82090 EVALUATION OF THE WORLD BANK GROUP’S ACTIVITIES IN THE EXTRACTIVE INDUSTRIES Background Paper Factoring in Governance September 1, 2004 Document of the Operations Evaluation Department This report was prepared for OED by M.A. Thomas of the Center for Institutional Reform and the Informal Sector, University of Maryland. The views expressed herein do not necessarily represent the views of the Center, OED or the World Bank. Acronyms and Abbreviations CAE Country Assistance Evaluation CAS Country Assistance Strategy CFAA Country Financial Accountability Assessment CPAR Country Procurement Assessment Review EI Extractive industry(ies) ESW Economic and sector work IBRD International Bank for Reconstruction and Development IDA International Development Association IFC International Finance Corporation GRICS Governance Research Indicator Country Snapshot LICUS Low-Income Countries Under Stress MIGA Multilateral Investment Guarantee Agency OED Operations Evaluation Department PER Public Expenditure Review SSM Small-scale mining WBG World Bank Group Background papers, as special working papers, are informal documents made available to disseminate the findings of key building blocks for major evaluation studies and to encourage the exchange of ideas about development effectiveness through evaluation. The findings, interpretations, and conclusions expressed in these papers are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries i Table of Contents 1. Introduction..................................................................................................................1 2. Governance and Extractive Industries ......................................................................3 3. Does the Bank Factor in Governance?.......................................................................8 Successful Projects without Development Impact? .................................................9 No Link from Governance ESW, to CAS, to EI Projects .......................................10 A Growing Concern for Governance.....................................................................13 The Need for Selectivity and Sequencing...............................................................14 4. Factoring Governance into the Bank’s Approach ..................................................17 Possible Standards for Bank Assistance ................................................................18 Establishing and Implementing Standards ............................................................21 5. Conclusion ..................................................................................................................23 Appendix. World Bank, IFC and MIGA Oil and Gas and Mining Projects from FY1993 ......................................................................................................................................27 References.........................................................................................................................29 1 1. Introduction 1.1 Many poor countries are rich in resources such as oil, gas, metals, and minerals. The judicious development of those resources (here termed “extractive industries” or “EI”) can bring benefits for development. Often the most important potential development benefit is increased government revenue, provided that the revenue is well-managed in the public interest. Depending on the resource, its location, the method of exploitation, and other factors, benefits can also include substantial temporary employment, the construction of infrastructure, the stimulation of growth in related sectors, technological innovation, and the transfer of training and technology. 1.2 Because such resources are typically limited and non-renewable, the development of EI cannot be a long-term foundation for the economy. However, it is possible for governments to exploit such resources and invest the revenues so as to create sustainable growth and development. An example of a country that has achieved this is Botswana, which over the 30-year period from 1966–96 attained a higher per capita growth rate than any country in the world, facilitated by the development of its mining sector. [Leith 1999] 1.3 But the exploitation of these resources is not without risk or cost. Risks and costs can include long-term environmental damage with accompanying health consequences, the destruction of the traditional (and more sustainable) economic foundations of local communities, involuntary displacement and property takings, economic dependence on such revenues and increased economic volatility, increased corruption, violence and civil war. 1.4 An additional complication of EI projects1 is that the benefits and risks are not evenly distributed. Local communities benefit from the employment and infrastructure improvements that such projects bring (which are temporary, but can last for extended periods), but they also bear the brunt of any accompanying environmental damage, health risks, property takings and damage, and changes to traditional life and culture. The distribution of revenues, on the other hand, is determined by the legal and regulatory framework of the country. Typically, most revenues are split between the investor (in lower income countries, often a foreign company) and the central government. 1.5 The World Bank Group (WBG) supports the development of EI sectors in a number of ways. One way is by helping governments attract more sectoral investment. The WBG works with governments to help them attract investors, by revising their fiscal and legislative frameworks to create an investment-friendly business environment. The International Finance Corporation (IFC) and the Multilateral Investment Guarantee Agency (MIGA) work to attract increased investment by supplying would-be investors with investments (loan and equity) and risk insurance.2 The Bank also works with governments to help them maximize the benefits of the EI sector, supporting reforms for better revenue collection, increased productivity, and better public financial management. Finally, the Bank works with 1. In this paper, the EI sector is defined to mean the oil, gas, and mining sectors. 2. By ensuring that projects comply with the WBG’s environmental, social, health, and safety requirements, IFC and MIGA also provide important demonstration effects for other projects. 2 governments to help them manage the risks of such projects, supporting revisions to laws and the strengthening of institutional capacity to manage environmental, health, and safety risks. 1.6 This paper examines the links between governance and the EI sector, and considers the implications of those links for the work of the World Bank (the International Bank for Reconstruction and Development and the International Development Association). This evaluation takes place on the heels of important complementary World Bank initiatives, upon which it draws. The Bank has pioneered a research initiative into the economics of civil war, crime and violence, which considered inter alia the effects of natural resource endowments.3 A task force has examined the special problems of Low Income Countries Under Stress.4 There is overlap between this group of countries and the group of resource-rich Bank borrowers. And the Bank has done substantial research linking governance to development outcomes. 1.7 Section 2 describes the links between governance and extractive industries. It draws on the body of work that links poverty reduction and aid effectiveness to good governance, and, more specifically, links the development impact of the EI sector to the quality of governance. 1.8 Section 3 evaluates the extent to which the Bank has factored the linkages between governance and the development impact of EI projects in its approach to the EI sectors in Chile, Ecuador, Ghana, Kazakhstan, Papua New Guinea, and Tanzania from 1993 to 2002.5 Only World Bank projects are considered, although IFC and MIGA projects are noted for context. These evaluations are based on a retrospective document review of World Bank project documents and reports available through the World Bank’s internal project documentation system, as well as secondary literature. The countries were selected ex ante for review based on variation in region, the size and importance of the EI sector, the quality of governance, and the degree of Bank intervention in the sector. The evaluation finds an absence of a strategic approach to the EI sector informed by the quality of governance, and concludes that the Bank’s approach could be improved with more explicit consideration of governance constraints and risks, and with better sequencing and selectivity. 1.9 Section 4 suggests how the Bank might better factor governance linkages into its approach to the EI sector in the future. Section 5 concludes. The Appendix lists World Bank and IFC mining, oil, and gas projects in the selected countries for the period under review. 3. See http://econ.worldbank.org/programs/3102. 4. See http://www1.worldbank.org/operations/licus/. 5. While one should be cautious about generalizing the results of small-sample studies, in this case it seems unlikely that expanding the sample of projects would yield substantially different results. This is because the findings are driven by institutional changes in the Bank-wide approach to governance that took hold in the late 1990s. There is no reason to suspect that the Bank’s approach to the EI sector in light of governance was significantly different than its approach to other sectors during the period under review. 3 2. Governance and Extractive Industries 2.1 If the promise and potential of development through EI is tempting, it is also elusive. While countries such as Botswana and Chile have benefited demonstrably from the judicious exploitation of their EI sectors, economists have long argued that, paradoxically, for most countries, resource abundance has proved to be problematic. Auty [1994] found that resource abundance is strongly associated with poor economic performance, while resource booms do not lead to sustained economic growth. Poor countries with abundant resources have grown more slowly than poor countries without. [Sachs and Warner 1995] 2.2 Many stakeholders play an important role in ensuring that extractive industry projects lead to positive development outcomes. Investors are needed who will shoulder the enormous costs and risks associated with exploration and resource generation, minimize the damage and risk that such projects can present, and respect local laws and regulations regarding fiscal responsibilities, environmental protection, and property rights. Civil society has a role to play by participating in decisions regarding the projects that affect them, respecting investor rights while insisting on compliance with law and best practice, and holding the government accountable for the responsible management of the sector and use of revenues. 2.3 But increasingly, scholars are recognizing that whether the growth of an EI export sector leads to positive development outcomes depends largely on the government. The government must play a demanding role if it is to ensure that EI projects provide benefits, and that those benefits are not outweighed by risks and costs. The government must maintain an institutional and policy environment that attracts private investment into the sector, and ensures that investor rights are protected. It must negotiate with investors so as to obtain the maximum benefit for the public, manage non-renewable resources with an eye to the future, and ensure that revenues from the EI sector are managed transparently and effectively in the public interest. The government must also monitor and control the social and environmental costs and risks of EI projects, protecting the environment, the national interest, and the interest of local communities. 