The country has, though, moved gradually from trade Figure Pakistan’s tax and nontax revenue position 16.2 Pakistan (1999/2000–2009/10) Pakistan (2010/11) 261 Pakistan (2011/12 estimate) India (2000–10) South Asia (2000–10) Latin America (developing, 2000) World (2001–10) 5 7 9 11 13 15 Tax to GDP ratio (percent) 2.0 Nontax revenue buoyancy Pakistan Bhutan 1.5 Afghanistan Nepal Maldives Bangladesh 1.0 India Sri Lanka 0.5 0.0 0 2 4 6 8 10 12 14 Nontax revenue (percent of GDP) Note: Pakistan’s tax ratio refers to federal revenue. Nonfederal tax revenue is about 0.5 percent of GDP. Buoyancy was calculated using log regression from data spanning a number of years (number of years varied for countries and was based on data availability). Nontax revenue as a share of GDP was based on the latest available GDP or estimates, in most cases 2010. Source: World Bank 2012a; International Monetary Fund database; World Bank staff calculations based on World Bank (2012b). Figure Declining revenues and rising deficits, 2000/01–2011/12 16.3 Overall scal de cit Nontax revenue Tax revenue 15 12 Percent of GDP 9 6 3 0 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 (estimated) Source: International Monetary Fund database. mobilizing revenue Figure Stylized facts about Pakistan’s taxation effort 16.4 a. Taxes by administrative division b. Direct and indirect taxes Total tax revenue Federal tax revenue Provincial tax revenue 1981–90 1991–2000 2001–11 262 12 Trade taxes 9 Income taxes Percent of GDP Sales tax (domestic) 6 Excise duties (domestic) 3 Surcharges Others 0 1999/ 2001/ 2003/ 2005/ 2007/ 2009/ 2011/12 0 2 4 6 2000 02 04 06 08 10 (estimate) Percent of GDP c. Indirect taxes d. Share of FBR taxes Sales tax Custom duties Sales tax Direct taxes Excise duties Custom duties FED 1999/2000 1999/2000 2000/01 2000/01 2001/02 2001/02 2002/03 2002/03 2003/04 2003/04 2004/05 2004/05 2005/06 2005/06 2006/07 2006/07 2007/08 2007/08 2008/09 2008/09 2009/10 2009/10 2010/11 2010/11 2011/12 2011/12 (estimate) (estimate) 0 0 2 4 6 8 25 50 75 100 Percent of GDP Percent e. Direct taxes f. Widening gap between budget revenue targets and actuals Corportate income tax Tax (target) Tax (actual) Individual income tax FBR (target) FBR (actual) 12 2000/01 2001/02 9 Percent of GDP 2002/03 2003/04 6 2004/05 2005/06 3 2006/07 2007/08 0 0 1 2 3 4 2006/ 2007/ 2008/ 2009/ 2010/ 2011/12 Percent of GDP 07 08 09 10 11 (provisional) FBR is Federal Board of Revenue; FED is federal excise duty. Source: Lopez-Calix and Touqeer forthcoming; World Bank 2009; International Monetary Fund database; World Bank staff calculations based on Government of Pakistan (2012b,c) and FBR (2012). and excise taxes toward income and sales taxes (panels b, c, and d). However, further efforts are needed to mobilize revenues from corporate and individ- ual income taxes, as well as the sales tax. Moreover, direct taxes are skewed heavily toward corporate income tax (CIT)—which provides disincentives to 263 ­ rofitable—and biased away from firms to increase investment and become more p personal income tax, whose revenues are dismal (panel  e). Overall, federal revenue collection remains quite low, and the budgeted revenue targets have been missed in recent years (panel f). Main Issues Pakistan’s tax system underperforms because of a complicated and unfriendly taxpayer system, a narrow tax base, low compliance, weak and inefficient tax administration, and inadequate subnational taxation. Complexity The taxpayer system is complex. Most of the country’s revenues—customs and general sales tax (GST)—are generated by trade. Customs regulatory duties are an example of complexity (as are sales and income taxes). The general- ized use of statutory regulatory orders (SROs) introduces wide deviations between applied and statutory official rates on an extremely disperse range of tax slabs. The 2010/11 tariff schedule, for example, had 17 slabs (16 plus zero rate) under most-favored nation statutory official ad valorem tariffs, ranging from 0 to 150 percent and with the greatest frequency (about 40 percent) on or below the low tariff slab of 5 percent. Yet Pakistan effectively applies some 40 tariff slabs and, including SROs and partial exemptions, around 55 percent of the effective tariffs are 5 percent or less in ad valorem terms (Figure 16.5). About 2 percent of trade tariffs are considered nuisance taxes (0–2 percent ad valorem), as they have marginal revenue, have high administrative costs, and are prone to corruption. Moreover, complexity also creates an anti-ex- port bias: a 1 percent increase in tariff “complexity� (defined by a measure of the number of changes in tariff brackets) leads to a 13.2 percent decrease in export growth (Reis and Taglioni 2013).4 Finally, as developing countries enter into trade liberalization that leads to lower tariff duties, they generally recover as little as a third of the customs duty revenue loss with increased domestic consumption taxes (Baunsgaard and Keen 2005). The complexity makes it harder and costlier for businesses to pay taxes. Paying taxes is the second biggest obstacle to doing business in Pakistan, after access to electricity (World Bank 2012d). Of 183 countries, Pakistan ranks low at 158, or lower than most South Asian countries. Tax-compliance costs are mobilizing revenue “Most-favored nation� statutory and effective tariff rates by Figure frequency, 2009/10 16.5 Distribution of most-favored nation tariff rates 40 264 30 Frequency (percent) 20 10 0 0 5 10 15 20 25 30 35 40 45 50 90 150 Speci c Protection rate (percent) Distribution of effective tariff rates 30 Frequency (percent) 20 10 0 0 2 5 7.5 10 15 20 25 30 35 40 50 60 75 90 >100 Protection rate (percent) Note: Effective tariff rates include regulatory duties. Source: Reis and Taglioni 2013. also very high, with firms taking about 560 hours (about 14 weeks) to make 47 tax payments a year. This is almost double the South Asian average (311 hours a year), and three times the Organisation for Economic Co-operation and Development average (176 hours a year; Table 16.1). Matters are undoubt- edly further aggravated by frequent and ad hoc changes to tax laws (Kularatne and Lopez-Calix 2012). A narrow tax base Key economic activities, like agriculture and services, are barely taxed, despite their large share in total output; and those taxed, like industry, enjoy significant Table Paying taxes indicators, 2011/12 16.1 Organisation for Economic Co-operation Indicator Pakistan South Asia and Development countries Payments (number per year) 47 30 12 265 Time (hours per year) 560 311 176 Total tax rate (percent of profit) 35.3 40.2 42.7 Source: World Bank 2012d,e. exemptions or concessions. This is the outcome of structural factors, leakages, and low registration. Industry bears most of the tax burden: relative to its share in GDP (25 percent) it pays about 60 times more than agriculture (21 percent of GDP) and 5 times more than services (54  percent of GDP). Even within industry, some sectors have large tax exemptions or are undertaxed, including textiles, fertilizers, and pharmaceuticals. The implementation of the GST also distorts tax collection, by following a system of discounts and exemptions (Table 16.2). Beyond hurting large businesses, this system forfeits government revenues, partly because the prefer- ential tax rate for small companies5 (and overregulation) induces firms to stay small or informal.6 The narrow tax base also stems from low registration. On GST registration, a reason for low GST collection is also the small and declining number of registered taxpayers among retailers and service providers. Indeed, for the estimated 210,000 registered taxpayers at the end of 2011/12 (manufacturers being the largest group), from 2009/10 to 2011/12 annual growth in GST registrations fell from 60 percent to 4 percent (Figure 16.6). More broadly, on national tax registration, there are barely 3.1 million holders of national tax numbers (NTNs). And though 47,800 companies have NTNs, this is a very low percentage of the 400,000 industrial electricity connections (besides, Effective and nominal tax rates (general sales tax) for selected Table sectors 16.2 (percent) Cotton Carpets Sports Surgical Printing and Bidis ginning and rugs Pharmaceuticals Footwear goods instruments publishing Glass Cement Nominal 0.0 0.0 0.0 0.0 1.4 0.0 0.0 0.0 20.1 12.1 Effective 6.5 2.5 9.1 16.8 7.3 6.7 5.5 8.0 40.0 27.1 Source: Ahmad 2010. mobilizing revenue Figure Trends in tax registration 16.6 General sales tax registration, 1997/98–2011/12 Number of registered personsa Growth 225 80 266 150 40 Thousands Percent 75 0 0 -40 2010/11 2011/12b 2000/01 2001/02 2003/04 2004/05 1997/98 1998/99 1999/2000 2002/03 2009/10 Sales tax registration, by activity (until 2011/12) Others 6% Exporters 7% Manufacturers 29% Service providers 11% Importers 21% Wholesalers and retailers 26% a. Cumulative registration of persons. b. Data include actual figures until February 21, 2012, and forecast for