wP5 07,9 POLICY RESEARCH WORKING PAPER 2969 Fostering Community-Driven Development What Role for the State? Monica Das Gupta Helene Grandvoinnet Mattia Romani The World Bank Development Research Group Public Services and Rural Development and Africa Technical Families Poverty Reduction and Economic Management January 2003 POLICY RESEARCH WORKING PAPER 2969 Abstract States can do much to tap community-level energies and the local level. These alliances can also be very effective resources for development if they seek to interact more in catalyzing collective action at community level, and synergistically with local communities. The broader spin- reducing "local capture" by vested interests. off is creating a developmental society and polity. Using There are several encouraging points that emerge from case studies from Asia and Latin America, Das Gupta, these case studies. First, these powerful institutional Granvoinnet, and Romani show how: changes do not necessarily take long to generate. Second, o State efforts to bring about land reform, tenancy they can be achieved in a diversity of settings: tightly knit reform, and expanding non-crop sources of income can or loose-knit communities; war-ravaged or relatively broaden the distribution of power in rural communities, stable; democratic or authoritarian; with land reform or laying the basis for more effective community-driven (if carefully managed) even without. Third, there are collective action; and strong political payoffs in terms of legitimacy and * Higher levels of government can form alliances with popular support for those who support such communities, putting pressure on local authorities from developmental action. above and below to improve development outcomes at This paper-a loint product of Public Services and Rural Development, Development Research Group, and Poverty Reduction and Economic Management, Africa Technical Families-is part of a larger effort in the Bank to understand how to foster effective community-driven development. Copies of the paper are available free from the World Bank, 1818 H StreetNW, Washington, DC 20433. Please contact Monica Das Gupta, room MC3-633, telephone 202-473-1983, fax 202- 522-1153, email address mdasgupta@worldbank.org. Policy Research Working Papers are also posted on theWeb at http:/ /econ.worldbank.org. The other authors may be contacted at hgrandvoinnet@worldbank.org or mromani@worldbank.org. January 2003. (29 pages) The Policy Research Workmig Paper Series dissenuates the findings of work in progress to encourage the exchange of ideas about development issues An objective of the senes is to get the findings out quickly, even if the presentations are less than filly polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countnes they represent. Produced by the Research Advisory Staff Fostering community-driven development: what role for the state? Monica Das Gupta Development Research Group, World Bank, 1818 H Street NW, Washington DC 20433 Mdasgupta(a,worldbank orgz Helene Grandvoinnet Development Research Group, World Bank, 1818 H Street NW, Washington DC 20433 hjzrandvoinnet(a-).worldbank. ore Mattia Romani Development Research Group, World Bank, 1818 H Street NW, Washington DC 20433 mromani(d&worldbank org This analysis was done as we prepared for writing the chapter on state institutions of the WDR 2000/1 on Poverty. We are very grateful for comments and suggestions received on earlier drafts, from Abhijit Banerjee, Pranab Bardhan, Tony Bebbington, Hans Binswanger, Stephens Commins, Samantha da Silva, Mrinal Datta Chaudhun, Jean-Jacques Dethier, Shantayanan Devarajan, Jean Dreze, Gershon Feder, Jonathon Fox, Ron Herring, Ravi Kanbur, Peter Lanjouw, Nora Lustig, William Mayville, Dilip Mookherjee, Jean-Philippe Platteau, Alice Sindzingre, Judith Tendler, Robert Wade, anonymous reviewers, and participants at the seminars at the World Bank Research Group and Social Development Group. Contents 1. Introduction .........................................................1 2. The State and the Community: reducing the potential for "local capture" ......................4 Patron-client relationships as constraints on collective action: ....................................4 Reducing inequalities through land reform and non-crop sources of income: .............5 Broader ramifications for the community and polity ...................................................7 3. The State and local administrations: creating synergies with communities ..................8 The importance of agency incentives and design: illustrations from Indian and South Korean canal irrigation systems .........................................................8 Case studies of successful efforts to re-organize local government agencies ............ 10 Less successful cases ........................................................ 15 4. Lessons from these case studies ........................................................ 16 Close links between the state and communities, tapping into informal community institutions ........................................................ 17 Changing local and supra-local agencies' mode of operation .................................... 18 Enhancing the incentives of grassroots level workers and their accountability to communities ........................................................ 19 Generating community demand for better public goods and services ....................... 20 Financial, technical and other support to communities ............................................. 20 Sustained political backing for reforms, and payoffs for the state .............; . ............. 20 5. Discussion ................................................ 22 Appendix I ..24 Bibliography ..25 i 1. Introduction A major development challenge facing us today is How can poor developing countries make their institutional settings more conducive to growth and poverty reduction? This has been shown to be key to better development outcomes. Studies have illustrated how development outcomes are shaped by the nature of formal and informal institutions, as well as by the type of political regimes and administrative organization.I However, these analyses tend to focus more on the comparative dynamics of these systems, and less on the more challenging question of how to move from dysfunctional institutional settings to those with greater development effectiveness. Here we focus on a subset of issues related to improving service delivery at community level, and more broadly to helping achieve a transition to better-functioning institutional settings. With growing awareness of the limitations of traditional "top-down" approaches for development and poverty reduction, interest has shifted to the potentially powerful role of community participation in enhancing local public service delivery. This requires understanding the dynamics of communities and of local administrations, and their interaction with higher levels of administration. Communities are characterized by a dense network of multi-stranded relationships, the dynamics of which can obstruct or facilitate development. Involving local communities in the development process has several advantages, because they have direct knowledge of the situation (despite asymmetries in information between members) and a direct stake in the outcomes. Local information can be put to good use in many ways for planning and implementing programs,2 though care must be exercised to avoid negative outcomes (Platteau 2002, Tendler 2000). Local information can also be very useful in monitoring the use of program funds and of service delivery.3 This is especially strong when those being monitored have close ties to the community - sanctions for offences are hard to ignore when they come from people with whom one necessarily has multi-stranded and longstanding relationships. Yet, many communities are characterized by serious power imbalances, which place severe constraints on community-based development. Where the poor are heavily dependent on vertical links with local elites, it is difficult to form the horizontal associations necessary for organizing collective action for the common good. Moreover, there is real danger of "local capture" of development programs, with local elites thwarting efforts at collective action or monopolizing their benefits.4 In such settings, higher levels 1 See for example Bardhan (1996), Evans (1995,1996), Manor (1999), North (1989, 1991), Ostrom, Schroeder, and Wynne (1993), Ostrom and Walker (1997), Putnam (1993) and World Bank (1998). 2 For example, providing credit to rural communities is often possible only with organized use of local knowledge of individual credit-worthiness and community sanctions against defaulters: without this, credit agencies shy away from providing credit to the poor, because of the heavy costs of information-gathering (Kohli 1987:135). 3 Wade (1997), Probe Team (1999). 4 Bardhan and Mookheree (1998, 1999). of government may actually be better motivated to help the disadvantaged than local agencies influenced by local vested interests.5 We argue that the possibility for generating developmental state-community synergies is greatly enhanced by interventions which reduce serious power imbalances within communities - for example through land reform, tenancy reform, and developing non-crop sources of income. This reduces poor people's dependence on local elites, helping prepare communities for initiating collective action, and for collaborating amongst themselves and with government agencies.6 As we illustrate, it takes much more careful design and effort to achieve similar development gains in more inegalitarian-settings. Secondly, we argue that higher levels of the state can catalyze institutional change to improve service delivery at the local level. By forming alliances with communities - enabling them to demand better service delivery from local agencies and monitor their performance - higher levels of government can potentially bypass local vested interests and make local agencies more responsive to citizens (Figure 1). This can help higher levels of the state to achieve their development goals, and, importantly, to gain legitimacy and grassroots support. The literature can sometimes suggest that institutional change is a slow process, and that the emergence of functional institutions is almost a matter of historical accident rather than of deliberate engineering.7 Yet, both history and institutional re-structuring shape institutional outcomes. We discuss examples from East Asia and Brazil, which illustrate how quickly such powerful institutional changes can be wrought. These changes can release local energies for development: catalyzing local initiatives and freeing people to engage in a range of activities to improve their economic and political conditions. Using case studies, we illustrate how the effectiveness of efforts at initiating institutional change at local and community level is enhanced by: * Reducing the potential for "local capture" * Maintaining close links between supra-local, local and community institutional structures, for easy collaboration and flow of information * Changing the mode of operation of central and local government agencies to support decentralized decision-making * Enhancing the incentives of frontline workers, and strengthening accountability to communities * Generating community demand for better public services and goods * Providing financial incentives and technical and other support to communities, with fiscal measures to promote ownership and accountability For example, the federal government in the United States has a long history of protecting minority civil rights better than state governments (Foner 1989). This may be because of ideological reasons, various political and electoral motives, or merely because the center is detached from specific local power relationships. 