2.4 Unfortunately, many poor country governments — the very governments that need to build sustainable development and are most tempted to seek the revenue infusion that extractive industry projects can bring — lack the capacity and the institutions needed to carry out these complex tasks. In the worst cases, the EI sector can bring little public benefit and leave long-term costs behind in the form of environmental destruction and war. 2.5 Some explanations for the poor performance of resource rich countries have focused on poor economic policies, such as government failure to promote a competitive manufacturing sector, or poor intersectoral allocation of resource windfalls. [Sarraf and Jiwanji 2001; Usui 1997; Auty 1998; McMahon 1997] But a number of scholars explain inappropriate economic policies in turn as the outcome of governance factors such as rent- seeking and patronage. [Eifert et al. 2002; Isham et al. 2002; Robinson et al. 2000; Ross 1999; Coolidge and Rose-Ackerman 1997] Good governance is the key to positive development outcomes. 4 2.6 In its 1992 report “Governance and Development” the World Bank defined “governance” as “the manner in which power is exercised in the management of a country’s economic and social resources for development.” [World Bank 1992a 1] This definition was further elaborated in 1994. “Good governance is epitomized by predictable, open and enlightened policy making (that is, transparent processes); a bureaucracy imbued with a professional ethos; an executive arm of government accountable for its actions; and a strong civil society participating in public affairs; and all behaving under the rule of law.” [World Bank 1994] 2.7 Good governance is a development objective in and of itself. The United Nations defines political participation and freedom as “fundamental parts of human development.” [UNDP 2002] The World Bank defines poverty as including “powerlessness, lack of representation and freedom.”6 But the quality of governance is also an important factor affecting other development goals. The World Development Report explains that “chronic ineffectiveness of the state” can lead to “social violence, crime, corruption, and instability, all of which undermine the state’s capacity to support development — or even to function at all.” [World Bank 1997] World Bank research has shown the impact of the quality of governance on economic growth [Kaufmann et al. 1999a; Kaufmann and Kraay 2002], the effectiveness of public spending [Rajkumar and Swaroop 2002] and on technological deepening [Clarke 2001]. 2.8 Likewise, the quality of governance is an important determinant of the development benefits of an extractive industry project. (See IFC 2003 30.) The quality of governance affects the process of building sustainable development from an extractive industry project at every stage, from managing non-renewable resources with an eye to the future, to building an enabling environment and attracting investment, to generating the resource, collecting the revenue, and investing resources well for longer-term growth. (See Table 1). 2.9 Governments must be able to capture and maximize the benefits of these projects, which implies good public financial management. They must also be able to control the costs and risks. This in turn implies the existence of an appropriate legal, regulatory, and institutional framework, government respect for its own laws and the capacity to implement them. Countries such as Chile and Botswana have successfully leveraged their wealth into sustainable growth through investment-friendly policies, fiscal discipline, and long-term planning. Perfect governance is not needed for an extractive industry project to be beneficial, but some minimum standards must be met to help ensure that the benefits of EI projects are not squandered and the citizens left with costs that can include environmental damage, health risks, and war. 6. See http://www.worldbank.org/poverty/mission/up1.htm. 5 Table 1. Some governance aspects of building sustainable development from EI projects Extractive Industry Project Enabling Environment Operation Collection Revenue Use • Elaboration of country • Environmental monitoring • Transparency of amounts • Consistency of budget development strategy and control agreed and amounts with country development • Objective government • Protection of local collected strategy capacity to weigh community interests; • Transparent and regular • Revenues from EI sector risks/benefits of EI monitoring; resolution of handling of revenues allocated and spent projects claims and compensation • Contracted amounts effectively consistent with • Management of schemes collected and transferred budget exhaustibility of EI • Protection of investor into budget • Transparency of resources interests; performance of government use of • Negotiation of agreement contract terms revenues in public interest • Government and • Oversight and • Framework for local government actor freedom accountability of executive community compensation from conflicts of interest branch use of revenues and participation • Transparency of conduct • Management of • Viable legal/regulatory of project and government exhaustibility of resources investment framework administration of project • Environmental regulatory • Availability of judicial framework recourse • Capacity to assure law and order • Transparency of arrangements (taxes, environmental standards, compensation regimes) • Government respect for laws and contracts 2.10 Unfortunately, the quality of governance in countries with strong extractive industries is often poor.7 The World Bank Institute’s Governance Research Indicators (GRICS) estimates six dimensions — voice and accountability, political stability, government effectiveness, regulatory quality, rule of law, and control of corruption.8 Figure 1 shows the association between the average of the GRICS indicators and EI dependence.9 7. There is a significant correlation between poverty and poor governance, and many EI dependent countries are poor. It is not asserted here that the quality of governance in EI countries is worse than in poor countries generally. 8. For additional information on GRICS, see http://www.worldbank.org/wbi/governance/data.html#dataset2001 9. This relationship, which is statistically significant at the 95 percent confidence level (t-statistic = 2.44), illustrates a conclusion that is widely accepted in the literature. No claim is made that EI dependence is the sole determinant of a country’s quality of governance. The figure includes all countries eligible for borrowing from the WBG with a population greater than 1 million as of 2000, for which data are available. 6 Figure 1: Worse Country Governance with Greater EI-Dependence 0 20 "Composite" GRICS Indicator 40 2000/2001 60 80 100 120 140 R k* 160 180 0 20 40 60 80 100 Average EI Exports / Total Exports, 1990-99 (%) * “Composite” GRICS (Governance Research Indicator Country Snapshot) ranks are a simple average of individual GRICS rankings for 2000/2001 for Voice and Accountability, Political Stability, Government Effectiveness, Regulatory Quality, Rule of Law and Control of Corruption. Source: http://www.worldbank.org/wbi/governance/pdf/2001 kkzcharts.xls 2.11 The consequence of poor governance for management of the EI sector is that the government is unable to carry out the necessary tasks to ensure that EI projects are in the public interest. The government is unable to collect or manage revenues well, although increased government revenue may be the principal benefit of the EI project.10 Where there is poor capacity or systemic corruption, government officials may not negotiate the best agreements for their country to begin with. When the revenues from projects begin to flow, they may not reach the Treasury, or they may be lifted right out again. 2.12 When public financial management is poor, EI revenues — like other revenues — are often not well spent. Just as studies have shown that governments with poor policies and poor institutions do not make effective use of aid money [World Bank 1998], studies have shown that when governments with poor governance get EI windfalls, they do not make effective use of them either. Lacking public accountability and dependent on the distribution of private goods such as jobs and public contracts for the continued enjoyment of political power, resource booms are frequently accompanied by mass public sector hiring and unsustainable budgetary policies. [Leite and Weidmann 1999] 2.13 Not only are governments with poor governance unable to capture or maximize the benefits of the EI sector, they are also unable to control the costs and risks to the environment, to health, and to safety. While an appropriate legal, regulatory, and institutional 10. In some cases, the principal benefit of EI investment is not increased government revenues. For example, small-scale mining provides substantial employment and may contribute to GDP even when smuggling means that taxes are evaded. [Phillips et al. 2001] In such cases, the government’s public financial management capacity is not a significant factor; rather the important government capacity is the government’s ability to manage risks. In small- scale mining, these can include dangerous working conditions, lawlessness, and environmental degradation. 7 framework to carry out these functions is necessary and often lacking, it is in any event not sufficient. Governments that lack the rule of law11 ignore their own laws and regulations. Governments that have weak administrative capacity may be unable, even with good intentions, to exercise effective supervision and control of EI projects far from the capital. In the absence of government, resource extractors may violate environmental and safety laws, while conflicts between communities and extractors, or between small-scale miners and large foreign mining companies, become violent. Armed groups may seek to conquer projects or hold them hostage. (See, e.g., Collier and Hoeffler 2001). Collier and Hoeffler found that the possession of natural resources has an effect on civil war, increasing both the risk and the duration. [Collier and Hoeffler 1999] 2.14 Economists have long known that resource wealth can have perverse economic effects. But research has also shown that the presence of a strong EI sector can have an impact on the quality of governance. Leite and Weidmann show that natural resources are often associated with increased corruption. [Leite and Weidmann 1999] Scholars have found that when governments have access to wealth that citizens cannot control, they are less accountable. Moore et al. [1999] found that “[g]overnments that are dependent on their own citizens for critical resources appear more effective at converting material resources into human development.” Other scholars have explained the relationship between resource dependence and authoritarianism [Wantchekon 1999; Karl 1999] and soft budget constraints and failure to reform [Bates 1981]. Knack [2000] makes a similar argument concerning aid dependence, claiming that it can “undermine institutional quality by weakening accountability, encouraging rent seeking and corruption, fomenting conflict over control of aid funds, siphoning off scarce talent from the bureaucracy, and alleviating pressures to reform inefficient policies and institutions.” 