remaining period. Note: In 2004/05, the sales tax base was increased (for retailers and manufacturers by PRs 8.5 million) and records of nil filers were separated, cutting numbers of registered persons. Source: World Bank staff calculations based on Ahmed and Ahmed (2012) and Federal Board of Revenue data. fewer than 16,500 filed tax returns). Similarly, NTN issuance to the four most important taxpayers fell from 157,030 in 2008/09 to 96,845 in 2011/12. These taxpayers are business individuals, companies, and associations of persons, accounting for about 43 percent of NTN holders in 2011/12. Only the final category of taxpayers—salaried individuals—recorded a propor- tional increase during this period (Figure 16.7). As a result, most corporate income and trade taxation comes from a few large corporations and manu- facturing firms and the imports of a small group of commodities (the top 10 commodities contribute about 81 percent of import taxes). Similarly, income tax rates are high and have some space for reduction. Pakistan falls in the category of countries with high rates, at 35 percent, worldwide (Figure 16.8).7 New national tax numbers issued to different categories, Figure 2008/09–2011/12 16.7 Salaried individuals Company Association of persons Business individuals 2008/09 267 2009/10 2010/11 2011/12a 0 25 50 75 100 Percent a. Data are forecasts based on actual figures until February 21, 2012. Source: World Bank staff calculations based on Ahmed and Ahmed (2012) and Federal Board of Revenue data. Figure Corporate income tax rates worldwide, 2009/10 16.8 Low (0%–20%) Medium (21%–30%) High (31%–46%) 50 Bahrain Chad Libya Seychelles 40 Bangladesh India Pakistan (35%) Argentina Percent Australia Bhutan Indonesia Philippines 30 United Brazil China Italy Malaysia Kingdom Luxembourg 20 Nepal Turkey Albania Cyprus 10 Switzerland 0 Source: USAID 2011. Low compliance Pakistan has a very poor tax filing record. Less than 1 percent of Pakistan’s population files for taxes, well below 5 percent of India or 16 percent of Argen- tina (Ahmad and Best 2012). About 70 percent of legislators do not file income tax returns (CIRP and CPDI 2012). Taxpayers evade taxes by simply not filing tax returns or by paying low taxes due to special privileges obtained through legal concessions. For instance, many GST-registered taxpayers do not pay taxes. Of the total registered persons (114,953) in 2004/05, only 70 percent filed sales tax returns, and those with the lowest compliance (wholesalers and retailers, at 45.4 percent) had the highest sales tax registration (Ahmed and Ahmed 2012). As a result, tax payments are concentrated among few taxpayers. In 2007, about 90 percent of GST was paid by only 3 percent of taxpayers (Ahmad 2010). Tax mobilizing revenue reforms improved compliance over 2008/09–2010/11, as electronic return fil- ers increased 18 percent for sales tax and 58 percent for income tax, and as registered and active tax payers for income tax and sales tax increased 29 per- 268 cent and 13 percent, respectively (Figure 16.9). Yet tax filing in 2010/11 was still low: only about 43 percent of companies, and 25–28 percent of business individ- uals and associations of persons filed tax returns; salaried individuals had the highest compliance, at 68 percent. Overall, 1.5 million taxpayers filed income tax returns out of about 3.1 million NTN holders (Ahmad and Best 2012). Pakistan’s low compliance is reflected by one of the world’s worst tax productiv- ity records. Low tax compliance is seen in low and declining GST productivity. Under the best possible case, the GST (value added tax–like) productivity ratio would approach unity.8 Worldwide estimates for 2010 show that Cyprus is closest to one, with a ratio of 0.8, but that Pakistan is in the lowest bracket at 0.2— below Nepal and Sri Lanka (0.3) and similar to the Philippines and Turkey, among others (Figure 16.10, top panel). Further, the ratio has been declining— especially since the 2008/09 global economic crisis—and it has declined faster than the FBR tax to GDP ratio (Figure 16.10, bottom panel), partly because of the recent economic shocks and partly because of poor enforcement of tax collection. Pakistan also has one of the lowest GST C-efficiency indexes (0.27 in 2009) in the world.9 Low compliance is also reflected in big tax gaps. A measure of tax evasion, the tax gap marks actual versus potential revenues if everyone complied with tax laws; and recent estimates confirm Pakistan’s low (and decreasing) tax compliance. In 2004/05, for example, the sales tax gap was estimated at PRs  45 billion, or about 31 percent of collections; by 2010/11, the gap had increased Figure Pakistan’s tax filing record, 2008/09–2010/11 16.9 Federal Board of Revenue Filers Active taxpayers 120 Percent of general sales tax 110 100 90 2008/09 2009/10 2010/11 Source: World Bank 2012c. Figure General sales tax and value added tax productivity 16.10 0.8 Value added tax productivity 0.6 269 0.4 Nepal Sri Lanka 0.2 Bangladesh Pakistan 0.0 2.5 3.0 3.5 4.0 4.5 5.0 GDP per capita, 2010 (PPP, log scale) General sales tax Federal Board of Revenue tax to GDP ratio 120 General Sales Tax Productivity Index 110 (2004/05 = 100) 100 90 80 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 Note: Value added tax (general sales tax) productivity versus GDP per capita figure relates to 2010 data. Source: World Bank 2012b,c; USAID 2011. to PRs 152.4 billion, or about 39 percent of collections. Services, mining, and manufacturing show large tax evasion, while agriculture shows minor evasion (Table 16.3). Some of the widest tax gaps by subsector are in cigarettes, paper and printing, chemical products, cement, electricity, retail trade, and hotels and restaurants. Other low-compliance subsectors include sugar, pharma- ceuticals, and fertilizers (Ahmed 2011). Tax gaps are related directly to lax enforcement and corruption that goes undetected or, if detected, unpunished. Tax expenditure is high and rising. Tax expenditure is the revenue loss due to preferential legal provisions (in the finance bill or through ad hoc SROs) in the tax laws that provide certain taxpayers with special concessions—zero or reduced rates and tariffs or duty and tax exemptions that are not available to other taxpayers or sectors and that result in the collection of fewer tax revenues than would be collected under the basic tax structure. Tax expenditure increased from 0.8 percent of GDP in 2000/01 to 0.9 percent in 2007/08 (World Bank mobilizing revenue Table Sales tax gaps by sector, 2010/11 16.3 (PRs billion) 270 Net sales tax Actual collection Tax gap Agriculture 0.00 –2.87 2.87 Mining 69.26 17.78 51.47 Manufacturing 288.20 257.88 30.31 Services 187.47 119.78 67.68 Total 544.92 392.58 152.34 Source: Ahmed 2011. 2009) and then, according to Ather (2013), doubled to 1.8 percent in 2010/11. This estimate should be considered as a lower bound, as significant exemptions in income tax were not included in these estimates. Results show that the highest losses are in customs duties (Table 16.4). For its part, the services sector exhibits large tax exemptions, with its tax expenditure amounting to PRs 82.5 billion, and the transport subsector represents about half such amount (Ather 2013). Inefficient tax administration The big tax gap and low productivity ratios are symptoms of weak tax adminis- tration. Pakistan’s tax administration is constrained by poor management, low capacity due to weak human resources, and a lack of effective key supporting information technology (IT) systems, which all together provide enough scope for discretion and corruption. The incidence of bribes paid to tax officers is high, particularly by large firms (World Bank 2009). The Global Corruption Barometer 2010/11 finds that at least two of every five surveyed households have bribed tax revenue authorities to escape paying taxes (Transparency International 2012). According to the government’s National Anti-Corruption Strategy, tax collection losses by corruption were highest in the corporate and Table Tax expenditure, 2010/11 16.4 (PRs billion) Tax category Amount Income and corporate tax 126.01 Sales tax import 58.40 Sales tax domestic 89.57 Custom duties 128.06 Total 402.04 Source: Ather 2013. personal income taxes (64 percent), followed by the customs (48 percent) and sales taxes (45 percent; National Accountability Bureau 2002). The government implemented some reforms during the mid-to-late 2000s but 271 failed to achieve most of its desired outcomes. Noteworthy reforms included improving the FBR’s physical and IT infrastructure, establishing an online tax registration system, establishing Large Taxpayer Units and Medium Taxpayer Units, attempting to provide some stability to the tenure of senior management,10 moving the FBR under the oversight of a Cabinet Committee on Finance and Revenue, approving a previously nonexistent human resource management policy framework, and preparing a rationalization plan for nonessential FBR