PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS A case study from Reducing Poverty, Sustaining Growth--What Works, What Doesn't, and Why A Global Exchange for Scaling Up Success Scaling Up Poverty Reduction: A Global Learning Process and Conference Shanghai, May 25­27, 2004 Pakistan: Scaling Up Rural Support Programs Stephen F. Rasmussen M. Mujtaba Piracha Rashid Bajwa Abdul Malik Aadil Mansoor Rural Support Programmes Network House 7, Street 49, Sector F 6/4, Islamabad, Pakistan Email : info@rspn.org.pk Telephone : +92 51.282.2476 The findings, interpretations, and conclusions expressed here 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. Copyright © 2004. The International Bank for Reconstruction and Development / THE WORLD BANK All rights reserved. The material in this work is copyrighted. No part of this work may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or inclusion in any information storage and retrieval system, without the prior written permission of the World Bank. The World Bank encourages dissemination of its work and will normally grant permission promptly. 1 CASE STUDIES IN SCALING UP POVERTY REDUCTION Executive summary Pakistan's Rural Support Program (RSP) movement pioneered bottom-up, community-driven development using a flexible, autonomous, politically neutral approach, which has been replicated successfully. RSPs mobilize and organize communities to stimulate more effective demand for better public goods and services, foster important linkages between the communities and service providers, and at times directly supply services. The establishment of core capital funds provides RSPs with a basis for sustainability and helps them retain autonomy. Over the long run, these programs show a substantial, direct impact on poverty reduction and per capita income growth with indirect improvement in education and health. RSPs have also had significant influence on approaches to local governance, and the adoption of microfinance and community-owned infrastructure as mainstream development strategies. The movement began in 1982 in remote, poor, and sparsely populated rural districts and gradually expanded to include larger districts and even some work in poor urban communities. Today the RSPs work with more than 43,000 community organizations that have more than one million member households. Expansion happened not by scaling up the work of a single organization but through setting up autonomous RSPs, working in different geographical areas, that replicated the RSP approach but had the advantages of being smaller and more adaptable to local needs and conditions. Faced with high poverty levels and inadequate service delivery, especially for the poor, stakeholders in government and civil society began to seek development solutions outside the public domain. In 1982 the Aga Khan Rural Support Program(AKRSP) was established in the Northern Areas of Pakistan. It acted as a catalyst for rural development by organizing communities, working with them to identify development opportunities and promoting the provision of services needed to tackle the specific problems of high mountain regions. Through the combination of direct service delivery and indirect influence on the policies and working of public and private sector players, AKRSP showed remarkable results in a short period of time. AKRSP focused on building up the long-term asset base (physical, human, social, and financial) of rural households on an equitable basis, both directly through its own interventions and indirectly through improving the efforts of government and the private sector. It was through the rapid accumulation and better distribution of such resources that economic growth in the Northern Areas directly translated into reductions in poverty levels and improved the share of the lowest income groups. The demonstrated effectiveness of the approach attracted local and international attention and other rural support programs were established, beginning in 1989. In 1992 the government of Pakistan showed an interest to support the countrywide replication of the RSP model. This culminated in the creation of the National Rural Support Program (NRSP). The economic impact of the work of RSPs is best captured from the work of the oldest RSP. Data from the districts in which the AKRSP works show that incomes were less than one- third of the national average in 1991, rising to more than half of the national average in 2001. While national economic growth slowed considerably in the 1990s, the Northern Areas economy 2 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS experienced an impressive growth in per capita income of 84 percent from 1991 to 2001. The incomes of village organization member households were found to be 15-20 percent higher than those of non-members. There is also evidence of significant improvements in health and education outcomes. What the RSPs have been able to do better than any other large-scale development effort has been to organize poor people and enable them to be included in mainstream development opportunities. The RSP movement has also had an impact on development policies and practices in the country. Today's widespread acceptance of the importance of community-driven development, the growth of community infrastructure as an important means of poverty reduction, and the growth of the microfinance sector were influenced by the RSP model. Much of what the RSPs have tried to do over the past twenty years is reflected in the commitments made by the government in the recently approved Punjab Rural Support Program, Sind Rural Support Organization, and a planned RSP for Azad Jammu and Kashmir. Several factors have driven the scaling up process. First, the existence of an effective model, developed in the field, that clearly demonstrated effectiveness at a grassroots level made it possible to sell the model to government, donors and--importantly--to other poor communities. Second, the RSPs consciously maintained a non-confrontational approach with government and other stakeholders. Advocacy for change was important, but it was done through demonstrating practical, useful results at the community level. Third, during the 1990s several key opportunities to expand the RSP movement were seized where government provided funding directly to begin a new RSP. The alignment of interests required to do this did not materialize often, but each opportunity was crucial to the growth of the movement. Fourth, although the government was instrumental in supporting the growth of the movement, it did not try to control the RSPs it established and instead set them up as independent, self-governing organizations with control over their own decisions and financial resources. Lastly, while the government initially expected the work of these RSPs to largely complement and help improve the delivery of social sector services, the government did not interfere when RSPs took on roles that went well beyond those initial expectations. Factors related to institutional innovation and learning also affected the scaling up of the RSP movement. The key to expansion was the social mobilization approach, the antithesis of the top down blueprint approach to development that was widely practiced until recently. The RSP approach fosters community organizations able to deal with a wide range of issues, not just one specific problem or sector. Reducing poverty requires the flexibility to adopt a holistic approach using multiple interventions. The main lessons that emerge from the RSP movement are: · It was essential to have organizations to support communities in the process of community- driven development, and it works best if these support organizations are autonomous of government and have some independent means of sustaining themselves. 3 CASE STUDIES IN SCALING UP POVERTY REDUCTION · Complementing and supporting government, private sector, and civil society initiatives to reduce poverty and improving service delivery can leverage limited resources for greater effect. · Scaling up through a process of replication, rather than expansion, provides advantages in terms of local ownership and support as well as adaptation to local needs and opportunities. · Impact requires consistent effort over time. In the case of the RSPs this meant at least ten years of concerted effort at the grassroots level before results began to be visible. · People count. A few people have played key leadership roles in sustaining and spreading the core values of the movement, but it was important from the beginning to allow many opportunities for leadership to emerge, rather than remain dependent on very few people. 4 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS Implementation process The model The Rural Support Programme approach is about organizing communities around their needs and strengths. Given the willingness of communities, the support organization helps them channel their resources in a more effective fashion. Broadly speaking, the RSPs play three roles. First, they mobilize and organize communities to stimulate more effective demand for better public goods and services, targeted at both the household and village levels. Second, they foster linkages between organized communities and service providers (government, private sector, or others) for the supply of services. Third, at times they directly supply services where there is a dearth of supply or the supply lacks quality. A Visionary, practitioner and teacher-- The inherent ability and willingness of poor people to improve Dr. Akhter Hameed Khan, originator of their lot is the underpinning assumption the RSP approach in South Asia of the RSP model. While it was first put Over five decades, Akhter Hameed Khan inspired and into practice in an organized fashion in motivated thousands of development professionals in South Asia with his innate qualities and vision, beginning Pakistan in 1982 when the Aga Khan with Comilla Thana Pilot Project, continuing through Rural Support Program (AKRSP) was Daudzai and AKRSP, and culminating with the Orangi established, it was based on ideas and Pilot Project. Akhter Hameed Khan always saw the brighter side. He spoke of the resilience and success of the practices learned from more than a poor and the informal sector and was also a firm believer century of experience, beginning from in making public sector more efficient. He used to advise, the cooperative movement in nineteenth however, that in Pakistan, development will not come from century Germany and continuing on the top. It will come from the bottom and happen in pockets ­ one island formed here, one island there and through the work done in Comilla (then "one will be formed by you." East Pakistan, now Bangladesh) in the 1950s and 1960s. Pakistan's rural development model is based on the approach advocated by Akhter Hameed Khan. The From a poverty reduction experimental phase was embodied in the Orangi Pilot perspective, the assumption behind this Project and the Aga Khan Rural Support Programme. As the RSP movement grew it manifested this model model is that people, especially the countrywide. After his demise, Nishan-I-Imtiaz, the poor, are willing to do many things highest civil award, was conferred on him in recognition of themselves and invest their own his services. resources in order to better their own lives and improve the overall welfare of the communities of which they are a part. The model draws its strength from being participatory in nature rather than representative. In economic terms, the model is geared towards achieving economies of scale for poor communities. 