2.15 In short, when governance is poor, resources can be a development “curse.” Unable to control corruption and manage revenues wisely, the government is unable to capture the benefits. Without the rule of law, the government is unable to implement legal, regulatory, and policy solutions that would allow it to control the costs and risks. There does not seem to be much of an argument in favor of developing or expanding the EI sector in such environments. 2.16 The EI sector itself often faces particular governance challenges. Marshall argues that the mining industry is “especially vulnerable to supply-side corruption.” [Marshall 2001 8- 13] Heavy regulation of investors with high sunk-costs and discount rates creates an environment conducive to corruption, which in turn undermines the quality of government oversight. Monitoring of investor compliance with local laws and of the rate of extraction can be difficult. Projects present environmental and safety risks, and EI projects can have enormous impacts on local communities. The success of efforts to include local communities in decision making or profit sharing depends critically on the quality of local community governance. Such sectoral-level governance problems can be addressed through government capacity building and appropriate policy, legal, and institutional reforms — provided that macrogovernance problems such as weak rule of law are not so severe as to prevent such efforts from having meaningful, sustainable impact. The EI sector is unlikely to be an island of integrity in an otherwise systemically corrupt government. 11. The term “rule of law” is used here in its primary meaning, as “the use of public power in accordance with law.” 8 3. Does the Bank Factor in Governance? 3.1 The World Bank’s mission is to reduce poverty. If the quality of governance is an important determinant of whether the development of the EI sector has a positive or negative development outcome, has the Bank shaped its support for the EI sector in light of the governance challenges faced by the borrower? To what extent has the Bank factored governance into its approach to assistance to the EI sector? 3.2 To answer these questions, the Bank’s approach to the EI sectors from 1993 to 2002 was briefly examined in six countries in light of the quality of macrogovernance and governance in the EI sector itself. With respect to macrogovernance issues, particular attention was paid to the quality of public financial management (including the control of corruption), and the capacity of the government to obey and implement its own laws (rule of law). These factors were chosen because the quality of public financial management bears heavily on the government’s ability to manage EI revenues effectively, while the ability of the government to obey and implement its own laws suggests the likely impact of efforts to reform the EI sector, particularly through changes to law. 3.3 The countries considered were Chile, Ecuador, Ghana, Kazakhstan, Papua New Guinea, and Tanzania. The six were chosen for variation in region, the size and importance of the EI sector, the quality of governance, and the degree of Bank intervention in the sector. Chile, Ghana, Kazakhstan, and Papua New Guinea have important mining sectors, while the potential of Tanzania’s mining sector is only now being developed. Ecuador, Kazakhstan, and Papua New Guinea have important oil sectors. For these countries, the paper reviewed select Bank project documents, reports, economic and sector work, and general literature. Project reports and documents were drawn from the Bank’s document databases, ImageBank and the OED database. Some operational documents, especially those in draft form, were not available for review. 3.4 This report relies on the Bank’s formal documentary record to consider the Bank’s decision making processes and the degree to which governance was taken into the account. The evaluation sketched the governance challenges at the macro and sectoral levels, and sought to determine on the basis of the documentary record: • Whether the Bank conducted ESW that would allow it to be well-informed about the borrower’s governance challenges at both the macro and sectoral levels. • Whether the Country Assistance Strategy reflects a strategic approach to the EI sector based on an informed judgment about the quality of governance. • Whether EI projects are consistent with the articulated strategy, and the degree to which they took governance risks explicitly into account. What kinds of decisions did the Bank make about promoting EI investment, versus supporting reforms to improve risk management in the existing sector, or supporting macro public sector reforms in light of the quality of governance? • What conclusions did OED draw about the effectiveness of the Bank’s approach. 3.5 The report draws on previously conducted evaluations of the impact of the Bank’s work in governance and the EI sector to make suggestions of how governance could better be taken 9 into account. No independent evaluation of the impact of the Bank’s projects or work was conducted, and the report comes to no new conclusions regarding impact. 3.6 The Appendix lists the World Bank Group’s Oil, Gas, and Mining projects from FY1993 onward. Only the World Bank’s activities are considered; however, IFC projects and MIGA guarantees are noted for context. 3.7 It should be noted that the Bank’s databases do not constitute the complete record and that electronically available project documents capture only part of the dialogue and strategy that make up operations. But the essentials should be captured in the CAS and project documents. The CAS in particular is “the central vehicle for Board review of the Bank Group’s assistance strategy for IDA and IBRD borrowers.”12 To the extent that the Bank fails to articulate and document its assumptions and strategic choices through the CAS and project documents, the conclusion is not that they do not exist, but rather that they are non-evaluable. SUCCESSFUL PROJECTS WITHOUT DEVELOPMENT IMPACT? 3.8 The Bank’s mission is the reduction of poverty. A Bank project should be justifiable if it has a credible likelihood of contributing to this goal. The outcome of a project should be considered successful if it succeeds in contributing to this goal. But if the Bank overlooks important risk factors, Bank projects may be successfully executed without producing the desired impact. 3.9 The Bank’s 1983 Papua New Guinea Petroleum Exploration Technical Assistance Project sought to attract investment into the country’s petroleum sector. OED rated the outcome as satisfactory, institutional development as modest, and sustainability as highly likely. A second project, the 1993 Petroleum Exploration and Development Technical Assistance Project, sought to improve sectoral management, leading eventually to the introduction of the Oil & Gas Act of 1998. The project addressed such issues as assessment of license holders’ work programs, approval of development plans, promotion of petroleum potential, evaluation of oil field performance, and monitoring of development costs. OED judged the project to have a satisfactory outcome, with high institutional development impact and likely sustainability. 3.10 But despite the success of these Bank projects in building Papua New Guinea’s petroleum sector, OED found that “the progress in managing the growth of the oil and gas industry has not led to sustained economic benefits to the country because of macroeconomic mismanagement of oil revenues.” [OED 2002] This is an assessment shared by many others, including a former Prime Minister of Papua New Guinea who termed the 1990s a “lost decade.” In other words, the Bank had two projects rated as successes that at best did not contribute to development, or at worst, may even have had a negative development impact if it resulted in the wasting of the nation’s nonrenewable resources. How can successful Bank projects have a negative development impact? 12. See World Bank Operational Manual BP 2.11 (January 1995). 10 3.11 Any number of factors could prevent a project from having the desired impact. The Bank operates in an uncertain world, and cannot be held responsible for factors beyond its control that affect the development impact of its projects — unless such factors were foreseeable from inception. This raises four questions: • Whether the Bank recognized or should have recognized that poor management of EI revenues is an important risk factor that could substantially reduce or even negate the development benefit of attracting increased investment in the EI sector. • Whether the Bank recognized or should have recognized that the quality of public financial management was extremely poor. • Whether the Bank considered this and other governance risk factors in its decision to support increased investment. • Whether it is possible to mitigate these risks and whether the Bank took reasonable steps to do so. 3.12 There is no evidence that the Bank considered governance risk factors when deciding to support the expansion of investment in Papua New Guinea’s petroleum sector, much less developed a strategy of risk mitigation. As is discussed below, both projects pre-dated the Bank’s increased focus on governance. 3.13 The Bank dialogue in PNG focused on governance issues as early as 1995, well ahead of the dialogue in most other countries, with substantial ESW and one of the earliest governance-focused CASs. But the Bank continued to work to promote EI investment through the 2000 Mining Sector Institutional Strengthening Technical Assistance Project or the 2000 Gas Development Technical Assistance Project. There is no record that the Bank considered the likely development benefits of such investment in light of governance risks in those projects either, even though public sector mismanagement and an inability to deliver services had been specifically flagged. 3.14 Similar questions are now beginning to be raised concerning the Bank’s assistance to Kazakhstan. The 1994 Petroleum Technical Assistance Project supported the restructuring of the EI sector to promote investment. OED judged the outcome to be moderately satisfactory, with substantial institutional development and likely sustainability. There is no question that foreign direct investment in Kazakhstan’s oil sector is important to its dramatic macroeconomic turnaround. But as the revenue flows to a government with weak accountability, it is unclear whether the government of Kazakhstan will be able to manage revenues and risks effectively for the public interest. There is no indication that the Bank considered governance risk factors before deciding to support an objective of attracting increased EI investment in Kazakhstan.13 NO LINK FROM GOVERNANCE ESW, TO CAS, TO EI PROJECTS 3.15 This report finds that most Bank EI interventions in the selected countries from 1993- 2002 were not the result of governance-informed strategies set out in the CAS. This finding 13. IFC investments between 1994 and 2002 also sought to increase EI investment in Kazakhstan. 11 is not surprising given the evolution of the Bank’s focus on governance. Because the Bank sharpened its focus on governance in the late 1990s, there is no reason to think that the Bank’s earlier handling of EI projects was substantially different from its handling of other sectoral projects in that respect. 