6 Das Gupta, Grandvoinnet and Romani (2000). 77 North (1989, 1991). 2 * Sustaining strong political backing for these reforms. We begin with a brief analysis of the challenges of fighting "local capture", especially in very hierarchical settings. We continue by illustrating some key obstacles to effective local service delivery and how they can be averted, drawing on Wade's (1982, 1997) comparison of the South Korean and Indian canal irrigation agencies. This is followed by discussion of some country case studies of efforts to redesign local agencies' incentives and increase community participation in service delivery. These cases were selected because they offer examples of successful and sustainable governance reform for community-driven development, under quite different circumstances. We then draw out some of the generalizable lessons from these case studies, indicating reasons for success or failure. Figure 1. State-community alliances for development and improved local public service delivery Central / State government (reward: political support) Organizational streamliningX Political & economic incentives\ Strategic use of resources INFORMATION, & OTHER SUPPORT - stimulate local development initiatives Local public agencies - raise aspirations, demand generation & grassroots workers - information on services offered (rewards: public image, material hfRnpfitv) Communities (rewards: development and nnlitirnl Pmnnwrnrmunt) Pressure for quality services & peer monitoring 3 2. The State and the Community: reducing the potential for "local capture" As many anthropological and economic studies show, strong socio-economic hierarchies seriously undermine the potential for collective action for the common good within communities, and increase the potential for "local capture" of benefits. Societies with considerable income and asset inequalities are typically characterized by sharp conflicts of interest between groups. Quantitative analyses find that collective action works better among people who are relatively equal and have similar needs and resources.8 However, moderate degrees of socio-economic differentiation can be less inimical to collective action. Relatively mild differentiation can actually be helpful by allowing scope for local leadership, as indicated in studies reviewed by Baland and Platteau (1999).9 We describe here how the dynamics of substantial inequalities within communities can hinder collective action. Patron-client relationships as constraints on collective action: Patron-client relationships, with their inter-linked contracts, are a familiar feature of many agrarian societies. The classic example of this is the patronage relationship between landlords and tenants, where a single contract has intertwined within it tenancy and credit arrangements.'0 Analyses of landlord-tenant "labor-tying contracts" have shown how richer, less risk-averse people can prey on the insecurity of the poor. " The crux of the matter is summarized in Breman's (1974) succinct phrase "patronage and exploitation": flipsides of the same logic of exercising one's power as a landlord. For the patron, such noblesse oblige enhances their social and economic power, while ensuring their clients' continued survival and participation in the relationship of vulnerable dependence. Patrons seek to circumscribe individual mobility among their clients, using if necessary the threat of withdrawing insurance and even physical force (Appendix 1). Under such circumstances, collective action is difficult to initiate. The elites will perceive collective action by the poor as a threat to their hegemony, and will strongly discourage it. For their part, the poor are unlikely to want to engage in collective action for fear that the elites will seek retribution or appropriate the gains. The literature on South 8 For example, Banerjee and others (2001) find that sugar cooperatives work less effectively in more unequal communities, and Bardhan (1999) finds the same in the context of local cooperation for managing irrigation systems. 9 See also Olson (1971). 10 See for example Braverman and Stiglitz (1982), Srinivasan (1979), Binswanger and Rosenzweig (1984), Ghatak and Pandey (2000). " For example in West Bengal India, Bardhan and Rudra (1986) concluded that landowners take advantage of the fact that security is the greatest need of the poor, they offer security in exchange for considerable amounts of foregone income and loss of freedom and autonomy. Thus many people would not go to work in a neighboring village despite higher wage rates, because they did not want to risk breakng their ties with their patron. 4 Asia is full of such examples'2, as also that on Latin America countries.'3 In many Latin American settings, the landowner-politician nexus has been successful in resisting reforms or in initiating counter-reforms, and thereby defending their status. Reducing inequalities through land reform and non-crop sources of income: Much can be done to mitigate socio-political inequalities in rural areas, through land reform and diversification into non-crop sources of income - opening up the possibility of effective collective action within a community. This can have enormous indirect impact on growth and poverty alleviation, far greater than the direct equity and efficiency gains of land reform on which studies of land reform are typically focused, and for which the empirical evidence is somewhat mixed.14 Discussing the impact of land reform in South Korea, Whang (1 981: 97-98) argues that the gains of land reform were strictly socio- political: (Land reform) helped to miniaturize farm size ... and in so doing helped the delay of agricultural modernization in terms of productivity and mechanization.... (But it) promoted egalitarianism in rural communities.. .and thus facilitated the disappearance of class-consciousness between landlords and tenants.... Thus, the land reform provided the social and psychological precondition necessary for co-operation among framers and that m turn prepared the way for the rural innovation that was later ignited by the Saemaul Undong. Making social relations within the community more egalitarian was an essential precondition to the imaginative creation of state-community synergies in South Korea, Taiwan, and China, as discussed below. Land reforms facilitated the growth of horizontal developmental associations within South Korean and Taiwanese communities, as well as in China when it opened up to a market economy. As mentioned above, Whang (1981) emphasizes that the potential for South Korea's rapid rural development was generated by 12 See references in the two precedmg footnotes. '3 See for example Deere and Leon 1997; Hall 1994; Kay and Silva 1992; Prosterman et al. 1996; Tannenbaum 1962; and Thiesenhusen 1995b. 14 Some studies have found strong evidence of productivity gams associated with giving sharecroppers and small farmers more secure rights over the land they cultivate. This was found mn analyses of data from Tunisia (Laffont and Matoussi 1995), Ghana (Besley 1995) and China (Lin 1992). In West Bengal state of India, increasing tenants' rights was associated with 17-18% average productivity gains (Banerjee and others 2001). In another study m India, Binswanger and others (1993) concluded that small farms are more productive in settngs with little mechanization and slow technical change. Other studies, however, have found inconclusive evidence of productivity gams (Besley and Burgess 1998). In some settings land reform has been carried out carefully enough to be effective in promoting equity and reducing poverty. Elsewhere, landowning elites have been successful in resisting it through active links with political elites and other evasion tactics. Examples include the counter-reforms initiated in several Latin American countries (Kay and Silva 1992, Thiesenhusen 1995a,b). Even where land has actually been redistributed, it can benefit mostly the upper echelons of the rural poor, increasing the gap between them and the poorest - as in the Philippines (Hayanii, Quisumbing and Lourdes 1990) and in some Latin American countries (Thiesenhusen 1995a,b). 5 the land reforns, enhanced further by equal access to primary schooling. A similar shift in the political economy of intra-community relationships following land reform is well documented in Taiwan's cooperative movement. Cooperatives were established in Taiwan under the Japanese, but they blossomed after the land reforms into much more active developmental organizations. With government support, farmers' organizations flourished, helping peasants to pool their savings, improve irrigation and obtain new technologies, and over time develop rural industry (Ranis and Stewart 1993). Radical land reform is typically politically difficult to carry out, and these East Asian countries were fortunate in the circumstances which facilitated it.15 Where such radical land reform is difficult, even strengthening tenants' rights can help make local social and political relationships more egalitarian. Without undertaking the politically more challenging task of altering legal ownership of land, this seeks to strengthen the bargaining position of tenants vis-a-vis their landlords. It gives tenants greater security of tenure, and often specifies lower ceilings on rents charged. As studies in West Bengal indicate, even this limited change can be effective at empowering people within communities (Banerjee, Gertler, and Ghatak 2001). However, it requires considerable political commitmnent on the part of the state, to overcome landlord resistance as well as tenants' fears of retribution from landlords,16 as indicated in Kohli's description of West Bengal (1987:127-31): In a local struggle involving inherently unequal social actors, sustained external support on the side of the underdog was essential to modify the power balance....The local police have moreover been instructed by the government at minimum to refrain from taking sides with the landlords and, at maximum, to provide the bargadars with protection in conflicts over crop-sharing and disputes over evictions. Wherever the landlords have hired their own "thugs" to settle conflicts by force, the party has attempted to counter by militant mobilization. Similar changes can take place, albeit more slowly, by widening access to agricultural inputs which increase the productivity of small-holders (Appendix 1). Given that large proportions of the rural poor in developing countries are landless or nearly landless - and that per capita land availability is rapidly diminishing in poorer regions of the world - it is important to look beyond land reform for ways to make the rural poor less dependent on patrons. Key to this is creating more diversified sources of rural income less dependant on owning land, such as livestock rearing and non-crop activities. Disseminating information on such opportunities, providing access to necessary 15 These reforms were carried out after the Second World War in these countries. In China, land reform encountered little resistance given the context of sweeping revolution. In Taiwan and South Korea, the threat of Communist revolution provided a spur to land reform as a bulwark against Communism. The new regimes owed little to pre-existing landowning elites. Taiwan came to be ruled by the Kuomintang from mainland China, with little need for support from local elites. In South Korea, the new political leadership was relatively unencumbered by existing political links, as it took root in the vacuum left by Japan and the chaos of the Korean War. Moreover, the Korean landowning class was deeply discredited by their perceived collaboration with Japanese colonists. Besides, the departing Japanese colonialists left large tracts of land vacant, available for the initial postwar round of land reform. 