staff. The FBR Act 2007 was enacted, supported by sufficient funding for restructuring the organization. As a result of these capacity-build- ing reforms, services to taxpayers improved initially. However, the following years showed slow and piecemeal implementation impairing the consolidation of tax administration reforms for the following reasons (World Bank 2012c): • The FBR board and management have changed frequently due to political interference. In 2012, the FBR had four chairpersons, and of the 14 board members that started the year, only 3 remained at the end of the year. • Integration of the FBR along functional lines has moved too slowly and remains incomplete due to resistance from staff. Staff complaints include a lack of IT and managerial staff support, ineffective monitoring and evaluation mechanisms from FBR headquarters, and frequent rotation of senior and mid-management at decentralized Large Taxpayer Units and Regional Taxpayer Units. • New IT-based systems, essential for effective cross-checking enforcement and audits and focused on large taxpayers, have remained underused due to their poor integration in business processes, weak governance and partial use by senior management, lack of training, and staff resistance to adopting new technologies, which would limit their opportunities for bribes and discretion. • The audit function has remained very weak due to a lack of an effective, centralized, parameter-based risk-audit function (supported by solid planning, staffing, and monitoring of results) and to poor training, leading to unfavorable results for the FBR on legal disputes. • There was a lack of political consensus on the approval of the government’s reformed GST (RGST) by the Parliament. This central policy decision slowed momentum for reform. Low and declining subnational tax revenue The revenue effort by provinces is extremely weak, is worsening, and falls well short of their new expenditure responsibilities under the 18th Amendment to mobilizing revenue the constitution approved in 2010. In the last two decades, the provincial tax to GDP ratio has oscillated between 0.35 and 0.55 percent of GDP but on a declining trend (Figure 16.11, top panel). In 2011/12, it stood at 0.5 percent 272 of GDP, whereas its share in national tax revenue was 5 percent. The falling trend and inadequacy of provincial revenue collection has become a serious concern because it puts greater stress on scarce federal resources in a con- text of expanded outlays associated with the new devolved functions from the 18th Amendment toward social sectors, infrastructure spending, and the like. In 2011/12, all provinces spent about nine times the revenue they collected from tax and nontax sources (provincial revenue accounted for 11.4 percent of expenditure in 2011/12); in 2007/08, spending was six times revenues (Figure 16.11, bottom panel). These gaps are a problem for all of them (Figure 16.12, top panel). Among the four provinces, Sindh was the largest collector (16  percent of its revenues were generated by taxes in 2011/12), followed by Punjab (7.1  percent), Khyber Pakhtunkhwa (1.7  percent), and Figure Trends in overall provincial tax revenue 16.11 0.6 Percent of GDP 0.5 0.4 0.3 1987/88 1990/91 1995/96 2000/01 2005/06 2011/12 (provisional) 20 15 Percent of expenditure 10 5 0 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 (provisional) Source: World Bank staff calculations based on Provincial Government Fiscal Operations (various years). Figure Provincial revenue position and composition 16.12 Provincial revenue, 2010/11 and 2011/12 2010/11 2011/12 273 Sindh Punjab Balochistan Khyber Pakhtunkhwa 0 5 10 15 20 Percent of expenditure Provincial revenue composition, 2011/12 Share in federal revenue Federal loans and grants Provincial nontax Provincial taxes Sindh Punjab Balochistan Khyber Pakhtunkhwa 0 25 50 75 100 Percent Note: Provinces’ own revenue includes provincial tax and nontax revenue. Source: World Bank staff calculations based on Government of Pakistan (2011, 2012c). Balochistan (0.8 percent; Figure 16.12, bottom panel). Provincial taxes are even smaller than federal grants and loans from the federal government, which also creates disincentives for provincial governments to increase their revenue.11 Stark interprovincial structural disparities lead to different tax bases and expenditure needs. Punjab and Sindh have greater potential to raise tax revenue, given their higher income per capita, while Balochistan and Khyber Pakhtunkhwa have greater social needs. According to the latest reliable estimate (2004/05), Sindh had the highest estimated income per capita at PRs 6,900, or 1.3 times the national average, followed by Punjab at PRs  5,400 (about the national average). Balochistan and Khyber Pakh- tunkhwa had incomes about half that of Sindh. Unsurprisingly, Balochistan and Khyber Pakhtunkhwa featured the highest poverty rates in 2004/05 mobilizing revenue (Lopez-Calix and Touqeer 2013). Thus, from a social perspective, Balochistan and Khyber Pakhtunkhwa would need to make a greater revenue effort and collect proportionally more revenues than the other two. 274 Different tax composition and administrative obstacles keep provincial tax bases low. Tax composition differs across provinces (Figure 16.13). Obstacles to tax collection include only a few, low-revenue sources; con- stitutional restrictions of federal policy (provinces can levy GST only on professions, trades, and callings while tax exemptions and preferential treat- ment make it hard even to reach the available tax base); and low rates on agricultural income, stamp duties, and capital gains and capital value taxes on immovable property. Issues facing three provincial taxes are as follows: • Sales tax on services. The main issue is its narrow base for only a few services. Poor incentives deter the federal government from expanding this tax base, keeping revenue low. In 2011/12, Punjab and Sindh started collecting this tax, and early results are encouraging. The provinces are interested in expanding the tax base by expanding to other services, such as the retail trade. • Motor vehicle tax. Collection of this tax remains overly cumbersome, and it imposes heavy compliance costs on payers. It has to be paid quarterly, and it is levied on all vehicles based on unladen weight for motorcycles, seating capac- ity for cars and buses, and laden weight for trucks. The many exemptions include agriculture tractors and trailers, ambulances, school buses, noncom- mercial government vehicles (including National Logistics Cell vehicles), and vehicles used by foreign missions. Rates vary across provinces (and federally Figure Provincial tax revenue by source, 2011/12 16.13 Property tax Excise duties Stamp duties Motor vehicle tax Other Sindh Punjab Balochistan Khyber Pakhtunkhwa 0 25 50 75 100 Percent Source: World Bank staff calculations based on Government of Pakistan (2012c). administered areas), providing incentives for vehicle owners to underpay the tax by registering the vehicle in low-tax areas and using it elsewhere. • Urban immovable property tax. Despite rapid urbanization and a sharp escalation in capital and rental values of urban properties, the revenue 275 from the urban immovable property tax is extremely low due to exemp- tions, underassessment of property values, rate differentials (leading to tax evasion), and weak administration. This tax is levied only on owners of build- ings and lands in urban areas,12 and properties of government, religious parties and their affiliates, charities, and education institutions are exempt. Other exemptions include properties of cantonment areas and segments of the population especially vulnerable to adverse shocks (such as the poor, widows, orphans, and retired government employees). Huge underas- sessment of rental values of taxable properties13 stem from infrequent revision of valuation tables, which are based on a flawed formula.14 (Property valuations are made annually based on property rental values.)15 Tax evasion is heightened by different rates between owner-occupied and rental proper- ties (the latter charged a rate about 6 times that of the former) and between industrial and commercial properties (the latter charged a rate about 10 times that of the former). Finally, administration is affected by confusion around jurisdiction of the tax.16 Policy Options If we separate nontax revenue as part of a different agenda, successful additional revenue mobilization for 3–4 percentage points of GDP in the next five years will have to walk on two legs: tax policy and administration. Desir- able features of tax policy are well known (Box 16.1). The ultimate objective is to raise revenues based on a system that is simple and predictable, encourages investment, and facilitates taxpayer compliance. This usually implies employ- ing an active tax policy, broadening the base by eliminating exemptions, simplifying rates and tariffs, offering user-friendly electronic