3.16 The late 1990s marked a major change in the way that the Bank thought of its mandate, and its approach to governance issues. [OED 2001 34] Previously, corruption was rarely discussed directly in Bank documents for fear of trespassing the boundaries of the Bank’s mandate.14 Political economy analyses were also rare, either because staff lacked the skills to undertake such analyses or because such analyses were similarly suspect. President Wolfensohn put corruption squarely on the Bank’s agenda in 1997 and thereby legitimated its discussion. Good governance — with its focus on transparency, accountability, and participation — is now seen not only as a development goal in itself but also as a prerequisite for the effective use of aid. The Bank’s shifting standards are reflected in recent OED Country Assistance Evaluations covering the 1990s that criticized the Bank for a belated, indirect, or muffled response to obvious governance issues in Ghana, Ecuador, Kazakhstan, and Papua New Guinea. 3.17 Governance ESW was very limited during this period.15 The Bank’s assistance to Papua New Guinea is notable for its early and substantial governance ESW, in particular a 1995 Public Expenditure Review (PER) on the delivery of public services, and a major report on governance in 1999. None of the case study countries had a Country Financial Accountability Assessment or a Country Procurement Assessment Review, products that diagnose aspects of borrower public financial management, and which are now considered necessary for due diligence.16 3.18 The finding of the paucity of governance-related diagnostic work is consistent with similar findings regarding ESW in IDA countries and in Low-Income Countries Under Stress (LICUS). [OED 2001; World bank 2002] The lack of governance ESW may be due to the fact that the Bank’s focus on governance became more acute only at the end of the period under review. Indeed, some of these ESW products were only developed in the latter part of the 1990s, such as the Institutional and Governance Reviews (1999), and the Country 14. It is worth repeating here an excerpt from a legal opinion prepared by the Bank’s General Counsel for the Board of Directors in 1990 regarding the extent which the Bank might address governance matters. The opinion stated, “the Bank is prohibited from taking political factors into account in all its decisions. However, when such political factors lead to direct and obvious economic results relevant to the Bank’s work, the Bank may properly take such results into account as economic considerations which only happen to have political causes or origins.” Issues of “Governance” in Borrowing Members — The Extent of Their Relevance under the Bank’s Articles of Agreement (SecM91-131) (February 5, 1991) p. 57, as quoted in OED 2001. 15. This observation is cautioned by the possibility that ESW may have been conducted but not captured in the Bank’s databases. 16. A variety of current Bank analytic products give insight into the quality of public financial management in specific countries. These include the Country Financial Accountability Assessment (CFAA), the Public Expenditure Review (PER), and the Country Procurement Assessment Review (CPAR), which cover various aspects of public financial management. The Institutional and Governance Review has a broader and more integrated approach. 12 Procurement Assessment Review (2000).17 Another explanation may be budget constraint, as noted in the Kazakhstan Country Assistance Evaluation. The LICUS Task Force pointed out that, perversely, the current internal budget allocation formula allocates fewer funds for ESW in countries with poor governance as measured by the CPIA score. [World Bank 2002] The worse the Bank staff think country governance is, the fewer resources they have to find out about it and the challenges it poses to development and assistance 3.19 The lack of governance ESW allows a conclusion that the CASs were, for the most part, not informed by substantial Bank governance ESW, although it does not allow for a conclusion that the Bank lacked knowledge of the quality of governance. Governance ESW is not the only means by which the Bank learns about the quality of governance. Bank staff may come to be aware of governance obstacles through dialogue or project supervision. Consultant reports generated in the course of projects may address or flag governance challenges. The Bank may also learn about governance issues from reports generated by other donors, the government, or from the press. Dichotomies between the Bank’s CPIA evaluations and information available from other sources suggest that in some cases the Bank would benefit from more information. The increasing move to selectivity based on the quality of policies and institutions suggests an affirmative duty on the part of the Bank to seek the necessary information. But in the absence of Bank governance ESW or any other documentation of the information on which the Bank relied, it is impossible to evaluate the degree to which the Bank had or sought out actual knowledge of governance challenges. 3.20 The earliest CASs with explicit discussion of governance issues appeared in the mid- 1990s, such as the 1995 Papua New Guinea CAS, but in most cases, not until the late 1990s or later.18 Of the later, governance-aware CASs, only the 1999 Papua New Guinea CAS also sets out an approach to the EI sector. This is probably because no engagement in the sector was anticipated in the other CASs. Even in the 1999 Papua New Guinea CAS, the link between governance challenges and the choice of the strategic approach to the EI sector is not very clear. In retrospect, the decision to support increased investment in mining and gas should have required more explanation given the OED conclusion that similar assistance in the petroleum sector did not contribute to development in Papua New Guinea. [OED 2000b] 3.21 This report identified ten World Bank EI projects in the selected countries during the period 1993 to 2001.19 Because eight of them pre-dated the Bank’s first governance-focused CAS in the country in which were conducted, they did not follow from a governance- informed strategic approach to the EI sector. Only the two EI projects approved in 2000 for Papua New Guinea were approved after the articulation of a governance-informed strategic approach to the EI sector in the CAS. However, both the strategy and the projects seem uninformed by the knowledge of macrogovernance challenges in Papua New Guinea. 17. See http://www.worldbank.org/html/opr/procure/CPAR.htm (CPAR designated as ESW in 2000); “Institutional and Governance Reviews — — a new type of economic and sector work.” PremNotes No. 75. Washington, D.C.: World Bank (November 2002). 18. This is reflective of the broader change in the content of the Bank’s CASs prompted by its greater focus on governance in the later 1990s. [OED 2001 34] 19. In addition to these projects, there were also IFC and MIGA projects that are outside the scope of this review. World Bank and IFC projects are listed in the Appendix. 13 3.22 Accordingly, of the projects that contained components designed to increase sectoral investment, there is no record that the decision to support increased investment was preceded by a consideration of the governance factors that might bear on the benefits of such investment. There is no record that the decision to support sectoral reforms was preceded by a consideration of the probable success and sustainability of such reforms in light of the quality of macro governance. The Bank is not yet conducting adequate governance analysis and linking its governance analysis to its sectoral operations. 3.23 In sum, the link between governance ESW, a governance-informed strategic approach to the EI sector set out in the CAS, and the design of EI projects was missing in the period under review, with the possible exception of Papua New Guinea. This lack of consideration of governance issues reflects the institutional history of the Bank’s approach to governance. That governance ESW and governance-aware CASs began in the late 1990s in most cases means that such a link is only now becoming possible for the current generation of EI projects. The opportunity is now presented for the Bank to develop a new way to work in the EI sector. 3.24 Procedurally, factoring governance into the Bank’s approach to the EI sector requires a more explicit consideration of governance risks. ESW is needed to gather information about the quality of governance to permit an evaluation of the estimated net development benefit of a project. The CAS should set out an approach to the EI sector in light of governance concerns and challenges. Projects should be justified in terms of a sufficient probability of net development benefit taking into account the impact of governance factors on outcomes. Assumptions and strategic approaches should be explicitly stated to permit later monitoring and evaluation. A GROWING CONCERN FOR GOVERNANCE 3.25 Despite the lack of a documentary trail showing the Bank’s consideration of governance factors, over the course of the decade the Bank has paid more and more attention to the quality of governance both in general and in the sector. Historically, the Bank’s approach to the EI sector appeared to treat attracting increased investment in the sector as the primary goal and a good in itself. The first priority of the 1980s was to support privatization of state-run EI sectors. The Bank argued that state-run mining concerns had not been managed to maximize productivity, but rather were subject to corruption and political interference, while revenues were consumed. 3.26 Once the state-run sectoral enterprises had been privatized, the Bank focused on promoting additional sectoral investment. For example, the Bank’s 1992 “Strategy for African Mining” stated, “Overall, the main objective of donor intervention in African mining — whether through technical assistance or investment financing — should be to facilitate private investment and help reduce the country and project-related risks for the private investor.” [World Bank 1992b] To support greater private investment, the Bank sought to liberalize EI sectors, privatize state companies, revamp investment legislation and tax regulations to be more attractive to investors, and build institutional capacity to administer the new mining codes. In parallel, the IFC and MIGA encouraged private investment in the sectors by making loans and political risk insurance available to investors. This “investment first” approach predates the institutional changes in the Bank that allowed governance to be 14 diagnosed, analyzed and considered, and it may predate much of the debate about the development impact of EI sectors. 3.27 Despite the lack of an explicit governance-focused approach to the EI sector, over the course of the 1990s, the Bank devoted increasing amounts of attention to sectoral governance issues. The Bank sought to support sectoral reforms that would increase the benefits of the current EI sector, by, for example, improving revenue collection or increasing the productivity of small-scale miners. The Bank also sought to strengthen the government’s ability to manage risks, focusing increasingly on environmental risks, improving the lot of small-scale miners, and later, on community participation and local land issues. The 1992 “Strategy for African Mining” flagged the need to focus on improving environmental, health and safety aspects of mining, and on legalizing and improving the organization of artisanal mining. [World Bank 1992b] Finally, on an independent track, the Bank worked to support better public financial management, and later, better accountability and transparency. By the end of the period under review, the Bank was working on most of the key governance issues tied to better management of the EI sector, and its strategies were increasingly relevant. 