5 CASE STUDIES IN SCALING UP POVERTY REDUCTION The need Poverty in Pakistan has historically shown mixed trends. It declined in the 1970s and 1980s, followed by an increase in the 1990s. Nonetheless, even in the best years of economic growth, the overall incidence of poverty has remained high. Some of the factors behind the general persistence and recent increase of poverty include weak governance, slow economic growth, lack of social services, and degradation of environmental resources (ADB 2002). Today poverty remains at a high level with over 38 percent of families living below the poverty line (CRPRID 2003). From the experience of last three decades, two important lessons cannot be ignored. First, for many reasons including political volatility, weak governance and corruption, successive governments failed to deliver on their basic responsibilities, particularly in reaching the poor and providing adequate social services. Second, even in the best years of economic activity, the so called `trickle down' effect has remained elusive, with most of the benefits of development going to the better off. The poor have mostly been excluded from these mainstream developments. These circumstances led stakeholders in government and civil society to begin to seek development solutions outside the public domain with the objective of supplementing and complementing the government's efforts to combat poverty. The idea was to fill the gap in service delivery wherever government services were weak or even non-existent by seeking solutions in partnership with communities. At the same time, wherever government has been active and economic opportunities were available, such non-government initiatives focused on helping poorer people build their resources and skills to be able to participate more actively in mainstream development. The inception of the RSP movement and its evolution can best be understood in the context of the service delivery gap in the country, on the one hand, and the need for including the poor in mainstream economic activities on the other. Scaling up The Aga Khan Foundation launched AKRSP in 1982 in the Northern Areas of Pakistan. AKRSP acted as a catalyst for rural development by organizing communities, working with them to identify development opportunities, and promoting the provision of services needed to tackle the specific problems of high mountain regions. With the support of a number of donor agencies, AKRSP was able to expand quite rapidly to cover almost all of the five districts in the Northern Areas as well as Chitral District in the North West Frontier Province, eventually working with a population of about one million people widely dispersed over about 72,000 square kilometers in the middle of four of the highest mountain ranges in the world. The World Bank's Operations and Evaluation Department carried out the first of its four evaluations of AKRSP as early as 1987 and highlighted the promise offered by the approach. By 1990 AKRSP had shown respectable progress that was acknowledged by the second World Bank evaluation: "At a time when `rural development' as a development strategy is out of favor, the AKRSP experience provides a hopeful prospect that 6 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS rural development can be made to work, given half-way favorable circumstances" (World Bank 1990). The North West Frontier Province was the first province to show a willingness to try out this rural development model. This led to the establishment of the Sarhad Rural Support Corporation (later renamed as the Sarhad Rural Support Programme, or SRSP) in 1989. SRSP began operations with initial financial support from the provincial government and a bilateral donor. At about the same time, the Pak German project in Baluchistan, which had commenced operations in 1982, was renamed and restructured as the Balochistan Rural Support Programme. In 1992 the Government of Pakistan took the initiative to support the countrywide replication of the RSP model. This culminated in the creation of the National Rural Support Program (NRSP). The government pledged financing of Rs. 1 billion as an endowment fund and awarded the first installment of Rs. 500 million to start operations in October 1992 in eight districts. Due to a change of government soon after, though, further funding to NRSP was discontinued and it had to postpone its ambitious expansion plan. Later on it did expand beyond the initial eight districts as other resources became available. In the mid 1990s, NRSP helped the Water and Power Development Authority (WAPDA) to establish the Ghazi Barotha Taraqiati Idara, a resettlement cum rural development project designed to complement the construction of a large hydro-electric project by working with displaced communities. The most recent examples of RSP replications are PRSP, established by the Punjab Government in 1998, and SRSO, established by the Sind Government in 2003. The evolution of RSPs over the years has mostly taken the form of older RSPs and their senior managers playing important roles in setting up new RSPs. For instance, in the establishment of SRSP, NRSP and PRSP, experienced staff from AKRSP helped to introduce the program to local communities and set up management systems. In the late 1990s, however, it became evident that a more organized approach was needed to support expansion as well as the evolving roles of RSPs. This initially led to the creation of a "resource group" within NRSP, which was later transformed into an independent organization called the Rural Support Program Network (RSPN). The RSPN has a small group of professionals with extensive experience in participatory rural development who provide support to the RSPs in the initial phase of their development and work on matters like expansion and portfolio diversification. More recently the RSPN has played a significant role in policy dialogue at the national level, linking the work of the RSPs with national level initiatives. 7 CASE STUDIES IN SCALING UP POVERTY REDUCTION The RSP Movement 1982 - Aga Khan Rural Support Programme: In December 1982 His Highness, Prince Karim Aga Khan commissioned the establishment of AKRSP. The task for setting up the program was entrusted to Mr. Shoaib Sultan Khan, an ex civil service officer with experience of working in the Comilla project in the 1960s in what was then East Pakistan and the Daudzai project in the NWFP in the 1970s. 1989 - Sarhad Rural Support Programme (initially called the Sarhad Rural Support Programme): When USAID decided to withdraw from Pakistan, the NWFP Government and USAID approached AKRSP to expand its work into other districts of NWFP. Instead AKRSP proposed, and USAID and the government agreed, to establish a new organization to replicate the AKRSP approach. 1991 - Balochistan Rural Support Programme: A GTZ funded project that had begun in the early 1980s was converted into the Balochistan Rural Support Programme in the late 1980s. 1992 - National Rural Support Programme: The opportunity to set up a large-scale replication of AKRSP came in 1991 after the Prime Minister visited AKRSP and subsequently requested Mr. Shoaib Sultan Khan to begin the same kind of program down-country. The federal government provided an initial grant to establish an autonomous NRSP, an approach that was successfully followed later on to set up provincial RSPs. 1995 - Ghazi Barotha Taraqiati Idara: GBTI was set up by NRSP following a commitment by the government, with support from the World Bank, to mitigate problems of communities being displaced by a large hydroelectric project. 1997 ­ Lachi Poverty Reduction Programme: This programme was begun in two districts of NWFP under UNDP's South Asia Poverty Alleviation Programme, with technical and organizational support from SRSP. 1997 ­ Tardeep Rural Development Programme: Save the Children UK had supported a relief and development program in one the most remote desert districts in Sind Province in the mid 1990s. In 1997 that project was transformed into TRDP. 1998 - Punjab Rural Support Programme: By the mid 1990s the relevance of the RSP approach in reducing poverty began to be more widely recognized. The Government of Punjab expressed its desire to have a provincial RSP, for which it committed $10 million, to concentrate on districts that were not already covered by NRSP. 2003 - Sind Rural Support Organisation: The SRSO was established on the same pattern as PRSP in Punjab, with initial funding from the Government of Sind and management support from NRSP. 2004 - Azad Jammu and Kashmir Rural Support Programme: This RSP is approved to be set up in 2004. It follows the same pattern as the Provincial RSPs in Punjab and Sind. 8 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS Figure 1: Scaling Up N u m b e r o f D is tr ic ts 8 0 6 0 4 0 2 0 0 1 9 8 0 1 9 9 0 1 9 9 5 2 0 0 0 2 0 0 3 N u m b e r o f C o m m u n ity O r g a n iz a tio n s 6 0 , 0 0 0 4 0 , 0 0 0 2 0 , 0 0 0 0 1 9 8 0 1 9 9 0 1 9 9 5 2 0 0 0 2 0 0 3 T o ta l M e m b e r In d iv id u a ls 1 ,2 0 0 ,0 0 0 8 0 0 ,0 0 0 4 0 0 ,0 0 0 0 1 9 8 0 1 9 9 0 1 9 9 5 2 0 0 0 2 0 0 3 9 CASE STUDIES IN SCALING UP POVERTY REDUCTION Impact analysis Economic impact: AKRSP Economic impact can best be assessed by examining changes in incomes and poverty in the communities and by calculating rates of return for investments made by