16 Kohli (1987); Boyce (1987). 6 resources and marketing can all help greatiy in helping the poor diversify and increase their sources of income.17 Help is required with obtaining credit, finding the appropriate inputs, acquiring know-how about production and marketing, ensuring quality control, and reaching wider markets. There is a great deal of evidence that non-crop incomes alleviate poverty in developing countries while contributing to growth.'8 Since income shocks in the agrarian and non-agrarian sectors are often uncorrelated, non-crop incomes also reduce dependence on patrons for help with tiding over income shocks. Expanding the potential for non-crop incomes is also politically far less challenging than land reform or tenancy reform. Broader ramificationsfor the community and polity In a multiplicity of ways, land reform and non-crop opportunities help reduce the potential for local elite "capture", and enable greater mobility in a community as a whole. The most important gain of reducing local power imbalances is that they lay the basis for a more inclusionary and developmental polity. This has been extensively argued in studies of India. Dreze and Sen (1996) have pointed out that the potential for "capture" by local elites is far higher in regions of India still characterized by deep socio-political inequalities than in more egalitarian regions.19 Banerjee and Iyer (2002) have shown that these inequalities are associated with poorer economic and social development outcomes, in terms of agricultural productivity, health and education. Even if the poorest do not benefit directly from the new opportunities, they can benefit enormously indirectly from the social churning generated in the community. The newly-mobile beneficiaries may raise wage rates and hire more labour (Appendix 1), and they take on more remunerative opportunities and free up niches for the landless. They may benefit everyone by seeking greater accountability in the use of local development funds and lobby to ensure that the community receives better public services. The community as a whole is better placed to provide the peer-evaluation and peer-monitoring required for key facilities such as credit co-operatives. Even relatively meager levels of redistribution can produce powerful and long- lasting changes. Studies of the Southern United States show that despite the very limited and truncated efforts at giving land to Blacks during the 1870s: ...well into the twentieth century, blacks who did acquire land were more likely to register, vote and run for office than other members of the rural community (Foner 1989: 109, cited in Herring 1999). 17 This has been successful in many contexts. Through a creative model of dairy co-operatives, the National Dairy Project in India has helped to significantly raise rural household incomes, including those of the landless, marginal and small farmers . See Candler and Kumar (1998) and Sampark (1997). 8 This is shown in many studies in Asia and Africa. See, for example, Lanjouw and Lanjouw (1995), Newman and Canagarajah (2000); and Eder (1993). 19 Harriss (1999) argues that the regions of India in which traditional agrarian power structures have been changed show better development outcomes. 7 Again, in the 1930s, landless black tenants who benefited from very limited land distribution created ". . .a permanent middle class that ultimately emerged in the 1960s as the backbone of the civil-rights movement in the rural South."20 From this discussion of how the state can facilitate collective action at community level, we turn now to what the state can do to create the conditions for effective governance of community-driven development. 3. The State and local administrations: creating synergies with communities The modern state has enormous potential for providing the institutional support, inputs and infrastructure necessary for upward mobility and for triggering local development initiatives. These case studies illustrate how effective alliances can be formed between higher levels of government and communities, to elicit better service delivery from local agencies (Figure 1). This requires careful design of bureaucratic organizations and staff incentives for perforrnance: a large question that has received considerable attention,2' and care to avoid "local capture." We also focus on the relatively neglected area of the incentives of grassroots level staff, who form the direct interface between communities and the state. Here, too, creative mixing of local and supra-local strengths can be highly productive. We begin by illustrating some key obstacles to performance by grassroots level staff and how they can be averted, drawing on Wade's (1982, 1997) comparison of South Korean and Indian irrigation agencies. This is followed by case studies of redesign of local agency incentives, both more successful and less successful. Section Ell draws out the lessons from these case studies, about the key ingredients for success in generating these state-community synergies. The importance of agency incentives and design: illustrations from Indian and South Korean canal irrigation systems Several case-studies illustrate how service delivery improves if the highly complementary strengths of communities and the state are blended: using local strengths in information and monitoring capacity to complement technical and other inputs from higher 22 levels of government. One study which has been especially carefully developed and 20 Salamon (1979:129), cited in Herring (1999). 21 See for examnple Manor (1999) and Bardhan (1997). 22 For examnple, the school program in Himachal Pradesh in India blended the technical skills of higher levels of govermment in designing curricula and textbooks, with community-level monitoring of school (Probe Team 1999). Similarly, the highly successful program to eradicate river blindness in Sub-Saharan Africa combined the logistical strengths of the state for aerial spraying, and with those of communities for managing smooth distribution of drugs through civil wars and other cnses (World Bank 2000a). For less complex tasks, bottom-up collective action is also very powerful, as Hirshman (1 984a,b) has illustrated through empirical examples from Latin Amenca. See also Ostrom, Schroeder, and Wynne (1993), Owen and van Domelen (1998), Barnentos (1999), and World Bank (2000a,b). 8 documented is Wade's comparison of the canal irrigation systems of South Korea and Andhra Pradesh state in South India. In the South Korean canal irrigation system, broad policy was developed at high levels of government, canal construction was managed by technical experts, and routine operation and maintenance carried out by people with strong stakes in the end-user community. The professional staff must belong to the same county as the end-users, and the actual patrollers are part-time staff who must also be rooted as farmers in the communities they serve. The patrollers, whose job it is to control the water flow and to do small-scale maintenance, are selected by the chiefs of the villages in their jurisdiction and must be re-nominated by them each year. When necessary, patrollers can draw on the technical and other support of the large organizational structure above them. As a result, local management of the South Korean canal irrigation system is done by associations with strong links to the state as well as the end-user communities. Moreover, staff salaries and their operational budget derive partly from the water charges collected. There are thus multiple incentives for a South Korean canal patroller to perform well. He has an interest in obtaining irrigation water himself, in avoiding sanctions from his own community, and in ensuring his continued employment. Not subject to transfer, he and his colleagues in the irrigation department are also able to develop strong working relationships and exert peer pressure on each other. By contrast, the Indian system is carefully designed to minimize the scope for collusion between irrigation staff and the farmers who actually use the canal water. The canal patroller must be a full-time employee and cannot farm more than a marginal amount of land. He must not be posted in the jurisdiction in which his native village is located, and must be rotated between posts within six years. His salary derives from the departmental budget, and thus unrelated to water-use. He is recruited by the irrigation department, and after a probationary period his post in made permanent. His incentives are furthered reduced by a virtual absence of promotion opportunities, even though he is expected to spend his life as a full-time employee of the department. The work of Indian canal staff is hampered by an elaborate system of checks and balances which separate control of funds from those who need to expend them. For example, if the canal develops a significant breach and labor needs to be hired to repair it, neither the foreman nor the assistant engineer is authorized to undertake the expenditure without the permission of the executive engineer, who is often not contactable at short notice. As a result "most assistant engineers are inclined to play it safe, allowing the breach to deteriorate until they can contact their superior" (Wade 1997: 124). Designed to avert corruption, these mechanisms have the resulting effect of discouraging initiative, and are counter-productive. Distanced from their clients and hampered from effective performance, the Indian canal patrollers have