3.28 The Bank did not have an articulated strategy for sequencing such interventions in the EI sectors or coordinating them with work done in other sectors.20 Instead, the sequence of interventions was shaped by the institutional evolution of the Bank’s understanding of the issues and its mandate. As a consequence, where the Bank sought to promote investment in the EI sector, it pursued this objective either before supporting better risk management, or in tandem. In Ghana, for example, the 1988 Mining Sector Rehabilitation Project sought to rehabilitate and privatize the State Gold Mining Corporation. According to the implementation completion report, the project paid little attention to environmental issues, reflecting Bank practice at the time. [cited in OED 2003 15] The subsequent 1995 Mining Sector Development and Environment Project sought to support the development of environmental guidelines and strengthen the government’s capacity for monitoring and enforcement. In Tanzania, both objectives were pursued simultaneously in the 1994 Mineral Sector Development Technical Assistance Project. Efforts to support better public financial management, and later, reflecting the increased attention to governance issues, greater accountability, and transparency, went forward on a completely independent track. THE NEED FOR SELECTIVITY AND SEQUENCING 3.29 The Bank should undertake a project only if there is an expected net development benefit at entry.21 The quality of governance is a constraint that must be considered in evaluating the expected net development benefit of an EI project. 20. This is a cross-cutting problem with the Bank’s approach to governance issues. As OED noted in its review of governance interventions in IDA countries, “Bank staff has tended to work on different sectors in isolation even when they are assigned to the same country. Although most institutional issues have systemic roots, sector staff tend to ponder the institutional needs of their sector, often oblivious of institutional approaches being followed by their colleagues in other sectors and other countries.” [OED 2001 53] 21. The counterfactual does not justify undertaking projects with no expected net development benefit. For example, the Bank should not promote investment in the EI sector if it knows that such investment has no expected net development benefit, on the grounds that such investment will take place in any event and would 15 3.30 For some EI projects, such as projects whose principal development benefit is the creation of employment, the benefits are less dependent on the quality of governance, although risk management still requires some governance capacity. But where the principal development benefit is increased government revenue, as is the case for projects whose aim is to promote sectoral investment, the quality of governance also affects the likely benefit. 3.31 Should progress in attracting investment outstrip progress in establishing the governance prerequisites for good development outcomes, the Bank risks facilitating the wastage of the country’s non-renewable resources, as well as contributing to environmental damage, violence, and weakening of the quality of governance itself.22 Moreover, when investment is increased in a poor governance environment, subsequent reforms are likely to become even more difficult. Actors develop vested interests in non-transparent revenue management or lax environmental regulation.23 Economists have long recognized that corruption is “sticky.” And to the extent that government revenue needs are met by foreign direct investment, the Bank will no longer be in a position to press difficult governance issues. [See McPherson] The Bank’s decision to focus its dialogue with Kazakhstan on governance issues comes after foreign direct investment has made the Bank relatively unimportant as a source of finance. 3.32 Because of the inherent costs associated with the exploitation of non-renewable resources, the Bank should distinguish between investment promotion in the EI sector, EI project closure and economic diversification, and improving sectoral management of benefits and risks.24 The Bank’s older “investment first” approach to the EI sector, already being overtaken by increasing attention to governance issues, should be replaced by a more selective policy. 3.33 Governments are likely to have different incentives with respect to the objectives of increasing revenue, maximizing public benefits, and controlling risks. Bank support for increased EI investment in a country where the government is desperate for revenue to pay its external debts is likely to be met with strong government ownership. By contrast, control of environmental damage or the distribution of compensation to property owners are public be worse without the Bank’s involvement. Such an approach casts the Bank as “the lesser of evils.” Moreover, Bank Operational Policy for economic evaluation of investment projects already requires that “[t]o be acceptable on economic grounds . . . the expected present value of the project’s net benefits must not be negative”. O.P. 10.04 (September 1994). 22. Concerns about the use of revenues in poor macrogovernance environments are not limited to revenues produced by EI projects. The same concerns have been raised about the effective use of aid, leading to debates about whether and how the Bank should lend to countries with poor governance. But a distinction must be drawn between these two types of revenue. Aid is renewable and can be refinanced or even forgiven. Revenue from EI projects comes from the exploitation of non-renewable resources. As a consequence, if an EI project fails to deliver the expected development benefit, the country is left poorer. Accordingly, the Bank should be more conservative in its risk-taking when it comes to EI projects of doubtful development benefit. 23. There is also the possibility that the level of corruption will determine what kind of foreign investors come to the sector. Investors whose success strategies depend on co-opting government officials in return for favored treatment may not be interested in subsequent good governance reforms either. 24. This report does not consider the relative merits of public versus private investment. If investment — whatever its form — is not likely to have a net positive development impact, then the Bank should not work to increase it. 16 services. They cost money, and they may run counter to the objective of maximizing investment revenues. They are, moreover, very often public services supplied to small, remote communities far from the government’s political constituency. If the government provides public services poorly overall, or if it lacks the capacity or will to implement its own laws, it is unlikely that it will supply EI-related public services well even if the law requires the government to do so. While the Bank has often had good success in supporting changes to the formal rules of the game (for example, supporting the drafting of laws and regulations), where the rule of law is weak, such changes often have little impact on behavior. Where political power depends on the ability to distribute government largess privately, or where governments are wholly unaccountable to their citizens, the Bank’s efforts to support reform of public financial management to maximize public benefits are likely to meet with resistance. 3.34 If the quality of governance at entry is not adequate to assure that a project has an expected net development benefit, proceeding on the assumption that the Bank can “fix” governance in time to assure project benefits is not a promising risk management strategy. As the Papua New Guinea case illustrates, working to establish the prerequisites for good development outcomes from EI investments in parallel with, or after promoting investment is a risky strategy. The results of support for parallel governance reforms are too uncertain for the Bank to assume that existing governance problems will be addressed in time to assure a positive development benefit from increased sectoral investment that would not otherwise be beneficial. The EI projects in the sample considered here suggest that: • Where the principle of private investment in the sector is agreed, the Bank’s efforts to support market-friendly environments have successfully contributed to greater private investment in the EI sector, as they did in Ecuador, Ghana, Kazakhstan, Tanzania, and Papua New Guinea. • Efforts to support sectoral reform to increase the government’s ability to manage the risks and costs of the EI sector have not always been successful, even in countries with good governance. In the 1992 Environmental Institutional Development Project in Chile, a component to strengthen environmental regulation of the mining sector made little progress because it conflicted with the government’s priority of attracting more investment. The sustainability of the 1995 Mining Sector Development and Environment Project in Ghana was found to be unlikely because of uncertain government commitment. The 1988 Environmental Management Technical Assistance Project in Ecuador was unable to make headway in energy and mines because of the rapid turnover of government officials. • Efforts to support better public financial management have similarly met with uneven results. In Ecuador, the Country Assistance Evaluation found that progress towards CAS objectives was “meager.” In Tanzania, the Bank noted that corruption was worsening. In Kazakhstan, OED found that while a public sector reform loan achieved the technocratic reforms it sought, it nevertheless did not achieve its stated purpose of improving the effectiveness of resource mobilization and improving the efficiency of the use of resources because of the absence of fundamental checks and balances. 3.35 Accordingly, the evaluation of the net development benefit of a proposed project should take existing governance constraints explicitly into account. Given that strengthening governance is time-consuming, uncertain and beyond the Bank’s control, the conservative 17 (and probably more realistic) approach would be to take governance constraints as fixed over the life of the project when estimating the expected net development benefits. It should be made on the basis of the quality of sectoral and macrogovernance as it exists and as it has historically been, rather than as the Bank hopes governance will one day be. 3.36 The next section proposes that the Bank’s approach to the EI sector be nuanced according to the quality of macro- and sectoral governance. 4. Factoring Governance into the Bank’s Approach 4.1 Since “Assessing Aid” [World Bank 1998], the Bank has focused on the importance of selectivity in aid. Selectivity means directing aid to governments that already have good policies and institutions in place. Such governments make more effective use of aid money. But selectivity is not only deciding whether to assist but also deciding how to assist. 