RSPs. But to do this requires extensive data collected over long periods of time, something that is only available so far from the AKRSP experience. Evidence on incomes and poverty from northern Pakistan (Northern Areas and Chitral, or NAC) is available from 1991 onwards and indicates significant growth in income per capita. Table 1 shows that incomes were less than one-third of the national average in 1991,rising to more than half of the national average in 2001. While national economic growth slowed considerably in the 1990s, the NAC economy experienced impressive growth in per capita income of 84 percent from 1991 to 2001. Table 1. Trends in Income Per Capita (US$) Pakistan Northern Areas and Chitral NAC as percentage of Pakistan 1991 424 131 31 1994 440 176 40 1997 487 232 48 2001 415 241 58 Source: Government of Pakistan, Federal Bureau of Statistics; AKRSP, Farm Household Income and Expenditure Surveys. This economic growth has had a direct impact on poverty as well. Table 2 shows that while poverty showed a rising trend in the national economy, it dropped dramatically in NAC from about two thirds to about one third of the population from 1991 to 2001. A further investigation into poverty dynamics in NAC reveals that even those who are currently living below the poverty line have benefited from this growth. The indicators for depth of poverty (poverty gap), intensity of poverty (poverty gap index) and severity of poverty all show that the consumption levels of even poor households have improved. Table 2. Trends in Poverty Pakistan Northern Areas & Chitral Head count Head count Poverty Poverty Severity of (percent) (percent) Gap Gap Index Poverty 1991 26.1 67 0.53 0.36 0.75 1994 28.7 54 0.49 0.27 0.55 1997 29.8 45 0.42 0.19 0.41 2001 32.1 34 0.38 0.19 0.27 Source: Government of Pakistan, Planning Commission; Malik and Wood (2003) 10 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS Notwithstanding the usual problems with attribution, the available evidence is strong enough to suggest that the contribution of AKRSP to economic development in NAC has been substantial. For instance, the incomes of village organization member households were found to be 15-20 percent higher than those of non-members. This difference increased with the number of members per household and the length of their membership. The economic rate of return on AKRSP investments also points towards the substantial benefits generated by AKRSP. Using conservative assumptions, the calculated ERR for AKRSP's investment falls in the range of 16 percent to 24 percent, well above the usual experience in similar programs (World Bank 2002). The impact of RSPs is not confined to household economics only. The work of AKRSP also influenced efforts to improve the education and health status of communities in NAC, albeit in a less direct manner. AKRSP often played an active role in forging partnerships between community organizations and Government and NGO service providers to improve social service delivery. The existence of organized communities created numerous opportunities to improve the delivery and quality of services, sometimes through partnerships and other times through efforts directly managed and financed by communities themselves. Table 3 provides some data reflecting the changes in the NAC. Table 3. Selected Social Indicators for Northern Areas Infant Mortality Rate Literacy Rate Primary school Enrolment Rates( (per 1000 live births) ( percent) percent) Boys Girls 1986 162 N/A N/A N/A 1994 50 14 N/A N/A 1997/1998/2002* 33 33 77 62 Note: Infant mortality figures are for 1997, literacy rates for 1998, and primary enrolment rates for 2002 Source: Government of Pakistan; AKHSP; AKESP Economic impact: NRSP In the case of NRSP, which was launched throughout Pakistan in 1992, it has been estimated that membership in a community organization resulted in about 7.5 percent higher household incomes annually. Similarly, NRSP estimates show that poverty levels were lower in member households (Khan 2002). The UNDP-supported Pakistan National Human Development Report 2003 states that 68 percent of the NRSP respondents ate their fill daily, 68 percent ate better than before, 50 percent felt improvements in health, and 82 percent experienced a sustainable increase in income after accessing credit. Finally, a study of 35 rural water supply schemes in Punjab (Khan 1999) concluded that NRSP schemes had a higher success rate in sustainable service delivery than schemes undertaken directly by Government and that such schemes were also more cost effective. 11 CASE STUDIES IN SCALING UP POVERTY REDUCTION Impact on policy The RSP movement has also had an impact on development policies and practices. When the RSP movement began in 1982 there was little talk of community based development in national policies, much less what later came to be called `community driven development'. The RSP experience was a major factor in influencing change and today terms such as community participation, community-based development and community driven development appear in all significant national and provincial development policies and projects, including the recently approved Pakistan PRSP. RSPs have also had significant influence on poverty reduction strategies, approaches to local governance, and the adoption of microfinance and community owned infrastructure as mainstream development strategies. Much of the influence of RSPs has come through demonstrating what could be accomplished in a way that directly affected many communities and households. Impact Issues It was through the combination of direct service delivery and indirect influence on the policies and working of public and private sector players that AKRSP was able to show remarkable results in a short period of time. AKRSP focused on building up the long-term asset base (physical, human, social, and financial) of rural households on an equitable basis, both directly through its own interventions and indirectly through improving the efforts of government and the private sector. It was through the rapid accumulation and better distribution of such resources that economic growth in NAC directly translated into reductions in poverty levels and improved the share of the lowest income groups (Malik and Wood 2003). At the same time, the choice of interventions made by AKRSP had important implications. Through mobilizing communities and improving infrastructure, AKRSP was able to enhance access to markets and to education and health services. Not only did health and education outcomes improve as a result of this integrated development approach but their vulnerability, for example exposure to health shocks or food shortages was also reduced. The improvement in the condition and position of women is another example of how a carefully chosen portfolio of interventions specifically designed for women has played a positive role. It is also important to note that the effect of RSP interventions is best felt when there is better macro-economic growth and when the communities are better linked to mainstream markets. This was demonstrated by the growth performance of incomes in northern Pakistan; higher in the first half of the 1990s and much lower in the second half, following broad national trends (Parvez and Rasmussen 2002). Similarly, better integration with the national markets and improved public policies, for example greater investment by government, clearly increased the effectiveness of AKRSP's interventions. Chitral and Gilgit showed differential gains owing to differential access and public investment in those two districts of NAC (Malik and Piracha 2003). The most recent World Bank evaluation of AKRSP (World Bank 2002) states that AKRSP's high level of inputs (consistent delivery of considerable resources) was effective in delivery of a high level of outputs, had a high economic rate of return, and had a positive, 12 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS measurable impact on reducing poverty. In an underserved area like northern Pakistan, pursuit of a high input model made more sense to create a dent in the widely prevalent and long persistent poverty. A less intensive approach that relies more on increasing the effectiveness of government, private sector and NGO service providers, through forging partnerships with organized communities, could also possibly work well in areas where service providers exist and can be mobilized, and where there is already some degree of integration into mainstream markets. Such a less intensive approach has been adopted by most of the RSPs that followed AKRSP. But measurable reductions in poverty for large numbers of people have remained elusive. Much more remains to be done, perhaps even requiring more resources than RSPs have so far been able to mobilize in order to adopt a high input/high output approach. Driving factors Today the RSPs form the largest group of non-government organizations working for poverty reduction in Pakistan. The RSPs work in over 70 percent of districts and are still expanding their coverage. The factors that have driven this scaling up process are discussed below. The section on "commitment and political economy for change" describes the external environment within which the RSP movement grew while the "institutional innovation, learning and experimentation" section largely deals with how the RSPs have worked. The final section describes external catalysts for the scaling up process. Commitment and political economy for change NRSP enjoys a special relationship with government. This is remarkably congenial at all levels. While this is a great advantage in linking NRSP to line agencies and drawing upon resources from government departments, a careful balance must be maintained to ensure NRSP's independence. (Interim Evaluation of NRSP, UNDP 1998) Developing and marketing the model For the first six years, from 1982 to 1988, there was only one RSP (AKRSP) operating in five remote districts of northern Pakistan. This was the period in which the ideas gleaned from earlier projects such as Comilla and Daudzai were put into practice in the field, beginning from a few villages and ultimately covering six districts. The ideas and practices adopted at that time continue to form the core of the RSP model. AKRSP had no direct Government backing, in terms of financing, although the Aga Khan had sought and received the support of the President of Pakistan to begin the project and the first General Manager was a well respected person from the bureaucracy who had excellent links within the government. So while the RSP movement began from outside the government, there was support at high levels of government. Even so, AKRSP worked in a remote region outside of the development limelight at that time, allowing it to focus on demonstrating what could be done in villages and first proving itself to people living in the villages. 