little positive reinforcement from client feedback or their own sense of job satisfaction. Meanwhile, their bureaucratic superiors have their own problems to deal with. Inadequately protected from political pressures, they are faced with arbitrary and 9 unwelcome transfers if they displease their political bosses. As Wade (1985) puts it "Frequently, the pressures on officials to be corrupt cause them to behave in ways contrary to the ostensible objectives of their departments." These concerns also reduce their responsiveness to requests or alerts from their junior staff, further reducing the latter's incentive to perform. Thus, in sharp contrast to their South Korean counterparts, Indian canal patrollers have little incentive to perform and little opportunity to develop the good working relationships necessary for good performance. These two case studies highlight the need to ensure that staff at every level have the right performance incentives, and the authority required for getting their job done. The Indian government's approach to discouraging corruption actually hampered service delivery and the agencies' responsiveness to the needs of end-users. Case studies of successful efforts to re-organize local government agencies Success has been achieved in very different political and administrative settings, in bringing about innovative institutional re-organization and fostering state-community alliances for development, averting problems of "local capture." Encouragingly, these case studies show that such institutional change can be achieved quite rapidly, given strong political will.23 For example, the above description might suggest that South Korean local government had a long history of good development outreach. Yet, the basic structure that Wade describes was put into place in the early 1 960s, less than two decades before his study. South Korea. Altering the organization and values of local administrative institutions in South Korea required considerable creative organization on the part of the state: Historically, local governments in Korea have been instruments of the central government in charge of controlling local resources, and oriented to law and order. Support of rural development for meeting development needs of rural communities seemed to be a new concept to them... .The organization of local governments has been restructured to introduce a functional transformation from control and regulation to development support administration. (Whang 1981: 17). In the 1970s, the South Korean government also undertook a program of rapid community development, the Saemaul Undong (Whang 1981, Wade 1982). The idea was to galvanize communities into upgrading their own living conditions, while also incorporating them into the central structure of the state. Through wide information- dissemination and propaganda about the prograrn, people were informed about the benefits of the program and encouraged to support it. This was combined at the local level with more direct exhortation and some coercion, reinforced by organized competition whereby more successful communities were rewarded by being allocated more resources by the 23 Simnilar but less dramatic examples of successful generation of state-community linkages for development are to be found in other settings, including in Indonesia and Vietnam (Ranis and Stewart 1993, Evans 1996). 10 state. Community leaders were trained and local officials encouraged to provide back-up and technical support as required. They were also encouraged to interact in many fora with higher officials, such that they could spontaneously develop new ideas and avenues of co- operation: ... the organizational channels provided for Saemaul leaders greatly encouraged the participation of village leadership in the decision-making processes both at the township and county levels. Indeed, a few Saemaul leaders worked as members with the Saemaul Consultation Committee at both township and county levels... .(They were also) given the prnvilege to freely meet with the county-chief or governor if they wanted to do so. (Whang 1981: 96). The results were to improve villages along a wide range of dimensions. Village infrastructure was extensively constructed and upgraded: including roads, water systems, bridges, small-scale irrigation, and electrification. A variety of income-generating projects and savings programs were initiated. Community and domestic hygiene were greatly improved through various measures, such as better cattle housing and improved water management. Moreover, the costs were borne largely by the communities themselves: the response was so enthusiastic that communities contributed around 70% of the total resources required for these projects (Whang 1981: 15). Clearly a community development initiative trying to upgrade living conditions required the active participation of women. In a sharp break from traditional constraints to women's participation in public life, this was encouraged (Whang 1981: 106-10). Women were obliged to participate with men in the village general assembly. The women responded with alacrity. to this new opportunity, becoming a driving force of change: The members of Saemaul Women's Association mitiated a campaign for burning up the cards with which their husbands gambled during the off-season. (They) also organized a campaign for closing down the liquor stores within their villages where their husbands used to dnnk and thus add more hardship to their pains of poverty-stncken housekeeping. The women's associations undertook many activities to improve their families' lives. They operated a small savings project which evolved into a "mothers' bank" with a small revolving credit union. Income-generating activities included running village cooperative stores and collecting waste for sale. Through a village kitchen project, they organized collective cooking in the busy season, and also taught ways of improving family nutrition. China. This model of developing a dense network of ties between communities and local administrations also underlies the stunning growth of township and village enterprises (TVEs) in China during the 1980s and 1990s.24 For various reasons, including a desire to reduce rural-urban migration, the central government took steps to favor industrial development in rural areas, by allowing local communities to organize their own enterprises. Community leaders are elected by the community, but are affirmed by higher 24 See, for example, Pei (1996), Johnson (1991), Oi (1999), Du (1990). The growth of TVEs is slowing down now, as problems of conimunal ownership of enterprises become more apparent. 11 authorities and function as members both of the state administrative hierarchies as well as of their own villages. Through their ties with the administration, community leaders obtain information on economic opportunities as well as help with accessing the credit and other resources for realizing these opportunities. The... .difference between governments of township and villages and governments above township level is that most officials of township and village governments are at the same time members of collective organizations, and are, therefore, also owners of TVEs....village officials can be considered as having double identities; they are members both of China's administrative hierarchy and of their own villages. (Pei 1996: 45-61) Collective action is greatly facilitated in Chinese communities by tightly knit kinship ties and local loyalties, within which framework people are accustomed to co-operating for mutual benefit.25 Community sanctions are used effectively in this system: The characteristics of local governments are also rooted in the organization of Chma's rural communities....because of the low information costs of the social group and the geographic immobility of village officials, the cost of choosing and the cost of evaluating performance are very low. The official cannot make a continued return for himself alone from (his administrative position). Should he try to do so, hls highly valued reputation and its interrelated benefits will be lost. So it is in the official's own long-term interests not to jeopardize them. Thus, the benefits of village officials must be realized through profit maximization of the village coalition. (Pei 1996: 45-61). The potential for using these longstanding social institutions could, however, only be realized after crucial changes in financial incentives. Before the market reforms, communities had little incentive to increase output, as the state would appropriate the surplus. The system of taxation now varies between settings, but a common practice is for taxes to be fixed in advance (Qian and Weingast 1996). Any income beyond that is shared by the community and officials, giving both a big stake in the success of the enterprise. At the same time, the state does not provide bailouts: the community faces hard budget constraints and if they cannot make the enterprise work successfully, they stand to bear the full loss. Their continued reliance on agriculture helps cushion them against industrial setbacks, as well as providing a source of capital for investment. Taiwan. In Taiwan, efforts to build active collaboration between the state and communities helped to simultaneously develop agriculture and rural industry. Following land reform, and with the need for fast and strong development successes to ensure the country's viability, the Taiwanese government stimulated rural communities to rapidly develop its agrarian economy. This is succinctly described in Fei, Ranis and Kuo (1979: 37-71)": 25 See, for example, Johnson (1991). This is greatly aided by the systematic cultural homogenization of the Chinese into membership of the dominant Han group. 26 See also Amsden (1985). 