4.2 Given that good governance is the key to good development outcomes from the EI sector, the Bank’s approach to the sector — like the Bank’s approach to assistance more generally — should be shaped by the quality of borrower governance. For the purposes of determining the scope of the Bank’s assistance to the EI sector, the Bank should consider both the quality of macrogovernance, and sectoral governance. While good governance has a much broader definition, two aspects of macrogovernance that are particularly essential to determining the net development impact of the Bank’s EI interventions are the quality of public financial management and the rule of law.25 Different strategies are required to assist borrowers with sound macrogovernance and sound sectoral governance; borrowers with sound macrogovernance, but weak sectoral governance; and borrowers with moderate or serious macrogovernance problems. Table 2 sets out possible standards for Bank assistance. Table 2. Shaping Bank assistance in light of the quality of governance Sound macrogovernance Sound macrogovernance Weak macrogovernance Sound sectoral Weak sectoral governance governance Bank assistance Support for increased √ investment Support for individual projects designed to √ √ compensate for sectoral governance weaknesses Support for sectoral governance reforms to maximize benefits and √ √ √ minimize risks; support for closure of operations Support for strengthened √ √ √ macrogovernance 25. Accountability could also be signaled as an essential element of macrogovernance for the purposes of this discussion; however, accountability is an element of sound public financial management and is necessary for the rule of law. 18 POSSIBLE STANDARDS FOR BANK ASSISTANCE 4.3 Sound macrogovernance, sound sectoral governance. A borrower with sound macrogovernance has satisfactory public financial management, including control of corruption, capacity to manage exhaustible resources, the ability to use revenues effectively and wisely. The government also operates under the rule of law, wielding public power in accordance with law and implementing laws. Sound sectoral governance means that the borrower has a satisfactory legal, regulatory, and institutional framework in place to manage environmental risks, involve and protect local communities, ensure investor compliance with the law, and protect investor contractual rights. Because the borrower has adequate rule of law, government officials wield public power in accordance with law and have the capacity to implement laws. This means that environmental, financial, and compensation regulations are enforced, and that both local community and investor rights are respected. The government’s management of the sector is transparent and subject to effective oversight. 4.4 Where macro and sectoral governance are sound, the Bank can support the borrowers as needed to further improve macro or sectoral governance or to attract new investment. However, very few of the Bank’s borrowers are likely to fall into this category. Where both macro and sectoral governance are sound, the borrower is likely to have access to international financial markets and investors will not need much encouragement to respond. 4.5 Sound macrogovernance, weak sectoral governance. A borrower with sound macrogovernance has satisfactory public financial management, including control of corruption, capacity to manage exhaustible resources, and the ability to use revenues effectively and wisely. However, governance of the EI sector itself is sufficiently underdeveloped or poor to call into question the net benefits of present EI investment or to leave room for substantial improvement. Sectoral governance is weak, with problems such as weak environmental monitoring and control; poor local community participation; lack of fair compensation to landowners for property taking or damage; confused institutional responsibilities, with agencies as both shareholders and regulators. Because the rule of law is adequate, the government can address these shortcomings directly and effectively through legislative, regulatory and policy reforms, as well as capacity building on the basis of expert technical assistance. 4.6 Where borrowers have sound macrogovernance but weak sectoral governance, the Bank should focus its support on helping the government better maximize the benefits, or minimize the risks and costs of the existing sector, including, if necessary, support for shutting down existing operations. An example of this type of assistance is the Bank’s support for better environmental management in Chile. Although governance in Chile is sound, and private investors list Chile as the most attractive mining venue in the world, industrial pollution from mining is posing serious health risks. 4.7 Because there are questions about the net development impact of the sector where sectoral governance is weak, the Bank should not work to expand the sector and attract new investment. However, the Bank may choose to support a specific EI project if the project can be designed so as to compensate for specific sectoral governance weaknesses. Examples include 19 requiring independent and public environmental monitoring, negotiating a compensation scheme for landowners affected by the particular project, or ring-fencing project revenues. A demonstration project of this type may have the effect of raising sectoral standards. 4.8 Weak macrogovernance. If borrower macrogovernance is weak, then the government may be affected by systemic corruption. The quality of public financial management is poor, and revenues are not effectively used for development. The borrower therefore lacks the ability to collect and use EI project revenues wisely. Such borrowers are also likely to have weak sectoral governance, and be unable to control the costs and risks of such projects. Moreover, because the rule of law is weak, the government also lacks the ability to address sectoral problems effectively through legal and institutional reforms. Environmental regulations, even if drafted, are likely to be ignored or used as an opportunity for bribe collection. Funds for local compensation schemes, if developed, are likely to go awry. Management of the EI sector is nontransparent, but no less so than the management of the rest of government. 4.9 The Bank should not promote EI investment in countries with macrogovernance that is so weak that the net development benefit of the project is dubious. If macrogovernance is very weak, increased investment designed to augment government revenues will have little benefit to the extent that the government lacks the ability to manage revenues well. Because the government also lacks the capacity to control the costs and risks, and because resources are non-renewable, the expansion of the sector is unlikely to have a positive development outcome. The Bank is currently experimenting to see whether innovative project design can compensate for poor macrogovernance in countries such as Chad by, for example, requiring the deposit of EI revenues into off-shore accounts to be spent for agreed purposes only when approved by a committee, and with expenditures to be subject to regular audits. [McPherson] It is still too early to tell whether such approaches will be successful, and the governance risks are high. The Bank should allow adequate time for monitoring and evaluation before expanding this approach to other countries. 4.10 Bank support to such countries should focus on resolving macrogovernance issues, and in particular, building effective, transparent, and accountable public financial management and the rule of law — which may in turn require addressing broader or more fundamental issues. Whereas governments with sound macrogovernance can address sectoral governance problems effectively, governments with weak macrogovernance cannot. They lack the capacity and the ability to abide by, implement and enforce their own laws. Accordingly, while there may be reasons to seize a window of opportunity for sectoral reforms, the Bank must do so as the result of a deliberate decision to work to establish conditions that are necessary, but not sufficient to allow EI projects to contribute to sustainable development. One reason to do so might be to maintain a presence in the sector if this is deemed important. The Bank may also support the closure of existing EI operations or reduction of environmental impacts where advisable. 4.11 Because of the Bank’s focus on helping poor countries, and the links between poverty and poor governance, a majority of the Bank’s borrowers will fall into this category. Oil and gas intensive countries in particular feature high corruption and weak rule of law. See Figures 2 and 3. Accordingly, the adoption of this type of selective approach will mean a fundamental reorientation of the Bank’s work in the EI sectors away from prioritizing the attraction of new 20 investment and toward capacity building and technical assistance focused on strengthening the government’s capacity to maximize the benefits and minimize the risks of existing EI investment, or projects in support of mine closure. Figure 2. Percentile ranking of oil and gas intensive World Bank clients in terms of Corruption Control and rule of law from the GRICS II database (100= best, 0= worst). Oil and Gas Intensive World Bank Clients 100 (GRICS II Percentile) Rule of Law 50 0 0 50 100 Corruption Control (GRICS II Percentile) Source: Kaufmann, Kraay and Zoido-Lobatón (KKZ). 2002. “Governance Matters II: Updated Governance Indicators for 2000-01.” Countries’ relative positions are subject to a high margin of error. A country is considered oil and gas intensive for the purposes of this table if oil and gas exports as a share of total exports from 1990- 1999 are greater than 11 percent. Source: World Development Indicators. **Oil and gas revenues as a percentage of exports for the year 2000. Figure 3. Percentile ranking of mining intensive developing countries in terms of Corruption Control and rule of law from the GRICS II database (100= best, 0 = worst). Mining Intensive Clients 100 (GRICS II Percentile) Rule of Law 50 0 0 50 100 Corruption Control (GRICS II Percentile) Source: Kaufmann, Kraay and Zoido-Lobatón (KKZ). 2002. “Governance Matters II: Updated Governance Indicators for 2000-01.” Countries’ relative positions are subject to a high margin of error. A country is considered mining intensive for the purposes of this table if the estimated share of mining exports of total exports during the period 1990-1999 is greater than 15 percent. Source: Mining Department and World Development Indicators. 21 4.12 Because sectoral reforms are not likely to have substantial impact unless macrogovernance issues are addressed, the Bank should ensure that EI sectoral governance concerns are integrated into the Bank’s diagnostic work and dialogue on macrogovernance issues.26 4.13 Assuming that macrogovernance improves with time, a borrower would become a country with sound macrogovernance and weak sectoral governance. The Bank could then work more successfully to strengthen sectoral governance, possibly supporting the introduction of demonstration projects with design compensations to set the sectoral governance standard. As sectoral governance improves, the Bank could work to help the government attract increased sectoral investment. ESTABLISHING AND IMPLEMENTING STANDARDS 4.14 Setting standards. In order to be able to practice selectivity and determine the appropriate approach to the EI sector, the Bank needs to gather information on the quality of borrower macrogovernance and sectoral governance. Key indicators of macrogovernance, from the discussion above, would be those of the quality of public financial management (corruption, one particular public financial management shortcoming, is a possible proxy measure); and rule of law, or the use of public power in accordance with law, as a measure of the government’s ability to address problems through formal institutional reforms. 