13 CASE STUDIES IN SCALING UP POVERTY REDUCTION Once having established something on the ground, it became easier for the founders of AKRSP, the Aga Khan Foundation, to `market' this model of rural development to a wider audience, one of the original objectives of the program. Many political leaders, senior government officials and donor representatives in Pakistan, as well as people from outside Pakistan, visited the villages where AKRSP worked. In fact, participants on courses that trained top government officials such as the Civil Service Academy and the National Defense College came to AKRSP to get field experience as part of their training. Gradually, a growing number of influential decision-makers who had some exposure to the RSP model was built up, and as time went on some of these people played important roles in expanding the RSP movement. As it developed, AKRSP greatly benefited from the strong support of the international Aga Khan Development Network. This allowed it to gain credibility more quickly and to establish patterns of governance and management that set the standards for the RSP movement as it expanded. In that sense, AKRSP was the successful pilot project that was key to creating the basis for scaling up. At a national level and beyond, acknowledgement and credibility came through World Bank evaluations of AKRSP that validated the efficacy and impact of the approach. Later on UNDP evaluations of NRSP added to this. A 1998 evaluation of NRSP observed: "In sum, the NRSP model is unique. It is highly responsive to community motives and aspirations within the context of community participation. NRSP has shown that institutional sustainability is an achievable goal through the creation of the endowment and access to a virtually unlimited line of credit [from a commercial bank]. Clearly, NRSP's experiences suggest that it is feasible to unlock the productive and entrepreneurial potential of Pakistan's rural people" (UNDP 1998). Such widely circulated reports put development in Pakistan in a positive light, building credibility with donors as well as pride in a locally developed approach that appeared to be working. The enthusiasm of the partner communities was crucial to building momentum for the process of scaling up. The confidence and exuberance of community members in northern Pakistan was evident to visitors. Beginning in the late 1980s when SRSP was created, a pattern was established whereby influential decision makers who supported the replication of the RSP model did so because they had visited villages where the model had worked. The inclusive nature of the RSP approach was also instrumental in attracting the poor to organize themselves. Once the process of forming organizations took off in an area, it quickly created a demonstration effect and gained a momentum of its own. This process of social mobilization gradually changed a long entrenched cultural attitude that the poor are merely passive recipients of services planned, financed and delivered by others. Through the social mobilization process people were not organized simply to receive services but to be catalysts of change and improvement. Maintaining political neutrality Many poor countries are characterized by volatile and unpredictable political environments. In such environments, it is often important that non-government programs try to remain noncontroversial by keeping themselves politically neutral if they are to be able to work effectively over long periods of time on a relatively large scale, or even to continue to exist. The RSPs maintained their autonomy and integrity through a non-confrontational approach. Advocacy 14 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS for changes in development policies and practices mostly took place through demonstrating that practical, useful results could be achieved. Also, the RSP model was replicated through several different organizations rather than expanding one single, large RSP that might have appeared threatening to established government structures. The focus was on fostering community organizations primarily engaged in practical development work. Despite its growing scale and influence, the RSP movement was largely, although not entirely successful in staying clear of political controversies. Government support and funding for expansion The consistency of the RSP movement that has allowed it to grow and have some impact on poverty reduction has largely been the result of internal factors, while the external environment in which the RSP movement was scaled up was quite volatile and unpredictable, both politically and economically. In such circumstances the RSPs tried to position themselves to take advantage of opportunities to scale up as and when they came along, and during times when the political environment was not so supportive, they continued to do what they could in the field. In general, the RSPs sought partnership with government, at least in terms of seeking high-level government support and identifying ways of complementing government efforts to improve service delivery for the poor. This kind of positioning allowed the RSPs to be opportunistic in taking advantage of the few key opportunities that came along when the political environment was supportive. This was the story of the origins of what have become three of the largest RSPs (SRSP, NRSP and PRSP) as well as two of the smaller, emerging RSPs (GBTI and SRSO). The key people involved in creating these opportunities had been persuaded by visits to AKRSP that the RSP model did work. The most important opportunities to go to scale came through making government funds available to create a new RSP (NRSP, GBTI, PRSP, and SRSO) or through government support to seek donor funding for expansion (SRSP). Government used its own financial resources, or donor funding it had some control over, to establish RSPs that operated entirely outside government control. This kind of support made the RSPs immediately credible to external donors, who watched carefully the lead taken by government and generated additional resources for RSPs. 15 CASE STUDIES IN SCALING UP POVERTY REDUCTION Scaling up: The National Rural Support Program The 1990 elections brought the Pakistan Muslim League to power at the national level. Nawaz Sharif became the Prime Minister and he appointed Sartaj Aziz as Finance Minister. The Prime Minister had previously been Chief Minister of the largest province, Punjab, and had observed how the common government approach of investing heavily in buildings and infrastructure was not good enough by itself to lead to positive outcomes. For example, he noted that there were many empty school buildings in rural villages, or that even if children came to school, teachers were often absent. The Finance Minister had been following the progress of AKRSP from its beginning, having first visited the Northern Areas in 1983. In light of his extensive experience international in rural development when he was Assistant President of the Rome based International Fund for Agricultural Development, he was convinced that this model could be widely replicated in Pakistan. When the first UNDP Human Development Report was published in 1991 showing that while Pakistan was doing relatively well economically, performance on social indicators was woeful at best, pressure mounted on the Government to do something that would quickly rectify this glaring imbalance. In 1991 at the donor roundtable in Paris the Government proposed what eventually became the Social Action Program, a program that was initiated in July 1992 and aimed to improve the "software" side of the Government's social sector efforts. The Finance Minister was of the view that these efforts could not work without a complementary "social mobilization" program to organize and empower communities to be effective partners in the development process. . He also held the view, based on what he knew of AKRSP and on experience in the 1980s of trying to provide social mobilization services through a representation based political structure, that the best way to do this was through support organizations that were autonomous of the government. Although the Government had presented the Social Action Program to the donors, it began implementation in July 1992 with its own funds. The Finance Minister then wrote a summary for the Prime Minister proposing that the model effectively demonstrated by AKRSP be replicated throughout the country by creating a national rural support program. The General Manager of AKRSP, Shoaib Sultan Khan, was invited to make a presentation of the RSP model to the Prime Minister, federal ministers and provincial chief ministers. The Prime Minister gave his approval immediately for a program that would cost Rs. 10 billion and take ten years to establish the presence of the RSPs in all districts of the country. The Finance Minister followed up by writing a note to the Prime Minister proposing that out of the funds earmarked in the budget for the Social Action Programme, Rs. 500 million be used to set up the National Rural Support Program to begin work in eight districts. Instead of sending the note through the normal channels under the signature of the Secretary Finance, the Finance Minister signed the note himself and personally took it directly to the Prime Minister for immediate approval. The funds were released to the new NRSP within two weeks. From the first mention of the idea of setting up NRSP to the release of funds and the beginning of the program had taken only a few weeks and mostly involved just three main actors. The Finance Minister's note actually mentioned Rs. 1 billion for 16 districts based on a plan that would release funds over several years to expand the program into all districts. But the Pakistan Muslim League government was dismissed in April 1993 and no further funds were released to NRSP. The People's Party of Pakistan won the elections in 1993 and tried to exert their control over NRSP. When this could not happen because of NRSP's autonomous legal status, the Government tried to