12 The farmers' associations and credit cooperatives, set up by the Japanese to facilitate agricultural extension programs and nce procurement, were top-down institutions dominated by landlords and nonfarmers. As a result, most farmers did not directly benefit from them. In 1952 government consolidated those institutions in multipurpose farmers' associations restricted to farmers and serving their interests. In addition to the original function of agricultural extension, the activities of farmers' associations expanded to include a credit department, which accepted deposits from farmers and made loans to them, and to provide facilities for purchasmg, marketing, warehousing, and processing. The associations thus became clearnghouses for farmers, who controlled and maintained them and viewed them as their own creatures. With skilful adaptation, these associations were also used to engender rapid growth of industrial production based in rural areas (Fei, Ranis, and Kuo 1981). Nor was this constrained by the relatively small volume of internal demand for industrial goods: the state made available the necessary information and other organizational support to enable rural industry to produce for the international market. Farmers' associations were expanded to include new functions - forming part of an organizational framework for providing important support activities for production and exchange processes, such as input supplies, quality control, warehousing, and marketing. In some ways, these associations were used to parallel the roles played by larger trading companies in Japan and the Republic of Korea. Brazil. In Ceara, northeast Brazil, Tendler (1997) describes how the state tried to overcome problems in service delivery by forming an alliance with local workers and communities. It thereby brought pressure on local municipalities from above and below, to improve their performance in areas such as public health, agricultural extension, drought relief and building infrastructure such as schools. This had two major prongs: increasing the accountability of frontline workers, and making finances available directly to communities. The state recruited and trained a fresh cadre of grassroots workers to provide these services, and motivated them highly by publicizing their work and offering much official recognition for their services: Simply by virtue of working in these programs, they were told, they stood above the rest as leaders, and this placed a special burden on them to behave responsibly. All this contributed to a new respect for these workers by the public --- remarkable in a time of widespread contempt for government. (Tendler 1997: 136). Budgetary allocations were made available directly to communities to enhance ownership and accountability, with technical support where necessary: ... the school maintenance and repair program drew neighborhood associations into the process of selecting small contractors. And to enable these associations to informally monitor the repair and reconstruction work, agency technicians instructed them in some basic procedures that contractors should be following, like the proper preparation of wall surfaces for receiving tiles. (Tendler 1997:140). 13 A publicity blitz served as a multi-pronged weapon in this strategy. It helped reinforce respect for the workers. At the same time, it encouraged people to have high expectations of the program and hold workers accountable for their performance. It also inforned people of what services they should receive, so that they could pressurize local government to provide them if the services were not forthcoming. This publicity campaign helped galvanize communities into collective action to improve their public goods and services. Multi-tasking was an important feature of these grassroots workers' job, and this made their services much more attractive to clients. Public health paraprofessionals' primary task was preventive care, but they were also authorized to provide some simple curative care, normally the monopoly of nurses and doctors. Agricultural extension agents were authorized to provide drought relief and to supervise small construction projects in the drought relief program, which is normally the job of engineers and supervisors. Instead of delivering standardized messages about improving farming, these extension workers were encouraged to tackle various problems raised by farmers, including matters such as getting access to an unsympathetic bank. This seeming jumble of tasks, sometimes involving more brokering than expertise, cohered together in a client-centric, problem-solving approach to service delivery. It gave rise to trusting and respectful relationships between clients and public servants... .(Tendler 1997: 138). Lacking the prior extensive land reform of East Asian countries, the state had to tackle problems of landlord interests and their political connections with local government. This created problems, for example, in providing drought relief: "the drought relief programs benefited the region's large landowners, private contractors and suppliers of relief more than those needing jobs, food and water" (Tendler 1997: 49). In the 1987 drought, agricultural extension workers were used to break the grip of patronage in the distribution of drought relief However, sustaining such successes requires much continuing effort, and at least in the early 1990s the powerful vested interests managed to re-introduce some clientelism in drought relief. As Tendler points out, the Brazilian case shows that the government can be a strong moral force in the realm of public service, and it can use this force to get citizens to monitor its workers and help protect them from meddling by local politicians and other powerful interest groups. Dedicated workers needed protection from others, including rent- seeking co-workers, superiors and elected officials, and showed much appreciation for receiving this protection. The state's government was less likely to respond to the demands expressed along localized socio-economic divisions and therefore a better ally of community development projects. All the more so that having services delivered at the local level and being able to claim this success as its own was in the state's political interest. As this model was extended to broader areas of development in Brazil, local political interference has been kept at bay by the state governments' insistence that municipal-level 14 councils for disbursing development funds have at least 80% representation from end-user communities (World Bank 1997). Moreover, if communities feel they are unfairly treated by the municipal councils, they can by-pass them to apply for funds directly from the state government. Gradually, these councils have gained experience and an estimated one-third of them have begun to actively monitor other municipal activities (Coirolo 2000). Less successful cases Mexico: the absence of sustained state support. The Mexican PIDER program, initiated by the federal government in 1973, was a large investment program for the implementation of small-scale, local projects in some of the poorest areas in the country. Small-scale irrigation, soil conservation, fruit-trees plantations, rural roads, fishponds, live stocks units, rural health points, schools, potable water systems, village electrification were the types of activities financed under the program. In order to be able to implement all these small-scales operations, emphasis was put early on in the process towards "promoting, postulating and organizing participation" (Cernea 1983). In describing how this program operated in the state of Nayarit, Fox (1992) says that it was initially very successful, with a dynamic team of community organizers bringing leaders from several agrarian communities together for the first time.27 The team raised awareness among peasants of the losses they sustained due to tied-markets and rent- seeking bureaucrats. The peasants mobilized successfully to obtain better access to credit and cheaper fertilizers, by putting pressure on the state's governor, the branch manager of the official agricultural bank, and the government's fertilizer agency. Following their early successes in asserting their rights, communities joined together and founded a Union in 1975 (UELC, Lazaro Cardenas' Union of Ejidos). However, in 1976, the presidency changed hands and reformists lost power, depriving the movement of its federal allies. It became easy for the state governor to subvert the movement. The movement had been too short-lived to create a sustainable force that could stand up alone -without federal support - against the interests of powerful local authorities. The Philippines: the impact ofpowerful landed interests. Efforts at institutional reform in the Philippines illustrate the problems of initiating reforms where powerful landed interests are at play and the society is riven by wealth disparities. Despite their efforts to centralize power, successive governments have needed the cooperation of the landlords to function (IRIS 2000). In this situation, the rush to organize mayoral elections in the Philippines after the war, before constructing local bureaucratic agencies, resulted in the mayors creating their own nepotistic bureaucracies and "creating a system of local bossism which survives to this day" (Manor 1999: 59). 27 Mexican agrarian reform communities, the ejidos, are institutions to which the government cedes land uses rights. They are corporatist institutions , whlch take decisions through democratic mass membership assemblies, but govermnent officials supervise and often intervene in internal elections. Ejidos can from regional unions to collaborate on agricultural development projects. 15 Rural landlords and urban bosses, acting as local power brokers with substantial control over local coercive and economic resources, developed at each electoral level: municipalities, congressional districts, and provinces (Carino 1998; Sidel 1996, 1997). "Elected officials enjoyed enornous discretion over the emerging State apparatus. Combined with ... the expanding economic role of the state with import-substitution industrialization after the Second World War, it allowed bosses to construct personal fiefdoms and retain control over the state apparatus" (Carino 1998). This was displaced by more centralized bossism during Marcos' regime (1966-85), but has re-emerged after his fall: The new local government code has restored - and indeed strengthened- the prerogatives of municipal mayors and provincial governors over local agencies and resources of the state apparatus. Given the manifold spoils of these multitudinous offices, elections have been fiercely contested, through machine mobilization, vote buying, fraud, manipulation and violence (Sidel 1996). Here mechanisms of monitoring and accountability were not in place to ensure that the benefits of the institutional re-organization of local government were not "captured" by a few. Reforms towards democratic decentralization, including the involvement of local NGOs in decision-making at the local level, have not been sufficient to counteract rent- seeking tendencies of local bosses and entrenched bureaucracies. Some efforts have been made at land reform (Hayami, Quisumbing and Lourdes 1990), which may gradually help pave the way for more effective community-level development. 