filing, per- mitting zero tolerance for noncompliance, and strengthening the technical capacity and accountability of tax administration (that is, the FBR). At the provincial level, for the relevant issues this implies introducing incentives for provincial taxes (linked to fiscal transfer mechanisms), enhancing the capacity of tax administration, and updating selected rates. Broadening the GST on services, enhancing collection efficiency of the motor vehicle tax, and revamping the property tax would also be desirable. Adopting this pro- posed mix of policy and administration at both the federal and provincial levels would produce a simplified tax system that is broad-based, efficient, and effective for addressing Pakistan’s revenue mobilization needs (now discussed in greater detail). mobilizing revenue Box Desirable features of tax policy 16.1 In lieu of an optimal tax structure, international best practice suggests the following benchmark pat- terns of tax policy: 276 • Minimize exemptions and tax incentives that jeopardize revenue and good governance and generate no clear, offsetting social benefit. • Remove minor taxes and fees that are costly to administer. • Build CITs that are simple, broad-based, and competitive by international standards and set effective tax rates that are reasonably low and uniform across investments. A single statutory rate is recom- mended. Adopt an accelerated depreciation schedule. • Extend the coverage of personal income tax with an effective rate structure that is consistent with the authorities’ distributional preferences, while keeping the effective maximum personal income tax rate equal or close to the uniform CIT effective rate. Keep the rate structure simple. • Replace the production tax or sales tax with a simple value added tax (VAT) that has a broad base and a high threshold—minimizing exemptions and the number of rates, preferably to one nonzero rate. • Lower trade tax rates that have fewer bands and replace the loss with domestic sources (for example, VAT). Eliminate export taxes. Source: World Bank 2012a. Activate tax policy and broaden the tax base This reform aims to increase collection (tax policy) and buoyancy (broaden- ing) by removing exemptions. The latter agenda covers the sales tax, income tax, and customs duties. The decision about the final mix of these measures would involve political considerations. • Tax policy. Publishing a tax expenditure annex in the annual budget would identify and facilitate gradual removal (under a sunset clause that takes up to three to five years) of most tax exemptions (except food, medical supplies, and the like) and zero ratings, along with their projected fiscal impact (while possibly leaving only those protected by the constitution). Creating a joint FBR-Ministry of Finance unit on tax policy would support sound policy design and solid revenue forecasting. • Sales tax. The goal is to have a simple broad-based and nationally integrated RGST on goods and a provincially harmonized RGST on services. To fill exiting leaks that affect compliance, serious consideration should be given to eliminating most exemptions and zero rates that would have an impact equivalent to about 1.8  percent of GDP. Alternatively, zero rates could be converted into exemptions for domestic sales of main exports. Exemptions on special regimes (particularly preferential trade agreements) should be scrutinized. • Corporate and individual income taxes. In the medium term, the high CIT needs to be lowered from 35  percent to a 25–30  percent international benchmark and its base widened with new registrations, as current coverage is low. A lower CIT would encourage investment and attract new businesses to file. Similarly, the individual income tax requires a massive registration and filing effort. Continually simplifying tax returns and 277 streamlining personal tax credits would also encourage tax compliance. • Other measures. A complementary but not exhaustive agenda for the government includes reintroducing special excise duties; introducing a retail tax similar to that introduced successfully by China, which favors lot- tery tickets for new registered sales taxpayers; bringing in a capital gains tax on property transfers; and increasing levies like those on petroleum and gas. Simplify taxes and make compliance effective and noncompliance expensive A simple tax system that does not rely on only a few people or sectors for revenues and that has low rates encourages voluntary compliance and reduces the incentives for tax evasion. Recent improvements in tax admin- istration allowing a high percentage of e-filers are encouraging, but further gains are needed: • For simplicity, make an immediate freeze in issuing the distortive trade, income, and GST-related SROs, and ensure the passage of a clear schedule of gradual elimination (cleansing) of the remaining SROs in three years; fully revamp customs duties by consolidating effective tariffs into fewer slabs (ideally three), while reducing tariff peaks to a predefined ceiling; streamline special tariff regimes for selected industries (such as automobiles and pharmaceuticals); and complete the gradual phasing out of the negative list with India in the first half of 2013, while allowing for a few still applicable, well-justified but temporary exceptions (list under preparation). • For strengthening compliance, carry out effective audits of a signifi- cant share (30–50 percent) of large taxpayers selected through parame- trized risk-based criteria, with quantitative performance benchmarks to attain; complete the audit of withholding agents and eliminate illegal input adjustments; and adopt a zero-tolerance policy against tax evasion detected through audits by penalizing nonfiling of or understatement in tax returns. Modernize tax administration Stability in tenure of senior tax managers, investment in key soft infrastruc- ture (IT) and qualified human resources, and governance improvements are all urgently required. These actions have been on the tax reform agenda for mobilizing revenue years, if not decades. Their success depends largely on the decision power— and sustained implementation c ­ apacity—of the political leadership. • Consider presenting a bill to the Parliament converting the FBR into a fully auton- 278 omous institution. This is consistent with best practices worldwide. It would help prevent political interference, foster accountability, and support its move into a performance-based institution. • Fully integrate automated computerized systems. As a myriad of different sys- tems work at various levels, functioning in “silos,� they are inefficient, disconnected, and unreliable. Upgrading and effectively integrat- ing the current IT software and databases (to facilitate multiple taxes cross-checking and data exchanges with NADRA and other national data- bases—credit cards, banking accounts, and so on), providing IT equip- ment to field offices, and making special arrangements for safe data storage for the interrelated systems would be beneficial. It would also help to improve e-filing, make the refund process more reliable, improve the auditing function by field offices, and facilitate their tasks by prepar- ing the long due tax ledger, using computerized discrepancies to cross- check or detect potential tax evaders and noncompliers. • Invest in human resource capacity building. Governance effectiveness of the automated IT-supported tax system and its piloting require massive training for senior and mid-level staff both at Large Taxpayer Units and Regional Taxpayer Units, in parallel to the new system’s design. Some accreditation mechanism should be considered. • Improve management and human resource policies . This implies improve- ment in the FBR’s human resource management policies, which should include new job descriptions, hiring policies, and performance, merit, and integrity criteria for staff evaluation. Bonuses should be attached to function-specific targets to evaluate job performance. Managers should be made accountable in enforcing codes of conduct across the institution. • Strengthen the FBR along functional lines. Its organization should strengthen performance reporting and monitoring tools and procedures per function. Preparing an annual action plan as a management tool includ- ing key performance indicators to be regularly assessed (perhaps monthly or quarterly) is essential; and the plan should include corrective actions. Stability of tenure at mid-management levels also needs to be ensured. FBR staff ownership of change management will require extensive and regular consultations with all stakeholders, as well as constant training. Increase provincial tax revenues The government probably needs to step back and look at the entire intergov- ernmental fiscal system, which is broken and unbalanced, and its revenue mobilization, which has a narrow lens. At the heart of the tax problem is that provinces have the wrong incentives to collect taxes, though from