take back some of the funds that had been given to NRSP, but this also could not be done. Despite facing considerable pressure for some time, NRSP was able to continue its work in the districts, gradually building up a successful, credible program. The NRSP precedent was followed later on by the provincial governments to establish the Punjab Rural Support Program in 1998 and the Sind Rural Support Organisation in 2002: an autonomous organization was set up under the Companies Ordinance and Rs. 500 million was given to be used as an "endowment." In fact, the NRSP precedent was also followed by the federal government in the establishment of the Pakistan Poverty Alleviation Fund in 1998. PPAF is an apex institution that provides funding and support to NGOs engaged in poverty alleviation work and has been an important part of the recent scaling up of the RSPs. 16 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS A different approach to scaling up The Ghazi Barotha Taraqiati Idara (GBTI) represents a unique approach to establishing a sustainable RSP based on the need to care for the needs of people being displaced by a large hydroelectric project. In 1994 the Government of Pakistan's Water and Power Development Authority (WAPDA) began the construction of the 1450 megawatt Ghazi Barotha Power generation project on the Indus River, backed by loans from the World Bank and other donors. In order to try to work in a more participatory way with the people affected by this project, especially those facing resettlement (about 60,000 households in three districts), an NGO was established to "promote the participatory development of the communities, villages, lands and settlements directly or indirectly affected by the Ghazi Barotha Hydropower project, and such adjacent areas as may be designated from time to time by the Company's Board, including advocacy of the rights of affectees in matters of (re)settlement and compensation." WAPDA sought help from NRSP to establish GBTI based on the RSP model. GBTI was setup as an autonomous, self-governing organization. Its Board of Directors included representatives elected from the communities as well as people with experience in development and resettlement matters. WAPDA initially provided a grant of $1.5 million that became an `endowment fund', the income from which was used to pay operating costs. NRSP organized GBTI and recruited and trained its staff. Participatory baseline surveys were done in the villages leading to the development of an integrated rural development plan for which an additional $5 million was committed by WAPDA. GBTI fostered the establishment of community organizations that had an unprecedented voice in the overall project. Acting as an intermediary, GBTI was able to help negotiate fair and generally acceptable market prices for land acquired from local people. Previously, all such projects had used an average "registration price" to acquire land through legal means, usually at prices that were 20 percent of market value, an approach that resulted in considerable conflict between WAPDA and affected people. GBTI worked with poor people to prepare land records and assess a fair value for the land. Also many people from the communities were provided jobs by the contractors. In addition, GBTI was able to mobilize funds to carry out many development projects together with community organizations. WAPDA was not always supportive of GBTI, but the autonomy of GBTI allowed it to continue to work and helped it to be more effective. Even though the power project came on stream in 2003, GBTI's endowment enables it to continue to work in the area on a much longer-term basis. Government support for RSP autonomy The few key government people that took decisions to expand the RSP movement agreed that the RSPs must remain free of government control and political or bureaucratic interference, in line with the approach taken by AKRSP. This largely happened through two mechanisms. The first was to establish the RSPs as private, not-for-profit companies regulated by the independent Securities and Exchange Commission that also regulates for-profit companies. This legal mechanism allowed for an autonomous, self-perpetuating governance structure. The second involved providing funding to the new RSPs in one lump sum from a source that did not require ongoing government oversight or audits. These amounts were used by RSPs to establish `endowments' that provided them some core financial stability and security, something that was especially important to sustaining momentum during difficult political and financial circumstances. These factors generally allowed for the RSPs to steadily build momentum and scale through volatile times. There was no consistently supportive environment for the RSPs over the years; rather they had to be hardy creatures that could weather difficult times. When an RSP was not set up in this way (see box below), it led to failure for the RSP movement in a province that remains one of the poorest in the country. 17 CASE STUDIES IN SCALING UP POVERTY REDUCTION Failure to Establish Autonomy The Balochistan Rural Support Programme did not begin life as an RSP although it gradually adopted the RSP approach and later on joined the movement. It began in 1982 as a rural development project funded by the German Government and managed by GTZ. The initial project was largely a success and in 1989 it was transformed into the Balochistan Rural Support Program, registered as a local organization in Pakistan. BRSP was a well managed organization that was establishing itself in Balochistan until the mid 1990s when the main donor began to interfere directly in the management of BRSP. This resulted in problems that eventually forced the closure of most operations. By 1997 BRSP was in serious decline although recent efforts to revive it have begun to meet with some success. In retrospect it is evident that two of the drivers of success from the experience of other RSPs were not put in place. Firstly, the donor retained a strong influence on BRSP even after setting it up as an autonomous local organization, appointing a senior management person and sitting on the Board of Directors. BRSP remained dependent on external donor funding and decisions such that when differences arose between BRSP and the donor, the donor stopped providing funding and BRSP had no core funds of its own to sustain its operations. BRSP was also not set up to be autonomous of government, primarily in the sense that government officers sat on the Board of Directors and also had some influence over decisions about the use of funds. So long as people in key positions were supportive of BRSP, things functioned well, but the organization was vulnerable to bureaucratic interference. Government expectations of RSPs The key decision-makers who created expansion opportunities for RSPs were convinced that governments in the past had not been able to provide adequate or effective economic and social development opportunities for the poor. The solution, in their opinion, lay in empowering communities to influence the creation of better service delivery mechanisms. On the other hand, political leaders who supported scaling up saw in the RSPs an approach that reached out to the poor, something that had positive political value for them. But government expectations of each RSP were not the same, and in some cases were not even clearly articulated. For instance, the creation of NRSP in 1992 was to complement the government's Social Action Programme, an initiative designed to improve primary education, primary healthcare and rural water supply and sanitation services. It was understood that if service delivery was to improve, expanded and improved services must be complemented by organizing the poor to be able to access those services. In the creation of PRSP there was also an expectation that the RSP would complement the development work of government, largely by creating demand for better services at the community level and by establishing community based mechanisms to make better use of services that government did provide. In essence, the government generally viewed RSPs as being able to operate outside political and bureaucratic control to organize communities with an objective of strengthening demand at grassroots level for more effective services from government departments. This complementary role is often still cited as the main reason for the support provided to the RSP movement. The understanding was that by fostering strong community organizations through social mobilization, the RSPs created a "public good", a mechanism that benefited communities as well as government line departments and other service delivery organizations. At the same time, there have always been examples of RSPs acting as service providers themselves, sometimes in parallel with government and sometimes as subcontractors to 18 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS government. The first RSP was supported by outside donor funding and almost from the beginning began to fill gaps in service provision, a role that was especially important in the relatively much poorer and underserved communities of mountainous northern Pakistan. At one time, for example, AKRSP had developed into the largest provider of inputs and advice to farmers in the Northern Areas. Similarly, in the late 1990s RSPs developed into the largest providers of microfinance services in the country, with combined outreach to about 80 percent of all clients of microfinance institutions in the country. There are other examples where RSPs partnered with government to improve management of basic health and education services. Although RSPs were not primarily established as service provider organizations, this role was taken up largely because they could be more efficient and effective providers of services, especially at a basic level where communities could play a direct role in managing and even partially financing service delivery. RSPs and Public Private Partnerships: Changing Roles and Expectations What have more recently come to be called public private partnerships are the basis upon which several of the RSPs were initially established; the government provided seed capital for private, independent organizations that were designed to work in partnership with government programs. The approach taken in setting up these RSPs later on became the basis for establishing other development organizations, for example the Pakistan