4. Lessons from these case studies The successful case studies illustrate how government efforts at redesigning the institutional structures of governance altered the incentives and the abilities of local agencies and communities to deliver public goods and to undertake new kinds of development initiatives. In Brazil, the government formed an alliance with local communities to improve the performance of the municipalities in diverse areas like public health, primary education and agricultural extension. In the Republic of Korea, the government's efforts through community development programs improved development and maintenance of both the physical and social infrastructure of the village communities. In Taiwan, the remarkable achievements in the sphere of rural development and industrial growth were achieved mainly through the activities of the multi-purpose farmers' associations promoted by the government in the early 1950s. In China, fiscal decentralization and granting of autonomy to local bodies helped create the basis for the remarkable achievements of township and village enterprises (TVEs). The close and multi-faceted collaboration between the apparatus of the state and community leaders becamne an important factor in these countries' economic development in recent years. Elsewhere, efforts were less successful. In the Philippines, the nexus of relationships between government officials and powerful landowners formed an obstacle to 16 program success. In Mexico, initial successes tapered off because of lack of sustained political support. There are several useful lessons to be drawn from these case studies. Some of the key ingredients for success are summarized here. Close links between the state and communities, tapping into informal community institutions In all these settings, efforts were made to build dense networks of interaction between higher echelons of government, local officials and community leaders - to make available to communities the necessary support with information, technical inputs, marketing, credit etc. needed to implement their own programs of social and economic development. In South Korea, China, and Taiwan, local leaders and officials had overlapping roles as simultaneously members of the community served and of the lowest tier of state administration. In Brazil, grassroots workers were not necessarily members of the community they serve, but were encouraged to develop close links with them. Through these dense networks of interaction, the state and communities were enabled to collaborate in development. The success of the model depends on productive convergence of interest between community members, field outreach staff, and local officials, as well as with successive echelons of the state administration. Through cultivating these networks, people could access more resources to help their communities prosper. The trick is to ensure that mechanisms of local accountability ensure that the benefits accrue to a wide section of the community (see Pei 1996 quotation above). In all these cases, and particularly in East Asia, the states built on local institutions, using local idioms to build their networks of collaboration. -Small-scale communities already have the dense network of ties of co-operation necessary for this purpose, and the state channeled these strengths to contribute to development goals. When these states decided to accelerate the pace of development, they were creative in tapping into existing local informal institutional structures - ignored by previous governments - and setting their norms of reciprocity, co-operation and trust to work to achieve development goals. The Brazilian example is of especial interest in this context. It shows that it is possible to achieve these ends even in settings without the strong clan and kinship ties characterizing rural East Asian communities.28 Even in less tightly knit communities, people will engage in collective action for common goals if they are given the opportunity and the required information.29 28 The strength of African communities, evidenced by their response to the availability of social funds and other communmty-based funds, could also be an asset to develop similar reforms in this continent (World Bank 2000c). 29 And the strength of African comrnunities, evidenced by their response to the availability of social funds and other conmmunity-based funds, could also be an asset to develop simiilar reforms (World Bank 2000c). 17 Changing local and supra-local agencies' mode of operation To make the most of the potential advantages of community participation in public service delivery, it is important to ensure that the mode of operation of local government agencies is such as to draw on local information and monitoring capacity. This requires careful attention to bureaucratic design and staff performance incentives.30 In many settings, local administrations are designed to be "instruments of the central government in change of controlling local resources, and oriented to law and order."3' Not only do they have little accountability to the people they serve, but sometimes (as illustrated above) the system is specifically designed to minimize interchange between communities and local administration, in order to reduce opportunities for corruption. Local agencies are naturally reluctant to accept demands for greater accountability, especially to clients over whom they are often accustomed to exercising considerable power. Besides, they develop work cultures which are difficult to shake. Officials can become quite adapted to working environments that offer few performance incentives, but that do offer them the benefits of status as the officials who must be petitioned by their clients. They can become accustomed to low work demands and to supplementing inadequate incomes with bribes and payoffs. To improve local agencies' performance, states orchestrated pressure on them from above as well as from the clients below (Figure 1), to motivate them into working collaboratively with communities. The modalities of pressure from above varied widely, depending on the country's political and administrative resources. South Korea's military government usedfiat to forcefully prescribe bureaucratic reform, while leaving considerable latitude for local initiative in service delivery. (Even so, officials tried to evade greater accountability, for example by hiding the complaints box behind a screen).32 Post-reform China used financial incentives, giving local governments a stake in the profits of community enterprises. In Northeast Brazil, the state used official job recognition to motivate staff. Pressure from below was orchestrated in all cases by getting communities to expect and demand better services. At the same time, there is a real risk of encroachment of the center on local administrations, which can potentially impede their functioning as strong entities. Accountability mechanisms are important, but these have to be structured such as to permit and facilitate the development of local administrative and governmental capacities. Also critical to the success of the enterprise was adjusting the roles of higher levels of government. Through creative management, work habits and perceptions of roles were altered amongst higher levels of bureaucracy. They were encouraged to shift away from '0 This is a large question that has received considerable attention. See for example Bardhan (2000), Crook and Manor (1994), and Manor (1999). 3' Whang (1981), describing local govermment in South Korea in the early 1960s. 3Z Wade (1997). 18 monitoring and co-ordinating outreach services, to focus more on the specialized tasks of managing overall strategy, providing technical support, and regulation and facilitation. Enhancing the incentives of grassroots level workers and their accountability to communities A variety of rewards and sanctions were used to elicit good job performance from frontline workers. In South Korea and China, a major incentive for good performance was built in by drawing the lowest echelon of workers from within the communities themselves - creating strong potential sanctions for non-performance as the workers had a long-run interest in good relations with their community. They also gained personally from good service delivery (e.g., successful irrigation or enterprises), along with the rest of their community. Another effective strategy was having their salary and/or job renewal dependent on community opinion. Where workers could be outsiders, as in Brazil, communities were exhorted to monitor their workers. To keep workers motivated, communities were encouraged to show their respect and appreciation of good workers, and official recognition was also given. Giving grassroots workers the flexibility to do their work properly was important, as also giving them the scope to multi-task such that they could meet client needs more effectively. This also provided positive reinforcement for workers. For example, if irrigation workers or drought relief staff can also provide agricultural extension, they are able to derive more status and satisfaction from being able to help their clients. A key aspect of grassroots workers' jobs in all these case studies was that they were expected not merely to deliver a specific service, but to act as brokers between the community and the wider world. Being more in touch with state facilities as well as those of the outside world, they were expected to help their clients expand their access to resources and help them negotiate their way through unfamiliar procedures. This is very important to people deprived of information, and in the absence of adequate state outreach they are forced to depend for such services on potentially exploitative patrons and middlemen.33 33 Note that instead of discouraging collusion, this model actually encourages collusion and channels it mto forms which can be productive for the common good (Baumol 1990). The success of the model depends on productive collusion between conmmunity members and their leaders, and between these leaders and local officials as well as with successive echelons of the state administration. Local accountability can help forestall an appropriation of gains by the leaders and officials alone, as Pei (1996) emphasizes in his discussion of China: The official cannot make a continued return for himself alone from (his administrative position). Should he try to do so, his highly-valued reputation and its interrelated benefits will be lost. So it is in the official's own long-term interests not to jeopardize them. Thus, the benefits of village officials must be realized through profit maximization of the village coalition. (parentheses ours). 