the pure perspective of raising taxes, the provincial agenda is quite straightforward. • Approve a provincial tax-friendly fiscal transfer mechanism . This implies 279 reviewing the current fiscal transfer mechanism to rebalance provin- cial revenue collection with expenditure needs. As the 18th Amendment stipulated that the share of provinces in the divisible pool of federal revenue cannot drop below the level defined by the 7th National Finance Commission Award (57.5 percent), this requires a constitutional amendment. • Enhance capacity of provincial tax administration. The original model adopted by Sindh and Punjab for creating a new revenue authority is a workable model of modernizing tax administration.17 • Broaden the base of GST on services. The split in the GST regime between federal and provincial entities is not optimal, and an obvious measure is to bring more services into the tax net, including retail trade. Merging profes- sional tax and stamp duties into the GST are also promising approaches. • Enhance collection efficiency of the motor vehicle tax. This would require harmonizing vehicle registration rates. Replacing the one-time registration tax and annual token tax on motor vehicles with an annual license tax and a fixed fuel levy (provincial) would also help increase tax collection. • Revamp the urban property tax. First is a need for removing the confu- sion surrounding ownership. The tax should be devolved to the larger Tehsil Municipal Administration (or municipal corporations), while smaller municipalities retain the provincial collection of the tax, with a transpar- ent revenue-sharing arrangement. Further measures include rationalizing exemptions and tax rates, re-indexing the base and preparing realistic property valuation tables to increase the tax base, and reducing dispari- ties between owner-owned and industrial and commercial properties. A potential second-best option would be to reform the urban property tax in combination with the property transfer tax and an agricultural income tax (Bahl, Wallace, and Cyan 2008). • Eliminate the myriad minor provincial taxes that generate little revenue. To sustain the proposed policy reforms, it is important to prioritize them in the correct sequence, which implies a mix of technical and political consider- ations (Table 16.5). One possibility is to give high priority to broadening tax bases of the GST and income tax and rationalizing tariff exemptions. After the expected results from broadening the tax base have materialized, the government should consider reducing income and GST rates under a phased program, then rationalize the federal excise duty while reforming cus- toms duties. It should accord CIT the lowest priority. The unification of and reduction in low-priority taxes are also important but may be put on hold until mobilizing revenue Table Sequencing of major tax policy reform priorities 16.5 First priority Second priority Third priority 280 Broadening tax bases Rationalizing taxes Reducing tax rates General sales tax/income tax Provincial taxes, individual income Customs duties, zero rates and exemptions tax, corporate income tax federal excise duties Source: Authors. domestic revenues, especially the GST, reach the level of revenues from trade taxes. Nevertheless, GST and CIT remain the most important tax reform areas for improving revenue mobilization. Under any of these scenarios providing alternative mixes of tax policy, however, reform of tax administration and of provincial taxation cannot wait and should be at the top of the priority list. Notes Information on provincial taxation drawn from the background paper prepared by Hanid Mukhtar. 1. Tax buoyancy measures the ratio of the proportional change in tax (or nontax) revenue to the proportional change in GDP. It is obtained by regressing the natural logarithms of tax (or nontax) revenue on GDP series. A ratio greater than 1 shows tax (or nontax) revenue growth above GDP growth. 2. Pakistan’s income tax, general sales tax, and customs tax buoyancies (with respect to their relevant bases) were 0.88, 0.85, and 0.5, respec- tively, in the last three years (World Bank 2012a). 3. According to recent rough estimates, Sindh collects about 50 percent of total provincial taxes; Punjab 25 percent; Khyber Pakhtunkhwa and Fed- erally Administered Tribal Areas 13 percent; and Balochistan 12 percent. 4. Sectors with high effective rates of protection tend to have low value added and to be domestically oriented, which creates an anti-export bias and restricts Pakistan’s graduation from producing low-value products (Lopez-Calix and Touqeer forthcoming). 5. The system of compensatory export rebates—applied to carpets, foot- wear, sporting goods, and surgical instruments—overcompensated for effective taxes that were significantly higher than nominal rates and formed a pure export subsidy (Ahmad 2010). 6. The size of the informal sector of Pakistan could be as high as 35 per- cent of the official economy (Kularatne and Lopez-Calix 2012)—higher than the averages for South Asia (33 percent) and East Asia (32 percent; Schneider and Buehn 2009; Schneider, Buehn, and Montenegro 2010). 7. Individual income tax is levied mainly on workers and salaried persons (as well as small unincorporated businesses). 8. (GST/GDP)/GST tax rate. 9. The C-efficiency index is the ratio of GST revenue to consumption, 281 divided by the standard tax rate. It also shows how effectively the base of the GST is used. It reflects both compliance and the narrowness of the tax base. Most developing countries are around 0.45–0.50, advanced middle-income countries (such as the Republic of Korea and Singapore) 0.6–0.7, and developed countries (such as New Zealand) much closer to 1.00 (Ahmad 2010). 10. In 2007, the government confirmed the tenure of the Chairman of the FBR for three years and renewed the terms of the members responsible for functional areas. But as described, frequent managerial changes continue to plague the institution. 11. Other financing sources of provincial expenditure are drawings from balances created by unfilled positions, slow disbursements of project funds, and bank financing. 12. The Local Government Ordinance of 2001 abolished the distinction between urban and rural municipalities, making it possible for local governments to levy the tax on all properties. But no local govern- ment taxed rural properties. With the subsequent lapsing of the 2001 Ordinance, the former distinction was reinstated and the tax levied only on urban properties. 13. In Punjab, for a market-set rent of PRs 50,000 a month, its tax-rental value is PRs 18,518 a year (3.6 percent of the market rate). 14. By law, these tables (prepared by excise and taxation departments and based on surveys of properties) have to be updated every five years—a long period when real estate and rental prices climb rapidly. Worse, provincial governments generally take much longer to update them. 15. The rental value is assessed based on the valuation tables. Net value deducts repair and maintenance costs, depreciation based on their age, and any land tax that may be paid by the property owner. Moves to change the tax base to capital value have been thwarted by local courts. 16. This tax falls under the jurisdiction of urban local governments, but pro- vincial governments have administered it through provincial excise and taxation departments. Tax revenue is transferred back to the urban local governments on a collection basis but only after making significant deduc- tions for collection fees and financing provincial development authorities. 17. The Punjab model intends to reform one provincial tax at a time, and the Punjab Revenue Authority would be given authority to collect the reformed tax. Over time, all provincial taxes would be reformed. This would remove fragmentation in the provincial tax administration. mobilizing revenue References Ahmad, Ehtisham. 2010. “The Political-Economy of Tax Reforms in Pakistan: 282 The Ongoing Saga of the GST.� Asia Research Centre Working Paper 33, London School of Economics, London. Ahmad, Ehtisham, and Michael Best. 2012. “Pakistan: Using Third Party Infor- mation to Plug Tax Gaps—Alternatives to Policy Reforms.� Mimeo, Octo- ber 2012. Ahmed, Athar M. 2011. “Revenue Analysis and Forecasting: Sectoral and Tax- Gap Analyses.� FBR-USAID-CSF Tax Assistance Program, Islamabad. Ahmed, Robina, and Athar M. Ahmed. 2012. “Estimates of Domestic Sales Tax Gap.� Presentation made in Islamabad. Ather, Robina. 2013. “Study on Tax Expenditures in Pakistan.� Pakistan Policy Paper Series, World Bank, Washington, DC. Bahl, Roy, Sally Wallace, and Musharraf Cyan. 2008. “Pakistan Provincial Government Taxation.� Working Paper 08–07, Georgia State University, Atlanta, Georgia. Baunsgaard, Thomas and Michael Keen. 2005. “Tax Revenue and (or?) Trade Liberalization.� IMF Working Paper 05/112, International Monetary Fund, Washington, DC. CIRP (Center for Investigative Reporting in Pakistan) and CPDI (Centre for Peace and Development Initiatives). 