4.15 Some comparative data are available on macrogovernance quality. Various aspects of governance quality are evaluated in surveys of experts, firm managers, and households. The Bank’s Governance Research Indicators are a measure developed by aggregating available information on the quality of governance from existing surveys. [Kaufmann et al. 2002] The Governance Research Indicators Country Snapshots (GRICS) II contain indicators for Controlling Corruption and the Rule of Law. 4.16 While GRICS II pulls together the sum of what is known, in the case of some countries, this isn’t much. There is an unfortunate relationship between poverty, poor governance, and poor data. In the words of the authors, “Our main finding is that the available data do not permit very precise estimates of governance. . . . The most striking feature . . . is that these confidence intervals are large relative to the units in which governance is measured.” [Kaufmann et al. 1999 15] Figure 4 shows the percentile rankings for the Control of Corruption indicator for six of the top EI dependent governments, with confidence intervals. While Bolivia and Nigeria fall below world average (50 percent), Colombia, Egypt, Guinea, and Mongolia cannot be placed as better or worse than average. Currently available comparative data therefore cannot be the sole basis for determining the quality of governance and the Bank’s approach to the EI sector. 26. The recently completed report of the task force on “World Bank Group Work in Low-Income Countries Under Stress” similarly suggests that the Bank put other objectives on hold, focusing on i) substantial and sustained improvement in policies, institutions, and governance; and ii) improvement in the provision of basic social services. [World bank 2002] 22 4.17 Instead, the Bank will have to evaluate macro and sectoral governance case by case through qualitative analytic work. A variety of current Bank analytic products give insight into the quality of public financial management. These include the Country Financial Accountability Assessment (CFAA), the Public Expenditure Review (PER), and the Country Procurement Assessment Review (CPAR), which cover various aspects of public financial management. The Institutional and Governance Review has a broader and more integrated approach. No generalized Bank analytic products currently exist that are targeted to a better understanding of governance in the EI sector; however, the Bank has conducted sectoral reviews that dealt in part with governance issues. Figure 4. GRICS II Control of Corruption percentile ranking shown with confidence interval for six EI dependent countries. Comparator bar is world average. 4.18 The absence of reliable, objective quantitative data concerning the quality of governance implies that once the Bank gathers information on the quality of governance, it will have to make a qualitative judgment about the gravity of macro or sectoral governance concerns. Instead of using a “bright-line rule,” the Bank must use a standard.27 While the use of a flexible standard raises concerns about the objectivity with which it will be applied, it is certainly better than making no judgment at all.28 27. A “bright-line” rule does not require judgment in its application. An example is a minimum test score needed for a student’s admission to a school. By contrast, a standard describes the factors to be considered and weighed, but requires judgment in its application. An example is an employee evaluation. 28. The Bank is already making such judgments about the quality of policies and institutions in other contexts. The Country Policy Institutional Assessment (CPIA) is used to allocate IDA resources. Note that the CPIA is a measure of the Bank’s opinion about a country’s governance. It is a useful proxy measure for the quality of governance only if the Bank is well-informed and ratings are unbiased. Because the Bank is allocating resources on the basis of its understanding of the quality of governance, it has an affirmative duty to collect information 23 4.19 Implementation. Factoring governance into the Bank’s approach to the EI sector requires a solid base of ESW that would permit the Bank to evaluate the quality of macro and sectoral governance. Various Bank reports have focused on the paucity of governance diagnostic work. While the Bank’s current ESW products would allow some evaluation of the quality of public financial management, no current Bank analytic product allows an evaluation of the rule of law (the degree to which the government exercises power according to law and implements laws), or of sectoral governance. It would be particularly useful for the Bank to develop a diagnostic product aimed at sectoral governance for countries that are EI-intensive. 4.20 On the basis of this information, the Bank could classify the country into one of the three categories above and develop a sectoral assistance strategy accordingly, weighing risks and benefits on the basis of explicit understandings and assumptions regarding governance. In particular, the Bank should distinguish clearly between the objectives of attracting increased sectoral investment, or supporting sectoral reforms to improve the benefits or reduce the risks of the existing sector. It is no longer acceptable to support increased investment everywhere, given the state of the Bank’s knowledge of the mixed results of such investment. The sectoral strategy should be described in the CAS or in a separate document to guide any projects in the EI sectors. Because the quality of governance changes slowly, it should not be necessary to develop a country sectoral assistance strategy more often than every four years. 4.21 Finally, monitoring and evaluation must review (a) the adequacy of the Bank’s governance ESW, (b) the degree to which the Bank’s sectoral assistance strategy is guided by the Bank’s understanding of governance issues and the reasonableness of its assumptions concerning risks and benefits; (c) the extent to which subsequent projects are guided by the sectoral assistance strategy; and (d) the extent to which the strategy itself, and subsequent projects, are guided by lessons learned as shown through monitoring and evaluation. 5. Conclusion 5.1 Increased investment in the EI sector has the potential to bring important development benefits, but it is not a universal good. In fact, the evidence suggests that it is likely to lead to bad development outcomes when governance is poor. This is a known risk, which the Bank must factor into its work in shaping its assistance strategies. 5.2 This report reviewed the Bank’s assistance to the EI sector in six countries from 1993 to 2002 in light of macro and sectoral governance problems. In particular, the evaluation reviewed the documentary record to see whether the Bank is factoring governance into its approach to the EI sector through governance ESW, a governance-informed CAS and sectoral strategy, and the subsequent design of projects. 5.3 This report does not suggest that EI projects were handled any differently than projects in other sectors with respect to governance. Instead, the treatment of governance in the Bank’s so as to be well-informed. Bias may also be a problem, inasmuch as the allocation of resources on the basis of the CPIA may create a conflict of interest for staff surveyed. 24 approach to the EI sector was determined largely by institution-wide changes in focus and approach. The late 1990s marked a fundamental shift in the Bank’s approach to governance issues, legitimizing discussion and analysis of them, and strategic planning in response to them. Governance-focused CASs became increasingly common in the late 1990s. The review of the Bank’s work found that most of the EI projects in the sample pre-dated the Bank’s sharpened focus on governance; accordingly, most EI projects in the sample are not the result of a governance-informed sector assistance strategy. There is no documentary record that EI projects were justified based on an analysis that considered the likely benefits and risks of such investment in light of the quality of governance. 5.4 Instead, the Bank’s sectoral assistance has evolved historically and continued to promote sectoral investment. However, as the 1990s progressed, the Bank subsequently or simultaneously devoted attention to issues such as improving environmental regulation and monitoring, improving revenue collection, or facilitating community participation. In parallel, and on an independent track, the Bank worked on strengthening public financial management. This parallel approach is inherently risky, if it leads the Bank to promote increased EI investment in a country whose government is unable to capture the benefits or control the risks. 5.5 The move to more governance-focused CASs opens new possibilities for the Bank’s work in EI. This report suggests that the Bank must develop a selective strategy in which the objectives of assistance are tailored to the quality of the borrower’s macro and sectoral governance. Doing so is likely to require a shift in the Bank’s support to EI sectors, prioritizing support to governments to maximize the benefits, and control the risks and costs of EI sectors, instead of prioritizing increasing investment. Specifically, this report concludes that: More governance diagnostic work should be conducted. If the quality of governance determines the development outcome of EI activities, due diligence requires the Bank to know about the quality of governance in order to evaluate the benefits and risks of supporting EI activities. Existing comparative data is not yet good enough to substitute for qualitative work. The development of a sectoral governance analytic product would be particularly useful. A sectoral assistance strategy should be developed, tailored to the governance capacity of the borrower. The strategy should set the objectives for Bank support to the sector. The Bank should support an objective of growing the sector only when there is a net development benefit, taking into account the quality of governance as it affects the government’s ability to manage benefits and risks. The sectoral assistance strategy, with a reference to the ESW upon which it relies, and any assumptions concerning governance risk, should be made explicit in the CAS or accompanying document. This will guide future project design, help the Bank have a more consistent approach to the sector, and facilitate later monitoring and evaluation. Monitoring and evaluation should ensure a link between the Bank’s knowledge of governance challenges, a strategy that considers governance risks, and project 25 design. If the Bank’s approach to an EI sector is to be the result of a strategic decision informed by the Bank’s knowledge of the quality of borrower governance, then monitoring and evaluation should evaluate the Bank’s effort to inform itself about the quality of governance, the degree to which a strategic approach was formulated in light of the this knowledge, and the extent to which subsequent projects followed the agreed strategic approach. The evaluation of the net development benefit of a proposed project should take existing governance constraints explicitly into account. Given that strengthening governance is time-consuming, uncertain and beyond the Bank’s control, the conservative (and probably more realistic) approach would be to take governance constraints as fixed over the life of the project when estimating the expected net development benefits. For example, if the primary development benefit of an EI project is increased government revenue, this benefit must be considered in light of the