19 Generating community demandfor better public goods and services Generating demand for grassroots services was a crucial component of successful community engagement. Through wide information dissemination, people were made aware of what services should be available to them, so that they would pressure the staff of their local agencies to provide them. This is crucial for informing communities of new opportunities and of new rights: ... the proclamation of a right (is) if not automatically self-fulfilling, at least the first step to any serious attempt to secure that right (Hirschman 1 984a: 9). Moreover, aspirations were systematically raised to create additional demand for services. For example, South Korea used organized competition between villages to generate demand for upgrading village facilities, where if left alone people's aspirations would not have risen so quickly. In the other settings too, information was actively disseminated on new opportunities in order to motivate clients into demanding and using services more actively and increase the pace of development. Financial, technical and other support to communities These case studies illustrate a range of ways in which financial, technical and other support can be channeled to communities. Fiscal decentralization can be used creatively to promote accountability and ownership. Financial inputs can be offered to help meet specific needs, such as for repairing schools in Brazil. Financial incentives can be used as a way of channeling communities' interests towards new areas, as was done in the Republic of Korea by setting up competitions to see which village was most successful in its Saemaul Undong activities. Similarly, many forms of technical support were effectively used. In Brazil, technical training was given to enable communities to maintain school buildings. In Taiwan and China, technical support extended to helping identify products, markets, production techniques etc., to help local communities produce both agricultural and industrial goods for a wide market. This was done in the context of decentralized decision- making, allowing communities to organize their own enterprises. Sustained political backing for reforms, and payoffs for the state It is important to note that these cases do not bear out some fears commonly associated with "going directly to communities." Firstly, higher levels of government did not lose power as a result of these changes, and their new roles offered greater scope for using professional expertise than before. Secondly, governments were not weakened by the process. Instead of being undermined, potentially weak governments were strengthened through this process. Communities were brought into direct contact with the central structure of the state. Governments were able to inculcate loyalty to the regime during the elaborate training of community workers in central locations. 20 These initiatives had substantial political payoffs. Politicians in democracies found an effective way of gaining votes, and quickly learned to adapt their electoral strategies towards offering people visible gains and empowerment. For example, in a municipal election in Brazil, candidates emphasized that they would have to support the new arrangements in order to be elected (Coirolo 2000). Authoritarian regimes gained legitimacy for their attention to their citizens' welfare. At the same time, the conditions were strengthened for a plural polity. Whether or not the center's primary motivation was ideological and developmental, well-understood immediate political interests have been a powerful driving force to mobilize communities and improve service deliveries. The lessons from the less successful efforts in the Philippines and Mexico illustrate that it is crucial for politicians to understand that this is a "win-win" strategy for them and their citizenry - and of sticking with the strategy through changes in administration. Caveats. While the above case studies show how much can be achieved through engaging communities, they also show that this process requires careful planning and active management. As Abraham and Platteau (forthcoming, 2002), Tendler (2000) and others have pointed out, there are serious dangers in seeking quick solutions from decentralizing powers to communities. These ideas need to be developed with careful thought to the outcomes sought and the nature of the communities involved. Platteau (2000) summarizes some of the potentially fatal oversights commonly made when people are in a hurry for a quick fix: * Informal community institutions and procedures can need adapting if they are required to be used for new purposes. Additional skills may also be needed, such as in maintaining accounts * Certain situations may require complementary interventions by the state, for example to offset dominance by local elites. By merely-devolving power and finances to communities with highly skewed power structures, the stage is set for "local capture." * Similarly, poor outcomes are also likely if prominent community members or NGOs are used to channel resources, without extensive efforts to disseminate information to and generate demand among the whole community about the expected products. The careful devising of ways to avoid "elite capture" in Brazil and the careful choice of development objectives to avoid areas most vulnerable to such capture34 is in sharp contrast to that pertaining in many development projects.35 What the case studies discussed here illustrate is that central agencies have an important role to play in managing the process of decentralization. As Platteau (2001) summarizes it: Brazil's success in decentralization of public service ... is that "it had at its core a strong and new role played by central government" (Tendler 1997: 73). ...The (state) government 'kept an iron hand' on some crucial components of the decentralized 34 Tendler (1997), World Bank (1997), Coirolo (2000). 35 Tendler (2000), Platteau (2000), Abraham and Platteau (2002). 21 programs to substantially reduce the opportunities for mayors and local power-holders (especially large landowners) to exercise patronage. 5. Discussion We analyze country experiences to examine the conditions required to tap the enormous potential of local energies and resources for community-level development. At the level of the community, this requires broadening the distribution of power to avert local capture and facilitate collective action. At the level of the state, much can be achieved by creating alliances between higher levels of government and communities, to provide incentives for local administration to collaborate with local communities and thereby improve public service delivery. We argue that land reform, tenancy reform and expansion of non-crop opportunities have powerful benefits for development by releasing people's energies for upward mobility- energies that are otherwise suppressed by fear that patrons will oppose them or capture the benefits. With greater intra-community equality, there is less of a trade-off between using comrnunities' advantages in local information, and elite capture. In turn, this helps build a more developmental polity and society. We also argue that higher levels of government can form alliances with communities to catalyze institutional change at the local level, in quite diverse settings. Through these alliances, they can increase the incentives and scope for local public agencies and communities to collaborate, creating powerful coalitions and developmental synergies with a potentially self-sustaining dynamic. Moreover, there are successful examples both in settings with democratic processes (where elected community representatives constituted the lowest echelon of the bureaucracy), as well as in others where community organizations function independently of these processes. In the East Asian settings, these institutional changes encountered little resistance from local elites and government officials, because at the time people were anxious to recover from the devastation of war and prepared to accept radical changes from the authoritarian regimes governing them. Also, for historical and ideological reasons it was relatively easy to push through radical land reforms and generate more egalitarian and developmentally oriented local communities. Given that most developing countries today face very different circumstances from Brazil, the Brazilian case is especially noteworthy. It shows that with some creative political thinking it is possible to effect rapid change even in inegalitarian institutional settings subject to "local elite capture." It illustrates the additional efforts required from the state to gradually overcome these obstacles, by carefully designing changes which bring grassroots electoral pressure to bear on local government and reduce "local capture" by elites. This model has proved to be very successful in achieving a wide range of outcomes, ranging from health and drought relief to generating agrarian and industrial economic 22 growth. As evidenced by China and Taiwan, state-community synergies can catalyze the expansion of secondary sector production in rural and semi-rural areas, thereby raising local incomes and catering not only to local consumption, but also to a rich export market for secondary sector products. These case studies also indicate that states, both authoritarian and democratic, have much to gain in legitimacy and popularity from initiating these institutional changes - if only they have the insight to realize that this is a more effective and sustainable way of gaining popular support than routine political machinations. At the same time, the example of Mexico shows how fragile such efforts can be if political support from above is prematurely withdrawn. As these studies show, building these state-community synergies requires careful channeling of community strengths, demand creation, and management of incentives for everyone concemed: communities, frontline workers, local agencies and supra-local agencies. The process depends on strong and sustained political backing. Moreover, careful choices need to be made regarding areas of action - while many kinds of development initiative benefit from decentralization and community engagement, others do not. For example in the field of public health, managing most aspects of environmental sanitation is best done at grassroots level where the problems are easily identified, but programs such as immunization are often most successful when they are centralized.36 This paper provides some initial insights into these processes, which deserve systematic consideration if the enormous potential of community-level energies and resources are to be effectively unleashed for development. 36 Khaleghian (2002). 23 Appendix 1. Breaking the power of landowning elites through development programs: a case study from Uttar Pradesh, India The socioeconomic hierarchy of the village is apparent even before the bus arrives at the village: there is one large brick whitewashed mansion, standing out among a sea of mud huts. The mansion is the home of the talukdar family, the erstwhile large landowner whose duty it was to collect land revenues for the colonial power. They lost some of their holdings when land ceilings were imposed in the 1950s, although they evaded much of this through fictitious division. For the next couple of decades, the talukdars consolidated their relationships with the new power structures of the state. In a typical pattern of diversifying family networks, the father arranged for one son to be in the police service, while another managed the land. They continued to be the critical source of credit and employment for the villagers, and their overwhelming social superiority was acknowledged by the villagers as they prefaced every interaction with the greeting "Touching your feet, Lord..." For this family, well-educated and well-connected, it was easy to divert development funds to their own benefit and the other villagers would typically not even come to know of the entitlements of which they were being deprived. Even if they did come to know of it, they could hardly protest about it when the talukdars had guns and were known to rape and maim at will. In the mid-1970s, information on tubewells reached some middle-level peasants, who pooled resources to sink a tubewell to irrigate their contiguous plots. Eager to maximize their profits, they began to sow cash crops and raised the wages they offered agricultural labor. The talukdar's son responded by striding around at the weekly market with a gun slung over his shoulder, threatening that he would shoot anyone who offered laborers more than the "going rate." This temporarily thwarted the middle peasants' efforts. However, new opportunities offered by tubewells and the opening of a milk- collection center in the village made it increasingly difficult for the talukdars to retain their position in the village society and economy. Over time, the middle-level peasants were successful in raising their incomes, and offered a greater multiplicity of sources of employment and credit for the poor. The bi-polar polity of the village changed into one with a broader distribution of power. In a study of another Uttar Pradesh village, Dreze, Lanjouw, and Sharma (1998) noted similar tensions between the same middle peasant group and the ex-talukdars. There, the talukdars had tried methods such as arson and election-rigging to maintain their power, but the middle peasants' determined use of the new agrarian opportunities was eventually successful in challenging the hegemony of the talukdars. By the 1990s, the middle peasants had become prosperous and educated, and formed a serious challenge to the talukdars in the village polity. Source: Vinod Jairath and Monica Das Gupta, fieldnotes from Hardoi district. 