2012. Representation without Taxation: An Analysis of MPs Income Tax Returns for 2011. Islamabad: CIRP and CPDI. FBR (Federal Board of Revenue). 2012. Year Book 2010–2011. Islamabad: Gov- ernment of Pakistan. Government of Pakistan. 2001. The SBNP Local Government Ordinance 2001. Islamabad. ———. 2011. Pakistan: Provincial Government Fiscal Operations, 2010/11. Islam- abad: Ministry of Finance. ———. 2012a. Pakistan: Consolidated Federal and Provincial Budgetary Operations, 2011–12. Islamabad: Ministry of Finance. ———. 2012b. Pakistan Economic Survey 2011/12. Islamabad: Ministry of Finance. ———. 2012c. Pakistan: Provincial Government Budgetary Operations, 2011–12. Islamabad: Ministry of Finance. ———. Various years. Pakistan Economic Survey. Islamabad: Ministry of Finance. ———. Various years. Provincial Government Fiscal Operations. Islamabad: Min- istry of Finance. Kularatne, Chandana, and Jose R. Lopez-Calix. 2012. “Addressing Regulatory Software Barriers to Business Growth.� Pakistan Policy Paper Series 07/12, World Bank, Islamabad. Lopez-Calix, Jose R., and Irum Touqeer. 2013. “Pakistan’s Development Strategies: Possible Synergies Between Federal and Provincial Strategies.� Background Paper for Pakistan: Finding the Path to Job-Creating Growth. World Bank, Washington DC. Lopez-Calix, Jose R., and Irum Touqeer. Forthcoming. “Pakistan: Revisiting the Constraints to Pakistan’s Growth.� Pakistan Policy Paper Series, World Bank, 283 Washington, DC. National Accountability Bureau. 2002. National Anti-Corruption Strategy. Islam- abad: Government of Pakistan. Reis, Jose G., and Daria Taglioni. 2013. “Reinvigorating the Agenda for Open Trade.� Pakistan Policy Note 2, World Bank, Washington, DC. Schneider, Friedrich, and Andreas Buehn. 2009. “Shadow Economies and Cor- ruption All over the World: Revised Estimates for 120 Countries.� Economics: the Open Access, Open-Assessment E-Journal 1 (2). Schneider, Friedrich, Andreas Buehn, and Claudio E. Montenegro. 2010. “Shadow Economies All over the World: New Estimates for 162 Countries from 1999–2007.� Policy Research Working Paper 5356, World Bank, Wash- ington, DC. Transparency International 2012. Global Corruption Barometer 2010/11. Berlin. Available at http://gcb.transparency.org/gcb201011/. USAID (United States Agency for International Development). 2011. “Collect- ing Taxes (2009–10).� Fiscal Reform and Economic Governance Project, Washington, DC. World Bank. 2009. Pakistan Tax Policy Report: Tapping Tax Bases for Development. Report 50078-PK. Washington, DC. ———. 2012a. South Asian Economic Focus, June 2012: A Review of Economic Develop- ments in South Asian Countries: Creating Fiscal Space through Revenue Mobilization. Washington, DC. ———. 2012b. World Development Indicators 2012. Washington, DC. ———. 2012c. Implementation Completion and Results Report for Pakistan Tax Admin- istration Reform Project. Report ICR00002147. Washington, DC. ———. 2012d. Doing Business 2012: Doing Business in a More Transparent World. Washington, DC. ———. 2012e. Doing Business database. Washington, DC. About the Authors 285 Rachid Benmessaoud has been the World Bank’s Country Director for Pakistan since 2010, based in Islamabad. Since joining the World Bank in 1990, he has held several positions in the Middle East and North Africa and Europe and Central Asia Regions, and worked on energy, infrastructure, and municipal finance projects. In 2003, he joined the South Asia Region as Operations Adviser for India, based in Delhi, where he oversaw the Bank’s India portfolio. He holds a BS in physics from Université d’Aix-Marseille II, an electrical engineering degree from Ecole Supérieure d’Electricité, a Doctor-Engineering Degree from Université de Paris XI, and an MBA from the University of Denver. Uzma Basim is working as the Senior Operations Officer at the Bank office in Pakistan. Her current focus areas are preparing the five-year Country Partnership Strategy for Pakistan and facilitating delivery of agreed programs and projects at the federal and provincial levels. She has an MPhil in devel- opment studies from Cambridge University and an MBA from Quaid-e-Azam University, Islamabad. Anthony Cholst is the Country Program Coordinator for Afghanistan, Bhutan, and Pakistan at the Bank. Before this he held several positions within the Bank including working on the South Caucasus and Central Asia Regions, and India. He has an MBA from University of Chicago and an MS in economics from the London School of Economics. M. José R. Lopez-Calix is the Lead Country Economist for Pakistan. He has about 20 years of experience across middle-income and lower income coun- tries. He has held a number of key Bank managerial and economist positions in most Latin American and Maghreb countries. Before joining the Bank, he was General Director for the Planning Ministry that governed the transition budget program between pre- and postwar periods in El Salvador. A former consultant in several international organizations and visiting professor in many American universities, he is the author of a number of Bank reports, studies, and academic articles on public finance, public expenditure tracking surveys, trade, growth, tax policy and administration, Public Expenditure and Financial Accountability assessments, risk assessments, and early warning sys- tems. He holds a PhD in economics from the Catholic University of Louvain University, an MSc international finance from the University of Pittsburgh, and 286 an MA from the Catholic University of Louvain. Rashid Aziz is a Senior Energy Specialist in the Bank’s South Asia Sustainable Development Department, and is responsible for energy sector operations in Pakistan. He has more than 25 years of experience working in the energy sector in Pakistan, including policy dialogue; project preparation, implemen- tation, and completion; donor coordination; and sector reviews and analysis. Prior to joining the Energy Unit, he worked as an economist in the Country Programs Department of the South Asia Region. Mohammad Saqib a Senior Energy Specialist at the Bank office in Islamabad. He is a CFA charter-holder and has more than 12 years of experience in the energy sector. He has worked on large infrastructure projects, such as hydropower and cross-border gas pipelines. His energy sector experience also includes analyzing and advising on commercial and policy matters related to tariffs, supply and demand, price forecasts, product marketing, and business development. José Guilherme Reis is an economist with the Bank’s International Trade Department, where he is leading the program on trade and competitiveness. He joined the Bank in 2004, and his work program has been focused on the micro- economic agenda for growth in Latin America and topics related to competitive- ness. He led several studies on competitiveness, trade, and growth in countries across the world. Past positions include Secretary for Economic Policy with Brazil’s Ministry of Finance (2001–02) and Chief of Economic Advisors with the Ministry of Planning and Budget (1999–2001). He holds a master’s in economics and public finance from the Catholic University of Rio de Janeiro, where he has been a professor in the Department of Economics since 1983. Daria Taglioni is a Senior Economist at the Bank. Previously she worked with the European Central Bank (2006–11) and the Organisation for Economic Co-operation and Development (2002–06). Her recent research focuses on international competitiveness, the economics of value chain production, and the link between financial markets and trade. She holds a PhD in inter- national economics from the Graduate Institute in Geneva, a master’s in economics from the College of Europe in Bruges, and an undergraduate degree from the University La Sapienza of Rome. Charles Kunaka is a Senior Trade Specialist with the Bank, and has more than 20 years of experience working on transport economics and policy, logistics, and trade facilitation. He holds a PhD in transport from University College London and a master’s in transport economics and policy from Cranfield University in the United Kingdom. He has published on various aspects of transport services, more recently on logistics in lagging regions and on 287 road transport services. He is currently finalizing a guidebook on trade and transport corridor management designed to help policy makers and project managers maximize the impact of corridor projects. He has previously worked for the Sub-Saharan Africa Transport Policy Program and the Southern African Development Community. Nihal Pitigala is an international trade economist with more than 15 years of experience in providing analysis, technical assistance, and capacity building to support sustainable trade and market-led development in developing countries. His recent research and analytical work has focused on regional integration, vertical specialization, global value chains, and trade competitiveness. He has a