government’s capacity to manage revenue for development. In particular, the Bank should not promote EI investment that would not otherwise be beneficial on the assumption that the Bank’s parallel efforts to strengthen governance will mitigate current risks. The Bank should not promote EI investment in countries whose governments currently lack the capacity to benefit from or manage such investment. Where macrogovernance is so poor that an EI investment is not likely to produce a net development benefit, the Bank should focus on helping the government to strengthen sectoral or macrogovernance. Examples include to strengthen management of EI revenues, support the government’s efforts to implicate local communities in EI decision-making processes, help the government regulate environmental impacts, or mitigate the local impact of mine closure. Where macrogovernance is sound, but sectoral governance is weak, the Bank may support specific EI investments that are specifically designed to overcome the identified sectoral governance weaknesses. Examples of this type of feature are projects that provide for independent audit of environmental impact, or that have specific provisions for community participation, community development, or compensation. If the quality of macrogovernance is sufficiently poor to call into question the net development benefit of EI projects, the CAS should address governance as a priority. This small sample report is necessarily limited. It is retrospective and based on the Bank’s documentary record as contained in Imagebank. As such, it does not consider innovative new work that is currently under development. Moreover, it does not consider Bank strategies that were not documented. If no documentary record exists of Bank strategic choices with respect to governance, one cannot conclude that they did not exist, but rather, that they are non-evaluable. In both cases, the Bank must conduct such analyses and document them if monitoring and evaluation are to be possible. The paucity of governance ESW required the report to rely primarily on available secondary sources for information about the quality of governance in specific countries. Finally, this report does not attempt an independent assessment of 26 the impact of Bank projects. Instead, the report draws on available documentation and secondary literature to suggest some approaches to better factoring governance into the Bank’s work. Soon, projects developed after the introduction of more explicitly governance-focused CASs will be closing and available for evaluation. An evaluation of the impact of these projects, taking governance constraints into account at entry and as they unfolded over the life of the project, would be of interest. 27 Appendix Appendix. World Bank, IFC and MIGA Oil and Gas and Mining Projects from FY1993 World Bank Projects Country Project Name Fiscal Year Project ID Instrument Status IBRD/IDA Commitment US$ Million Ghana Mining Sector 1995 966 SIL Closed 9 Dev. And Env. Cap. Tanzania Mineral Sector 1995 2812 TAL Closed 12 Dev. Papua New Mining Sector 2000 60330 TAL Active 10 Guinea Inst. Strengthening TA Ecuador Mining Dev. 1994 7129 TAL Closed 11 and Env. Control TA Papua New Petroleum 1994 4391 TAL Closed 11 Guinea Exploration and Development Papua New Gas Dev. TA 2000 66954 TAL Active 7 Guinea Kazakhstan Petroleum TA 1994 8501 TAL Closed 16 Kazakhstan Uzen Oil Field 1997 8507 SIL Active 109 Rehabilitation IFC Projects Country Project Name Fiscal Year Project Status Project Size IFC Gross Description US$m Investment Ghana BOGOSU 1993 Gold mine, Active 0 0 Foreign companies Ghana GAGL 1996 Gold mine, Active 11.5 10.06 Australian company Ghana GAGL 1995 Gold mine, Active 13.5 4.5 Australian company Ghana GAGL 2000 Gold mine, Active 13.5 0.54 Australian company Chile Escondida 1999 Copper mine, Active 25 25 RTZ, BHP, others Chile Refimet 1995 Copper Closed 91.2 79 smelter, local/N. American company Chile Refimet 1996 Copper Closed 6 5 smelter, local/N. American company 28 Appendix Country Project Name Fiscal Year Project Status Project Size IFC Gross Description US$m Investment Ecuador Tripetrol 1993 Gas Closed 32 10 development, Argentinean company Kazakhstan Akshabulak 1996 Oil Active 266.9 65.73 development, WG/French company Kazakhstan FIOC 2000 Oil Active 0 .01 development, small US company Kazakhstan FIOC 2000 Oil Active 45 20 development, small US company Kazakhstan Lukoil 2002 Oil Active 575 150 development, Russian company MIGA Guarantee Projects Country Project Fiscal Year Sector Status MAL FDI Enterprise Ghana GSM Gold 1993 Mining Cancelled 9,850,000 71,600,000 Limited Chile Compania 1994 Mining Cancelled 19,800,000 527,400,000 Contractual Minera Candelaria Papua New Southern Gold 1996 Mining Cancelled 10,000,000 892,000,000 Guinea (Bahamas) Limited, Lihir 66,600,000 Gold Limited Chile Minera Los 1998 Mining Cancelled 31,263,750 1,114,000,000 Pelambres Tanzania Kahama 2000 Mining Active 115,830,000 505,300,000 Mining Corporation 2001 56,250,000 Limited 29 References References Auty, R. 1994. “Industrial Policy Reform in Six Large Newly Industrializing Countries: The Resource Curse Thesis.” World Development 22:(1) 11-26. Clarke, G. 2001. “How the Quality of Institutions Affects Technological Deepening in Developing Countries.” Policy Research Working Paper 2603. Washington, D.C.: World Bank. Collier, P. and A. Hoeffler. 1998. “On Economic Causes of Civil War.” Oxford Economic Papers October 50:563-73. Collier, P. and A. Hoeffler. 2001. “Greed and Grievance in Civil War.” Typescript. Coolidge, J. and S. Rose-Ackerman. 1997. “High-level rent-seeking and corruption in African regimes: theory and cases.” Policy Research Working Paper 1780. World Bank: Washington, D.C. Eifert, B., Gelb, A. and N. Tallroth. 2002. “The Political Economy of Fiscal Policy and Economic Management in Oil-Exporting Countries.” Policy Research Working Paper 2899. World Bank: Washington, D.C. Environment Canada. 1996. “Environmental Management in Chile — Report.” (Canadian National Secretariat, Canada-Chile Commission for Environmental Cooperation.) International Finance Corporation. 2003. “Extractive Industries and Sustainable Development: An Evaluation of World Bank Group Experience: Volume III: IFC Experience.” (July 29, 2003). Washington, D.C.: World Bank. Isham, J. M. Woolcock, L. Pritchett and G. Busby. 2002. “The Varieties of Rentier Experience: How Natural Resource Export Structures Affect the Political Economy of Economic Growth.” Typescript. Karl, T. 1999. “The Perils of the Petro-State: Reflections on the Paradox of Plenty.” Journal of International Affairs. 53 (1): 31-48. Kaufmann, D., A. Kraay and P. Zoido-Lobatón. 1999a. “Governance Matters.” Policy Research Working Paper 2196. Washington, D.C.: World Bank. Kaufmann, D., A. Kraay and P. Zoido-Lobatón. 1999b. “Aggregating Governance Indicators.” Policy Research Working Paper 2195. Washington, D.C.: World Bank. Kaufmann, D., A. Kraay and P. Zoido-Lobatón. 2002. “Governance Matters II: Updated Governance Indicators for 2000-01”. Policy Research Working Paper 3106. Washington, D.C.: World Bank. Kaufmann, D., and A. Kraay. 2002. “Growth Without Governance.” Policy Research Working Paper 2928. Washington, D.C.: World Bank. Knack, S. 2000. “Aid Dependence and the Quality of Governance: A Cross-Country Empirical Analysis.” Policy Research Working Paper 2396. Washington, D.C.: World Bank. Leite, C. and J. Weidmann. 1999. “Does Mother Nature Corrupt? Natural Resources, Corruption, and Economic Growth”. Working Paper. Washington, D.C.: International Monetary Fund. Leith, J. 1999. “Botswana: A Case Study of Economic Policy Prudence and Growth.” Working Paper Report 20004. World Bank: Washington, D.C. Marshall, Ian E. 2001. “A Survey of Corruption Issues in the Mining and Mineral Sector.” Mining, Minerals and Sustainable Development Project, International Institute for Environment and Development. (Sept. 7, 2001). 30 References McMahon, G., Evia, J., Pascó-Font, A. and J. Sánchez. 1999. “An Environmental Study of Artisanal, Small, and Medium Mining in Bolivia, Chile and Peru.” World Bank Technical Paper No. 429. (May 1999). World Bank: Washington, D.C. McPherson, C. “Petroleum Revenue Management in Developing Countries.” No date. < http://www2.ifc.org/ogmc/files/ParofPlentyCH2.pdf Moore, M., J. Leavy, P. Houtzager, and H. White. 1999. “Polity Qualities: How Governance Affects Poverty.” Typescript. OED. 2001. “Review of Governance—the Critical Factor.” OED IDA Review. (May 1, 2001). Washington, D.C.: World Bank. (Available at http://wbln0018.worldbank.org/oed/) OED. 2002. “Project Performance Assessment Report: Papua New Guinea, Petroleum Exploration Technical Assistance Project.” (June 20, 2002). Washington, D.C.: World Bank. OED. 2003. “Project Performance Assessment Report: Ghana: Mining Sector Rehabilitation Project (Credit 1921-GH): Mining Sector Development and Environment Project (Credit 2743-GH).” (July 1, 2003). Washington, D.C.: World Bank. Rajkumar, A. and V. Swaroop. 2002. “Public Spending and Outcomes: Does Governance Matter?” Policy Research Working Paper 2840. Washington, D.C.: World Bank. Robinson, J., R. Torvik and T. Verdier. 2002. “Political Foundations of the Resource Curse.” Discussion Paper No. 3422. London: Centre for Economic Policy Research (June 2002). Ross, M. 1999. “The Political Economy of the Resource Curse.” World Politics 51(2) (297-322). Sachs, Jeffrey and Andrew Warner. 1995. “Natural resource abundance and economic growth” NBER Working Paper Series, No. 5398, Cambridge: MA, later published in Gerald Meier and James Rauch (eds.) Leading Issues in Economic Development. New York: Oxford University Press. Sarraf, M. and M. Jiwanji. 2001. “Beating the Resource Curse: The Case of Botswana.” World Bank Environment Department Papers 24753. World Bank: Washington, D.C. United Nations Development Program. 2002. Human Development Report 2002: Deepening Democracy in a Fragmented World. New York: Oxford University Press. United States Energy Information Administration. 2002. “Country Analysis Briefs: Chile.” Usui, N. 1997. “Dutch Disease and Policy Adjustments to the Oil Boom: A Comparative Study of Indonesia and Mexico.” Resources Policy 23:4 (151-162). Wantchekon, L. 1999. “Why Do Resource Dependent Countries Have Authoritarian Governments?” Typescript. World Bank. “Papua New Guinea: Economic Overview.” (Available at www.worldbank.org./html/extdr/offrep/eap/papua_ng.htm). World Bank. “Special Focus: Public Financial Accountability in East Asia.” (Available at http://www1.worldbank.org/publicsector/pe/specialfocus.pdf). World Bank. 1992a. Governance and Development. Washington, D.C.: World Bank. World Bank. 1992b. “Strategy for African Mining” (Executive Summary). World Bank Technical Paper 181. Washington, D.C.: World Bank. World Bank. 1994. “Governance: The World Bank’s Experience.” Development in Practice. Washington, D.C.: World Bank. 31 References World Bank. 1997. World Development Report: The State in a Changing World. Washington, D.C.: World Bank. World Bank. 1998. Assessing Aid—What Works, What Doesn’t and Why. Washington, D.C.: World Bank. World Bank. 2001a. “A Decade of Environmental Management in Chile.” (July 31, 2001). Washington, D.C.: World Bank. World Bank. 2001b. “Poverty and Income Distribution in a High Growth Economy: The Case of Chile 1987-1998.” (August 30, 2001). Washington, D.C.: World Bank. (Available at http://poverty.worldbank.org/library/view/11513/) World Bank. 2002. “World Bank Group Work in Low-Income Countries Under Stress: A Task Force Report”. (September 2002). Washington, D.C.: World Bank.