Poverty Alleviation Fund, the National Commission for Human Development and the Devolution Trust for Community Empowerment. The PPP approach also makes up a large part of the development work undertaken by RSPs and has recently been expanded into the social sectors to try to improve service delivery. The RSPs have been engaged in partnership with many grassroots, poverty reduction projects of the government, with community mobilization and capacity building done by the RSPs in order to provide a base for service delivery by a government line department. Examples of this include the Khusshal Pakistan Programme, area development projects in Punjab, and agriculture development projects in NWFP and Punjab. More recently new forms of public private partnership have been piloted that try to improve the poor record of service delivery in the social sectors. One of the very few `successes' in the Social Action Programme that began in 1992 was the government/AKRSP partnership in the Northern Areas that resulted in establishing 500 `community schools' providing access to primary education, especially for girls, in remote communities. This model was later on adopted and modified by the Punjab Government in partnership with PRSP, and there are now 400 such community schools, with plans for more. In 2001 NRSP entered into a partnership to manage three basic health units in a rural part of southern Punjab on behalf of the government. This experiment was later on expanded to the entire district of Rahim Yar Khan, with PRSP contracting with government to manage all 105 basic health units in the district. While it is too early to judge the impact of this experiment, management of the primary health care system has improved and already other districts have followed suit. There are now six districts in Punjab where management of basic health units has been contracted out to PRSP and a parallel approach for education is just getting off the ground where the RSPs are to create community based school management committees to be responsible to improve the performance of government primary schools in six districts. The Local Government Ordinance 2001 has devolved the management of departments involved in the provision of basic services from provincial to district level, and therefore provides an opportunity for the RSPs to be involved in more partnership agreements with district governments. The ordinance also envisages the creation of community-based organizations called `community citizen boards' to undertake projects for which 25 percent of the development budget would be allocated. Through an MOU signed with the Development Trust for Community Empowerment, the RSPs have been asked to undertake the task of mobilizing these CCBs. 19 CASE STUDIES IN SCALING UP POVERTY REDUCTION RSP influence on government development practices As the RSP model gained more visibility, it began to have a broader influence on development practices. Social mobilization to organize communities became a part of many development projects whereas it had not previously been given much importance. Provincial governments started large area development projects in rural districts using funding provided by some of the multilateral lenders. The first example of an attempt by government to directly manage the replication of the RSP model was when the Northwest Frontier Province received funding from ADB and IFAD to establish the Chitral Area Development Program in 1990. Much bigger projects followed in NWFP as well as other provinces. Government was able to get donor funding partly through the promise of greatly expanding a model that the donors already viewed as being successful. But after spending tens of millions of dollars on many projects, the track record of success is mixed. An attempt was made to parallel the structures, staffing patterns and program portfolio of an RSP but the systems used to manage such projects were not autonomous of normal government procedures and accountability mechanisms. This sometimes resulted in projects that appeared to look like an RSP but did not have the required autonomy to work independently of political interference or often stifling bureaucratic procedures and systems. The public viewed such projects as extensions of the government system and were reluctant to take on an effective partnership role, preferring to treat them as sources of funding to be demanded through the normal representative governance system. So although the RSP model had considerable influence on development thinking and practices, the drivers of success were not always well understood. Institutional innovation, learning and experimentation Expansion is a slow process even with skilled facilitators, as there are few economies of scale. However, expansion by grafting new programs into locations with similar circumstances offers substantial leverage... (Fourth Evaluation of AKRSP, World Bank 2002) Social mobilization lies at the heart of the RSP approach. It is based on the premise that people's potential can best be harnessed through social mobilization, something that is especially true for poor people. It also recognizes development as a continuous, ongoing process rather than a time-bound phenomenon. When the RSP movement began in the early 1980s, it was a departure from the classical, `blueprint' approach to rural development that was widely practiced then, whereby a uniform, modeled set of interventions was carefully planned to try to achieve rapid economic growth. Instead, the RSP approach began from what organized communities identified as their priority needs and required that communities invest some of their own resources before becoming eligible for the benefits of partnership that an RSP could provide, including improved access to services provided by government or other service providers. The approach was one of learning by doing and then adapting to what had been learned, rather than attempting to plan everything in advance. The RSPs share a commitment to maintaining this "purity of concept", a phrase that is often invoked even as the RSPs grow, innovate and develop differentiated programs. Social mobilization focused on the household, bringing households with common interests together into an organization where each one could potentially have a voice. It 20 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS encouraged participation at a small, local level in contrast to the process of representation that characterized elected government systems. Communities were given centre stage by this bottom- up approach. But empowerment was not just about increasing the assets of poor people and improving their access to better services, as important as those things were, there was also an element of strategic empowerment in the sense that people became effectively involved in making decisions that affected their lives. Practical benefits became the means to more important strategic gains. An example of this can be seen in the influence that organized communities had in choosing their local government representatives in the elections held throughout 2001. Many community activists were elected to local government positions. The RSPs did not set out to do things for poor people, rather it was expected that people would do things for themselves, making optimum use of whatever assets they had. Particular emphasis was placed on promoting collective action, building human capital through training people in leadership, management and vocational skills, and increasing financial capital through encouraging people to save money. The basic philosophy behind investing in these various forms of capital at the village level was to make the development process locally sustainable. This also included identifying and developing activists from the communities who could lead the process in their respective communities. As time went on, these activists played an increasingly important role. Expansion through replication Many movements go to scale through expanding a single organization. The RSP movement deliberately chose a different path in which replication of the RSP approach happened by creating new, independent organizations in each new environment. When an opportunity arose for scaling up in the NWFP in the late 1980s, SRSP was created as a separate, autonomous organization, even though the initial request of the provincial government and the donor was for AKRSP to expand its operations to new districts in the province. The same philosophy was applied in the creation of other RSPs, and today the process of encouraging regional autonomy within the larger RSPs is taking localization further. This approach allowed for local ownership to develop, for example provincial governments could provide support for `their own' RSP, as well as for customization of programs to suit diverse conditions in each locality. As noted before, this approach to scaling up has also received more support from government. Eventually there was the creation of a network organization to help support the movement as a whole, sustain necessary links within the movement, and use the collective experience and strengths of the RSPs to influence national policies and practices. Today there appears to be a shared core philosophy and approach, the essence of the RSP movement, married with increasingly differentiated programs. A further manifestation of this approach to scaling up outside Pakistan came when the SAARC countries began the South Asia Poverty Alleviation Program in 1994, creating autonomous implementing organizations in each country. In addition, the Aga Khan Development Network has replicated the RSP approach in several countries where it is engaged in rural development work. 21 CASE STUDIES IN SCALING UP POVERTY REDUCTION A holistic approach RSPs do not mobilize communities for a specific intervention or sectoral program. Instead, they foster community organizations that can deal with a wide range of issues that concern their livelihoods. To the extent that funding constraints can be overcome, an RSP's program in a region is a function of the needs of the communities it works with, and changing needs of the communities reflect in changing priorities for the RSP. In the AKRSP program, for example, the development of village infrastructure was followed by the demands for skill building programs, which in turn led to the expansion of microfinance and enterprise development activities. Experience shows that reducing poverty requires the flexibility