PhD in economics from the University of Sussex. Mehnaz Safavian is a Senior Economist in the Bank’s South Asia Financial and Private Sector Department. She has worked extensively on financial inclusion issues, including micro, small, and medium-size enterprise finance, gender, creditors’ rights, informality, and responsible finance. She holds a PhD in economics from Ohio State University, and has published extensively in World Bank and scholarly journals. Before joining the Bank, she held different positions with the U.S. government, development nongovernmental organiza- tions, and academia. Sarwat Aftab is a Senior Private Sector Development Specialist with the Bank’s South Asia Financial and Private Sector Development team in Islamabad. She joined the Bank in February 2012, and has worked on such areas as finan- cial inclusion, business regulations, corporate governance, state-owned enter- prises reforms, consumer protection, and women entrepreneurship. She has more than 17 years of work experience. Before joining the Bank, she worked for more than 10 years at the Securities and Exchange Commission of Paki- stan as a financial sector regulator for corporate sector, capital market, and nonbank financial institutions. She also taught at the Central Queensland University in Australia. At the start of her career, she worked with the Competition Commission of Pakistan. Sarmad Shaikh is an analyst with the Bank’s South Asia Financial and Private Sector Development team in Islamabad since September 2012, focusing on financial inclusion, business regulations, state-owned enterprise reforms, and women entrepreneurship. He has previously worked with the Bank’s Doing Business team in early 2012. Prior to joining the Bank, he worked as an investment banking analyst at Citibank on a number of mergers and acquisitions, project finance, and privatization transactions. He is a 288 Fulbright Scholar, holds a master’s in public administration from Cor- nell University, and is currently enrolled in the Chartered Financial Analyst program. Dan Biller is the Sustainable Development Lead Economist for the Bank’s South Asia Region. Previously, he was the Lead Economist for the East Asia and Pacific Region, and Environment and Natural Resources Program Leader at the World Bank Institute. He was also a senior economist at the Organisa- tion for Economic Co-operation and Development in Paris, and has taught at the Getulio Vargas Foundation and the Saint Ursula University in Brazil. He received his MS and PhD in economics from the University of Illinois at Urbana-Champaign, and a BS in economics with a minor in geophysics from the University of Kansas. He has published extensively on topics such as economic development, natural resource and environmental economics, sustainable consumption, urban and rural linkages, infrastructure, climate change, and social development issues. Ernesto Sánchez-Triana is the Lead Environmental Specialist for the Bank’s Latin America and the Caribbean Vice-Presidency. He has led the prepa- ration of numerous policy-based programs, investment projects, technical assistance operations, and analytical works. From 2006 to 2012, he worked in the South Asia Region, where he led multiple operations, including “Green- ing Growth in Pakistan through Transport Sector Reforms.� He joined the Bank in 2002 after having served as a government officer and university professor. He holds MS and PhD degrees from Stanford University, and an engineering degree from Universidad de los Andes in Colombia. He has numerous publications in the areas of environmental economics, energy efficiency, poverty reduction, strategic environmental assessment, pollution manage- ment, and green growth. S. Amer Ahmed is an economist with the Bank’s Development Prospects Group, where he contributes to analytical work on long-term global eco- nomic issues. At the Bank, he has conducted policy research on agriculture, climate change, growth, and international trade, among other topics. His most recent contributions to the South Asia Region have been a simulation analysis of climate change impacts on Bangladesh’s economic growth, and reviews of agricultural policies in India and Pakistan. His research has been published in peer-reviewed research journals and World Bank reports. He holds a PhD in agricultural economics from Purdue University. Madhur Gautam is a Lead Economist for agriculture, irrigation, and natural resources in the Sustainable Development Department of the Bank’s South Asia Region. Before joining the South Asia Region in 2011, he worked with the Development Economics Vice-Presidency (World Bank’s Research Department), 289 Independent Evaluation Group (then Operations Evaluation Department), and analytical and operational work in the Africa and South Asia Regions. Among his major accomplishments are an impact evaluation of the Agricultural Exten- sion System in Kenya; a review of the HIPC Debt Reduction Initiative; and major analytical reports (economic and sector work) for Ethiopia, Kenya, Tan- zania, and Uganda. His current focus is on strategy, analytical work, and policy dialogue on agriculture and rural development issues across the South Asian Region. He received his master’s degree from the Delhi School of Economics at the University of Delhi, and a PhD from the University of Maryland. Cem Mete is a Lead Economist in the Social Protection Unit of the Bank’s South Asia Region. He has operational experience in both middle- and low-income countries, including several fragile states. In analytical work, he has led pov- erty assessments, various economic and sectoral studies (on labor, skills, social protection, demography, and the like), and a Bank regional study on the economics of disability and chronic illness. He has also published academic papers in a wide range of journals including the Economic Development and Cultural Change, Economics of Education Review, Health Economics, and Population and Development Review. His work on impact evaluation includes an assessment of the universal health insurance in Taiwan, China, on elderly labor force participation, the impact evaluation of a significant payroll tax increase in Romania on informal sector employment, and the impact evalua- tion of a pilot conditional cash transfer program in Pakistan. Rehan R. Jamil is currently working as a consultant with the Bank’s South Asia Human Development Unit, focusing primarily on social protection and labor-re- lated analytical work in the region. Before joining the Bank, he worked with the United Nations Development Programme country office in Pakistan. He has a master’s in international affairs from Columbia University with a concentration in economic and political development. His research interests include social pro- tection, political economy, poverty, and rights and social exclusion in South Asia. John Newman is the Lead Poverty Economist in the Economic Policy and Poverty Sector of the Bank’s South Asia Region. He has held several positions in the operational and research departments at the Bank, most recently as Lead Economist in the Poverty and Gender Group of the Latin Amer- ica and the Caribbean Region and as a Bank representative in Bolivia and Peru. Prior to his Bank experience, he was assistant professor in economics and Latin American studies at Tulane University. He holds a PhD in economics from Yale University and MA and BA degrees from Johns Hopkins University. He has been published in the areas of impact evaluation, labor economics, and 290 poverty analysis. Halil Dundar is a Lead Education Specialist in the Human Development Department of the Bank’s South Asia Region. He joined the Bank in 1998 as an education economist, and is currently leading the Bank’s education policy dialogue and operational and analytical work in Pakistan’s tertiary education sector and Sri Lanka’s skills development sector. Before joining the South Asia Region, he worked in the Europe and Central Asia Region and the Africa Region, where he was responsible for the Bank’s education policy dialogue and operational and analytical work in Albania, Azerbaijan, Ethio- pia, and Nigeria. He received his PhD from the University of Minnesota in education policy, with a focus on the economics of education. He completed a postdoctoral fellowship at the University of Minnesota and worked as a research associate at the university’s Office of Planning and Analysis, and subsequently held the position of assistant professor of educational policy at the Middle East Technical University in Ankara, Turkey. He has authored or co-authored more than 30 monographs, journal articles, and book chapters in the field of education economics. Huma Ali Waheed is the Operations Officer in the Human Development Department of the Bank’s South Asia Region in Islamabad. Shahnaz Kazi received her doctorate in economics from the University of London’s School of Oriental and African Studies. She served as Senior Economist in the World Bank office in Islamabad from 2000 to 2011. During this period, she supervised and evaluated the Bank’s lending operations in the health sector and contributed to research and sector work including the Country Poverty Assessment and health sector performance reviews. Before joining the Bank, she worked as Chief of Research in the Pakistan Institute of Development Economics. Her research encompasses gender issues, reproduc- tive health, employment in the informal sector, and international migration. Inaam Ul Haq is working as a Lead Health Specialist in the Bank’s South Asia Human Development Department, and is responsible for operations in Afghanistan and Pakistan. He is a public health physician with 25 years of experience in health systems development and management in governmental, nongovernmental, and international organizations. Before joining the Bank, he worked with the U.K. Department for International Development, and taught management and planning of health systems at Aga Khan University’s Department of Community Health Sciences. He is a physician by training, having graduated from the Rawalpindi Medical College in Pakistan. He received on-the-job training in public health at Aga Khan University, and holds a master’s in public health from Harvard University’s School of Public Health, 291 with a concentration in international health. Luc Laviolette is the Senior Nutrition Specialist of the Bank’s South Asia Region. Before joining the Bank in 2009, he held several positions including Regional Director for Asia at the Micronutrient Initiative, and various manage- ment positions at the Canadian International Development Agency, including Deputy Director of the Program against Hunger, Malnutrition and Disease. He also served as Counselor to the Executive Director for Canada, China, Korea, Kuwait, and Spain at the African Development Bank, and worked at the International Development Research Centre. He holds a master’s in international affairs and a postgraduate diploma in health systems management. Kees Kostermans, MD, MPH, is a Lead Public Health Specialist in the Bank’s South Asia Region. He has published on the impact of polio eradication efforts on routine immunization programs, health infrastructure, gender, HIV, and malnutrition in Afghanistan. Before joining the Bank in 1993, he worked for the Dutch section of Doctors without Borders in Afghanistan, Colombia, Mozam- bique, and Pakistan. Iftikhar Malik is working with the Bank’s South Asia Region as Senior Social Protection Specialist and Pakistan Country Sector Coordinator for Human Development. He has extensive experience in designing and managing social sector operations, and co-authored various analytical pieces on social protection policy, targeting, health insurance, and safety net graduation. He has overseen the design and implementation of various cash and conditional cash transfer (CCT) programs in Pakistan as well as a CCT program in Kenya. He has closely contributed to the evolution of the national social safety net system in Pakistan, which is now used as a platform for coordinating and integrating various pro-poor programs. He has helped Bangladesh and other countries replicate the system. He has professional interest and expe- rience in social protection policy, institutional development, and safety net operations. Lucian Pop is a Senior Economist in the Bank’s South Asia Social Protection Sec- tor. His experience includes work in the Africa and Europe and Central Asia Regions. He holds a PhD from the University of Bucharest, and has authored sev- eral scholarly articles and institutional reports on poverty, inequality, and social protection. Haris Khan is the Disaster Risk Management Country Coordinator for the World Bank office in Pakistan. He has been with the Bank for more than seven years, working on improving preparedness and postdisaster reconstruc- 292 tion projects specifically in Afghanistan, Bangladesh, Bhutan, Pakistan, and Sri Lanka. He has also worked on four damage and needs assessments in the region, and is part of the Bank’s Rapid Response Team, which, among many others, assisted in the 2004 Indian Ocean tsunami, 2005 Pakistan earthquake, 2008 Bhutan earthquake, 2008 Pakistan cyclone, 2008 internally displaced persons crisis, and 2010 and 2011 Pakistan floods. He has previously worked with the Australian Government Overseas Aid Program, leading their humanitarian and governance program in Pakistan and working in East Asia, including in the Phillipines. He holds a master’s in transport management, with majors in transport economics from the University of Antwerp in Belgium. Marc Forni is a Senior Disaster Risk Management (DRM) Specialist. He joined the World Bank in 2003 in the Latin America and the Caribbean Region, where he helped expand the DRM program from $40 million to $2.2 billion. Notably, he helped create the world’s first multinational disaster risk insurance facility—the Caribbean Catastrophe Risk Insurance Facility. After spending three years as an investment banker in New York, he returned to the Bank in the aftermath of the Haiti earthquake in 2010. Currently with the South Asia Region, he is leading programmatic engagements across South Asia, with a focus on understanding and mitigating disaster risk. Naseer Ahmad Rana is a Governance Advisor in the Governance and Public Sector Department of the Bank’s South Asia Region. He is currently lead- ing the Bank’s governance work program nationally and subnationally in Pakistan. His prior postings include India, Indonesia, and the United States, with work experience in all countries of the South Asia Region and sev- eral countries of the East Asia and Pacific Region. He holds a master’s in systems engineering from the University of Southern California and a bachelor’s degree from University of Illinois at Urbana-Champaign. Iffat Idris is currently working as a consultant with the Bank’s South Asia Governance and Public Sector, focusing on the promotion of right to infor- mation in Pakistan. She has worked as a freelance consultant with different organizations for several years, and has extensive experience in the fields of governance (right to information, accountability, devolution), capacity development, social protection, and disaster management. She holds a PhD in political science from the University of Hull. Irum Touqeer is an analyst with the Bank’s South Asia Economic Policy and Governance teams. She worked on various economic and governance issues, including economic policy, tax policy, regional trade and integration, accountability institutions, anticorruption reforms, and right to informa- 293 tion. She has also worked with the United Nations Development Programme on Pakistan’s subnational Millennium Development Goals program. She has co-authored the Bank’s Pakistan Policy Papers on revenue mobilization, on revamping governance, on binding constraints to Pakistan’s growth, on possible synergies between federal and provincial growth strategies of Pakistan, and on institutions of accountability. She holds a master’s in interna- tional development economics from the Australian National University, and an MBA in management and finance from Hamdard University in Pakistan. Saeeda Sabah Rashid is the Senior Financial Management Specialist of the Bank’s Financial Management Department, focusing on Pakistan’s public financial management issues for about seven years. Before joining the Bank, she worked in the exploration and production sectors of Pakistan, and gained audit and accounting experience from PricewaterhouseCoopers in 2001. She is a fellow member of ACCA UK and the Institute of Chartered Accountants of Pakistan, with additional specialization in international financial report- ing standards and exploration and production accounting. Uzma Sadaf is a senior procurement specialist in the Bank's South Asian Region. She has a master of engineering from the Asian Institute of Technol- ogy in Thailand. Hanid Mukhtar is a Senior Economist with the Bank’s Poverty Reduction and Economic Management Unit in Pakistan. He has extensively worked on Pakistan’s fiscal issues with a special focus on tax policy and expenditure management, and on different Bank reports on Pakistan since 1990. Cur- rently, he is working on a Public Expenditure Review of the Punjab govern- ment. Before joining the Bank in 1990, he worked exclusively in academia with considerable experience in teaching and research at Quaid-i-Azam University at Islamabad, University of Massachusetts at Boston, Northeastern University in Boston, Boston University, and Applied Economics Research Center of Karachi University. For a brief period, he worked with the United Nations Economic and Social Commission for East Asia and the Pacific office in Bangkok, devel- oping a macroeconometric model for Pakistan to be integrated in the global LINK Model. He holds PhD in economics from the Boston University. He has authored several publications in the field of micro and behavioral economics.