to adopt a holistic approach using multiple interventions. Core capital and leveraging funds Just as with any private company or financial institution, it has been important for RSPs to have core capital that they control and that gives them a basis for sustainability. For some RSPs this came through initial grants provided by government, while some other RSPs have been able to build up core capital over the years through grants provided by donors. In cases where this has not been possible, RSPs have often been left in a precarious position and have struggled to continue their work. Government and donors generally provide grants against which expenditures must be shown. But it can be difficult for an organization to live from hand to mouth in this fashion if this is its only source of funding, whereas an organization with its own capital base can take on such additional funding to expand the work it is already doing. In fact, RSPs have a good track record of leveraging their core funds through the addition of donor or government funded projects. Some RSPs have also used their core capital as collateral to obtain lines of credit from commercial banks to expand their microfinance programs. Core capital allows considerable flexibility to the RSPs and makes it possible to think long-term. Patience and Perseverance Looking back over the twenty years since AKRSP began, it is clear that to have an impact requires longer-term vision and sustained effort. It took AKRSP ten to twelve years to foster community organizations that included 80 percent of all households in six districts, 80 percent coverage being a measure of critical mass that allowed the process of social mobilization to be embedded in the communities and gain momentum of its own. It was at the same time that a positive impact on reducing poverty began to become more evident and measurable. Whether coincidence or not, eleven years after AKRSP began its work, the Northern Areas Government officially adopted an approach to its development work of partnership with the community organizations, recognizing them as part of a community-based movement, not a part of AKRSP. It could also be said that influence on national policies and practices took the same amount of time to take off. Although there were earlier examples of trying to replicate the RSP approach, the creation of NRSP in 1992, ten years after AKRSP began its work, was the first time that government directly and significantly contributed to scaling up the RSP movement. Since RSPs take a holistic approach to development that follows an organic process of growth based on the capacities and initiative of communities, the process of scaling up is neither easy nor predictable. 22 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS Consistency and continuity are required over a long period of time to achieve results at any significant scale. Clearly, personalities have also played a key role in the scaling up of RSPs. This begins from the role of the Aga Khan in establishing and promoting AKRSP and the role of Akhtar Hameed Khan as the thinker and activist who inspired the RSP movement, and goes on to include important politicians and senior government figures. Then there is, of course, the role that has been played by Shoaib Sultan Khan who began the movement as the first General Manager of AKRSP and has been central to the establishment and success of each subsequent RSP. But the RSP movement has not been dependent on any single figure, or even on just a very few people, and it has suffered less from the succession battles or failures that have characterized many organizations and movements. One reason for this has been the considerable investment made in developing competent people to take on management roles. Another reason has been the creation of strong and accountable governance structures that require periodic re-election of Directors and allow for Directors to be changed. Over the years many different people have contributed to the growth of the RSPs by serving on their boards. In addition, the replication approach used by the RSPs has made a particularly useful contribution towards making the RSP movement less reliant on very few personalities. Replication of autonomous organizations put pressure on the RSP movement from its early days to identify and develop quite a few people who could lead independent organizations. Shoaib Sultan Khan: The principal driver of the RSP movement Shoaib Sultan Khan accepted Akhter Hameed as his mentor from the time of the Comilla Project in the early 1960s. Akhter Hameed Khan was ex Indian Civil Service and Shoaib Sultan Khan would in time become ex Civil Service of Pakistan. In the process both acquired a profound understanding of why and where governments fail. Akhter Hameed Khan showed that development could happen in pockets, Comilla and Orangi being the outstanding examples, provided the communities were empowered by `social guidance' to take charge of their own destiny. Shoaib Sultan Khan extended the principles of social guidance to AKRSP and from there to what became the RSP movement. If Akhter Hameed Khan was a social scientist, Shoaib Sultan Khan has been a social activist who believes that the art of development can be practiced country-wide, not just in pockets. Despite occasional setbacks over the years, Shoaib Sultan Khan never wavered from his drive to scale up the RSP movement and make it a success at national and regional levels. 5.3 External catalysts The role of donor community The RSP movement has greatly benefited from strong donor support. For example, from 1983 to 2003 donors provided more than $100 million to AKRSP. Soon after the start of AKRSP, donors realized the potential of the RSP approach and began to provide support. Some donors, in particular CIDA and DFID continued to support AKRSP over many years, providing the continuity that was crucial to achieving results. Donors were also a key part of the process of creating credibility for the movement and convincing government of its value. This was especially important at the beginning of the movement when it would not have been possible to 23 CASE STUDIES IN SCALING UP POVERTY REDUCTION go ahead without donor support. But even after the scaling up process was well under way, donor support for the RSPs continued to be important. The Pakistan Poverty Alleviation Fund In 1998 government created the Pakistan Poverty Alleviation Fund, putting up Rs. 500 million as an endowment from which to pay core, administrative costs and then working with the World Bank to secure a $90 million soft loan funds for programs to be carried out with NGO partners, primarily microcredit and community infrastructure. In its first phase from 1998 to 2004, PPAF was able to spend all of its funds before the end of the planned period, and a much larger second phase with new funding that will double the endowment and also add $238 million from the World Bank has recently been approved. On the one hand PPAF became a source of funding that allowed the RSPs to considerably expand their work, while on the other hand PPAF has been able to expand its programs because the RSPs had already achieved some level of scale and could absorb the large amount of funding being made available. In fact, the RSPs soon became the largest PPAF partners, absorbing more than 60 percent of PPAF funding given out in the first phase. It is expected that PPAF support for RSPs will continue to be an invaluable aspect of future plans to scale up even more. Lessons learned Autonomous and sustainable support organizations The history of the RSPs shows that Government can support scaling up poverty reduction efforts in other ways than doing things exclusively through its own existing service delivery and governance mechanisms. In fact, the RSP experience is about Government support for a scaling up process through organizations that could be effective because they were autonomous of Government control, with independent governance structures and sustainable core funding, as well as relatively free access to donor funding, the use of which was controlled by the RSP, not government. This was not an abdication of government responsibility, rather it recognized that development that had an impact on the lives of poor and marginalized people could often be more effective if it was partly driven by autonomous support organizations. Government is often not in a position to be able to directly manage a social mobilization process itself and in an environment where poor people remained marginalized from mainstream developments for so long, government agencies often did not have the credibility with poor people to form effective partnerships on their own. The RSP movement first worked to establish credibility and trust with poor people, partly through remaining largely free of direct political and bureaucratic interference. This autonomy not only allowed RSPs to replicate structures and systems that had proven to be effective, but it also fostered a process of internalizing values that were essential to partnership with communities. 24 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS Complementing and supporting development efforts The RSPs sought to complement government, private sector and civil society initiatives aimed at reducing poverty and improving service delivery for poor people. Inherent to the RSP movement was a commitment to seeking out and enhancing partnerships that would benefit communities, leveraging limited resources for greater effect. Partnership has taken on many different forms over the years and there has been a constant effort to evolve and change in line with the changing needs of communities and with emerging opportunities. To be successful in doing this it was important that the RSPs remained `a-political' in an environment that was highly charged and politically volatile. This included managing through some periods when the political environment was hostile to the RSPs. The RSPs did position themselves to benefit from government support as and when it was offered, but the political neutrality and autonomy of the RSPs was guarded quite effectively. The RSP movement managed to largely transcend political divides, retaining its base of support within communities and gaining support from other stakeholders on the basis of performance and demonstrated effectiveness. Replication instead of expansion The RSP movement scaled up by replicating the model through many different, local organizations rather than expanding one single large organization. This required having a clear conceptual model and core values that could be expressed with considerable flexibility through several different organizations at the same time. A balance had to be found between standardization of key aspects of the model and diversification of programmatic interventions to adjust to local conditions and opportunities. All RSPs adopted the social mobilization approach that has so characterized the movement, on the one hand, while each RSP was left to find the best fit to its environment, the needs of poor people, and opportunities for partnership and for mobilizing resources. This has led to increasing diversity within the RSP movement and the eventual need for a more formal network to help enhance and sustain core approaches and values. The replication of the RSP model through many organizations of varying size and scope also helped to reinforce the `a-political' nature of the movement since smaller, more localized organizations were better able to gain local ownership and access to local resources while also being less threatening than one large organization might have been. This approach also supported the emergence of a wider set of leaders rather than being dependent on the single leadership of one organization. Impact requires consistent effort over time The RSP approach of providing social guidance and capacity building to poor and marginalized communities created the opportunity for poor people to benefit from macroeconomic growth and development. This complemented mainstream development by creating a better `distribution effect' that benefited poor people. But this was not done overnight. It required an organic and often slow process of growth that clearly demonstrated the potential and effectiveness of what was being done. It relied on the pace at which communities could organize themselves and bring their own resources to bear. The experience of the first RSP, AKRSP, was that it took ten or more 25 CASE STUDIES IN SCALING UP POVERTY REDUCTION years to create `critical mass' in terms of organized communities with the scale to influence public policy and become an effective force for development. It also took ten years to begin to see a measurable impact on reducing poverty. Advocacy for change and for expansion of the work of the RSPs depended on showing success in the field, again something for which time and consistent effort was required. This did not fit well with the usual government and donor approaches to development that expected much quicker results with much higher initial visibility via short term projects. So the RSPs had to find ways to sustain their work to suit much longer time horizons and less top down control focused on achieving pre-planned outputs. Champions are important All movements require leadership in the form of people who are committed to its values and practices. Replication needed champions at all levels of the movement. For the RSPs this came in the form of some influential political figures and people in government who created important opportunities for the movement to grow and scale up. At the same time, the RSP movement had several important, highly visible and influential personalities who provided overall continuity of leadership and made sure that the movement retained its core values. But the RSP movement was also able to mobilize thousands of community activists who saw the potential to change long established and well entrenched patterns of exclusion from the mainstream. It further succeeded relatively early on to attract many competent managers and technical experts, some of whom developed into second generation leaders of the movement as time went on. 26 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS References _____. 1990. The Aga Khan Rural Support Program in Pakistan: A Second Interim Evaluation. A World Bank Operations Evaluation Study. Washington, D.C.: World Bank. _____. 1996. The Aga Khan Rural Support Program. A Third Evaluation. A World Bank Operations Evaluation Study. Washington, D.C.: World Bank. _____. 1998. Interim Evaluation of the National Rural Support Program. United Nations Development Program (UNDP). Islamabad, Pakistan. _____. 1998-1999. First Annual Report, Punjab Rural Support Program (PRSP). Lahore, Pakistan. _____. 2000. An Assessment of Socio-Economic Trends and Impact in Northern Pakistan (1991­ 1997). Findings from AKRSP's Farm Household Income and Expenditure Surveys. Policy and Research Section, AKRSP Core Office, Gilgit, Northern Areas, Pakistan. _____. 2000. Sustainability and AKRSP: Plans for the Future. Aga Khan Rural Support Pakistan. Pakistan. _____. 2002. Poverty in Pakistan: Issues, Causes and Institutional Responses. Asia Development Bank. Islamabad, Pakistan. _____. 2002. The Aga Khan Rural Support Program. The Next Ascent: A Fourth Evaluation. A World Bank Operations Evaluation Study. Washington, D.C.: World Bank. _____. 2003. Northern Areas and Chitral Integrated Household Survey, forthcoming, AKRSP _____. 2003. Operational Insights from Twenty Years of AKRSP (1982-2002). Aga Khan Rural Support Program. Islamabad, Pakistan. _____. 2003. Pakistan Human Condition report 2003, CRPRID, UNDP and UNOPS _____. 2003. Pakistan Poverty Reduction Strategy Paper. Accelerating Economic Growth and Reducing Poverty: The Road Ahead. Government of Pakistan, Islamabad, Pakistan. _____. 2003. Pakistan Poverty Assessment: National Report. Government of Pakistan, Islamabad, Pakistan. _____.n/d. SRSC Evaluation. NORAD. Khan, Shoaib Sultan. 2002. Partnerships with Communities for Integrated Water Resources Management. A paper prepared for International Water Conference, Dialogue on Water, Food and Environment, 13-17 October 2002, Hanoi, Vietnam. Malik, Abdul and Piracha, Mujtaba. 2003. Poverty Trends and Issues in Northern Pakistan, (Paper in Progress, Findings presented at Chronic Poverty Conference, University of Manchester, UK, 2003) Parvez, Safdar and Rasmussen, Stephen F. 2002. Sustaining Mountain Economies: Sustainable Livelihoods and Poverty Alleviation, AKRSP 27 CASE STUDIES IN SCALING UP POVERTY REDUCTION Tahir, Pervez. 2002. Poverty, Participation and Devolution. In "The RSPs of Pakistan: Exploring Options for Working With Local Government", Rural Support Programmes Network, Islamabad, Pakistan. Wood, Geof and Malik, Abdul. 2003. Poverty and Livelihoods. Aga Khan Rural Support Pakistan. Islamabad, Pakistan. Wood, Geof with Shakil, Sofia. 2003. Collective Action: From Outside to Inside. Aga Khan Rural Support Pakistan. Islamabad, Pakistan. 28 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS Annex 1: Acknowledgments and Acronyms In order to try to be more objective and gain wider insight into the process of scaling up RSPs, a working group representing the RSPs, government and others was constituted at the beginning of the process. The members of the working group made considerable contributions to the content of the case as well as commenting upon and improving working drafts. The working group comprised Dr. Pervez Tahir, Mr. Abdul Wajid Rana, Mr. Qazi Azmat Isa, Mr. Farooq Haroon, Mr. Masood ul Mulk and the authors. We are also especially grateful for the time taken by Mr. Shoaib Sultan Khan, Dr. Sartaj Aziz, Mr. Jehangir Khan Tareen, Mr. Tariq Sultan, Mr. Azam Khan and Mr. Kamal Hyat to provide useful insights that helped the authors understand important lessons. We are grateful to Mr. Ghias M. Khan and his staff, especially Ghulam Hafeez at SASMoN for providing the figures and tables as well as technical support. This case study was written and produced with financial support from the World Bank and the Department for International Development, Government of the United Kingdom. Acronyms ADB Asian Development Bank AKES Aga Khan Education Services AKF Aga Khan Foundation AKHS Aga Khan Health Services AKRSP Aga Khan Rural Support Programme BRSP Balochistan Rural Support Programme FHIES Farm Household Income and Expenditure Survey GBTI Ghazi Barotha Taraqiati Idara LPRP Lachi Poverty Reduction Project MER Monitoring, Evaluation and Research Section NAC Northern Areas and Chitral NGO Non-Governmental Organization NRSP National Rural Support Programme PPAF Pakistan Poverty Alleviation Fund PRSP Punjab Rural Support Programme RSPs Rural Support Programmes SGA Sindh Graduates Association SRSO Sindh Rural Support Organization SRSC/SRSP Sarhad Rural Support Corporation, later renamed the Sarhad Rural Support Programme TRDP Thardeep Rural Development Programme UNDP United Nations Development Programme WAPDA Water and Power Development Authority WB World Bank 29 CASE STUDIES IN SCALING UP POVERTY REDUCTION Annex 2: List of Important Meetings Meeting With Relevance Date/ Place Topic Discussed Mr. Sartaj Aziz Former Finance 10th October 2003, Poverty in Pakistan; NRSP's Minister, Govt. of Islamabad establishment Pakistan. Mr. Jehangir Khan Tareen Advisor to Chief 2nd October 2003, PRSP's establishment Minister, Punjab Lahore Mr. Tariq Sultan Former Chairman, 2nd October 2003, PRSP's establishment; Planning and Dev. Lahore reasons for Govt. support and Deptt. Govt. of endowment and current Punjab situation PRSP Senior RSP working in the 2nd October 2003, Perceptions on the RSP Management most populous Lahore model and the drivers in its province, scaling up innovations in Public Private Partnerships Dr Pervez Tahir Planning and 27th September Evolution of RSP movement Development 2003, Islamabad and how the Govt sees it Division, Govt. of Pakistan Mr Azam Khan Former Additional 3rd October 2003, SRSP's establishment and Chief Secretary Peshawar current situation (Development), Govt of NWFP SRSP Senior The first RSP, 3rd October 2003, Perceptions on the RSP Management replicated by Peshawar model and the drivers in its AKRSP scaling up Mr Shoaib Sultan Khan The founder and first 6th, 8th and 15th Evolution of RSP movement GM of AKRSP and October 2003, and the fundamental concepts the main driving Islamabad/ Bhurban force behind the up- scaling Mr Kamal Hayat Heads the Pakistan 10th October 2003, PPAF as a driver of scale for Poverty Alleviation Islamabad the RSPs Fund; RSPs being its biggest partners NRSP Management and The largest RSP in 4th October, 2003, How goal of poverty Staff Pakistan, working in Murree alleviation was being all provinces and in achieved and what were the various partnerships perceptions about scaling up 30 PAKISTAN: SCALING UP RURAL SUPPORT PROGRAMS Annex 3: RSP Map 31