24 Bibliography The word "processed" describes informally reproduced works that may not be commonly available through library systems. Abraham, Anita, and Jean-Philippe Platteau. "Participatory Development: where culture creeps in." In Vijayendra Rao and Michael Walton, eds., Culture and Public Actzon. Stanford CA: Stanford University Press. Forthcoming. Abraham, Anita, and Jean-Philippe Platteau. 2002. The Risk of Resource Misappropriaton in Community- Based Development Projects, Namur. Belgium: Centre for Research on the Economics of Development. Amsden, Alice. 1985. "The State and Taiwan's Economic Development." In Peter Evans, Dietrich Rueschemeyer and Theda Skocpol, eds., Bringing the State Back In. New York: Cambridge University Press. Baland, Jean-Marie, and Jean-Philippe Platteau. 1999. "The Ambiguous Impact of Inequality on Local Resource Management." World Development 27(5): 773-88. Banerjee, Abhijit V., Paul Gertler, and Maitreesh Ghatak. 2001. "Empowerment and Efficiency: tenancy reform in West Bengal." Journal of Political Economy 110(2):239-280. Banerjee, Abbijit V., Dilip Mookherjee, Kaivan Munshi and Debraj Ray. 2001. "Inequality, Control Rights and Rent Seeking: Sugar Cooperatives in Maharashtra." Journal of Political Economy 109(1, February): 138-90. Banerjee, Abhijit, and Lakshmi Iyer. 2002. "History, Institutions and Economic Performance: the legacy of colonial land tenure systems in India." MIT working paper 02-27. Cambridge, Mass. Bardhan, Pranab. 1995. "Rational fools and Cooperation in a Poor Hydrauhc Economy." In K. Basu, P. Pattanaik, and K. Suzumura, eds., Choice, welfare, and development: afestschrift in honour of Amartya K Sen. New York: Clarendon Press. Bardhan, Pranab. 1997. The role of governance in economic development : a political economy approach. Paris : Development Centre of the Organisation for Economic Co-operation and Development; Washington, D.C.: OECD Washington Center. .1996. "Decentralised development." Indian Economic Review XXXI(2): 139-56. . 1999. "Water Community: An Empirical Analysis of Cooperation on Irrigation m South India." http://globetrotter.berkeley.edu/faculty/Bardhan.html. . 2000. "Decentralization and Public Goods." In Dethier, Jean-Jaques, eds,. Governance, Decentralization and Reform in China, India and Russia. Dordrecht: Kluwer Academic Publishers. Forthcoming. Bardhan, Pranab, and Ashok Rudra. 1986. "Labour mobility and the boundaries of the village moral economy." Journal of Peasant Studies 13 (April): 90-115. Bardhan, Pranab, and Dilip Mookherjee. 1998. "Expenditure Decentralization and the Delivery of Public Services in Developmg Countries." CIDER Paper C98-104, University of California at Berkeley, Berkeley, CA. Bardhan, Pranab, and Dilip Mookherjee. 1999. "Capture and governance at local and national levels." Processed. Barrientos, Jorge C. 1999. "Coordinating Poverty Alleviation Programs with Regional and Local Governments. The Experience of the Chilean Social fund (FOSIS)." Social Protection Team, Human Development Network. World Bank, Washington DC. Baumol, William J. 1990. "Entrepreneurship: productive, unproductive, and destructive." Journal of Political Economy 98: 893-921. 25 Besley, Tim. 1995. "Property rights and investment incentives: theory and evidence from Ghana." Journal of Political Economy 103:903-37. Besley, Tim, and Robin Burgess. 1998. "Land reform, poverty reduction and growth: evidence from India." The London School of Economics, STICERD Development Discussion Paper Series No.13., London. Binswanger, Hans P., Shahidur R Khandker, and Mark R. Rosenzweig. 1993. "How infrastructure and financial institutions affect agricultural output and investment in India." Journal of Development Economies 41(2): 337-66. Binswanger, Hans, and Mark Rosenzweig. 1984. Contractual Arrangements, Employment and Wages in Rural Labour Markets in Asia. New Haven, Conn.: Yale University Press. Boyce, James K. 1987. Agrarian Impasse in Bengal: agricultural growth in Bangladesh and West Bengal 1949-1980. New York: Oxford University Press. Braverman, Avishay, and Joseph Stiglitz. 1982. "Sharecropping and the interlinking of agrarian markets." American Economic Review 72(4). Breman, Jan. 1974. Patronage and Exploitation: changing agrarian relations in South Gujarat, India. Berkeley: University of California Press. Candler, Wilfred, and Nalini Kumar. 1998. India: the dairy revolution. Washington, D.C.: World Bank, Operations Evaluation Department. Carino, Ledivina V. 1998. "Initiatives Taken against corruption: the Philippine case." University of the Philippines. Processed. Cermea, Michael M. 1983. << A Social Methodology for Community Participation in Local Investments, The Experience of Mexico's PIDER Program." World Bank Staff Working Papers, No 598. Washington, D.C. Coirolo, Luis. 2000. Personal Communication. Crook, Richard, and James Manor. 1994. Enhancing participation and institutional performance: democratic decentralization in South Asia and West Africa., London: Overseas Development Administration. Das Gupta, Monica. 1987. "Informal Security Mechanisms and Population Retention in Rural India", Economic Development and Cultural Change 13 (1, March): 101-20. Das Gupta, Monica, Helene Grandvoinnet, and Mattia Romani. 2000. "State-Community Synergies in Development: laying the basis for collective action." Background paper for the WDR 2000/1 "Attacking Poverty." Deere, Carmen Diana, and Magdalena Le6n. 1997. "Women and land rights in the Latin American neo- liberal counter-reforms." Women in International Development Working Paper #264. Michigan State University, East Lansing, MI. Dreze, Jean. Comments to chapter 8 of Dethier, Jean-Jaques, eds,. Governance, Decentralization and Reform in China, India and Russia. Kluwer Academic Publishers, Dordrecht. Forthcoming. Dreze, Jean, and Amartya Sen, eds. 1996. Indian Development: selected regionalperspectives. Delhi: Oxford University Press. Dreze, Jean, Peter Lanjouw, and Naresh Sharma. 1998. "Economic Development in Palanpur, 1957-93." In P. Lanjouw and N. Stern, eds., Economic Development in Palanpur over Five Decades. Oxford: Clarendon Press. Du, H. 1990. "Causes of Rapid Rural Industrial Development." In W. Byrd, and Q. Lin. eds., China's Rural Industry: Structure, Development, and Reform. New York: Oxford University Press for the World Bank. Eder, James F. 1993. "Family Farming and Household Enterprise in Philippine community, 1971-1988: Persistance or Proletarization?" Journal ofAsian Studies 52(3): 647-7 1. Evans, Peter. 1995. Embedded Autonomy: states and industrial transformation Princeton NJ: Princeton University Press. Evans, Peter. 1996. Government action, social capital and development: reviewing the evidence on synergy, World Development 24(6):1119-1132. 26 Fei, John C.H., Gustav Ranis and Shirley W.Y.Kuo. 1979. Growth with Equity: the Taiwan case, Washington DC: World Bank (published by New York: Oxford Umversity Press). 1981. The Taiwan success story: rapid growth with improved distribution in the Republic of China. Boulder, Colorado: Westview Press. Foner, Enc. 1989. Reconstruction A America's unfinished revolution, 1863-1877. New York: Harper & Row. Fox, Jonathan. 1992. Democratic Rural Development: Leadership Accountability in Regional Peasants Organizations, Development and Change 23(2). Galasso, Enmnanuela and Martin Ravallion. 2000. "Distributional Outcomes of a Decentralized Welfare Program", Washington DC: World Bank. Ghatak, Maitreesh and Priyanka Pandey. 2000. "Contract Choice in Agriculture with Joint Moral Hazard in Effort and Risk." Journal of Development Economics 63(2, December):303-26. Hall, Anthony. 1994. "Grassroots action for resettlement planning: Brazil and beyond." World Development 22:1793-1809. Harnss, John. 1999. "How Much Difference Does Politics Make? Regime Differences Across Indian States And Rural Poverty Reduction." Development Studies Institute, LSE. Hayami, Yujiro, Agnes R. Quisumbmg, and Adriano S. Lourdes. 1990. Toward an alternative land reform paradigm : a Philippine perspective. Manila: Ateneo de Manila University Press. Hemng, Ronald J. 1999. "Political conditions for land reform and poverty alleviation." Paper prepared for the DFID conference on 2001 World Development Report on Poverty, Birmingham England August 16-17 1999. Hirshman, Albert. 1984a. "Grassroots development in Latin America." Challenge Sept./Oct.: 4-9. Hirshman, Albert. 1984b. "Getting Ahead Collectively: Grassroots Experience in Latin America." World Development (special supplement). Hoddhnott, John, M. Adato, Tim Besley, and Lawrence Haddad. 2001. "Participation and Poverty Reduction: Issues, Theory, and New Evidence from South Africa." FCND Discussion Paper N° 98. International Food Policy Research Institute, Washington, D.C. Azfar, Omnar, Tugrul Gurgur, Satu Kahkonen, Anthony Lanyi, and Patrick Meagher. 2000. "Decentralization and Governance: An Empirical Investigation of Public Service Delivery in the Philippines." A joint paper by Center for Institutional Reform and the Informal Sector, University of Maryland and the World Bank. Available at http://wwwl .worldbank.org/publicsector/decentralization/DraftPhilipRinespaperl 21800.doc. Johnson, G.E. 1991. "Family strategies and economic transformation in rural China: some evidence from the Pearl River Delta." In D. Davis and S. Harrell, eds., Chinese Families in the Post-Mao Era. Berkeley: University of California Press. Kay, Crist6bal, and Patricio Silva. 1992. "Rural Development, Social Change and the Democratic Transition." In Crist6bal Kay and Patricio Silva, eds., Development and social Change in the Chilean Countryside: from the pre-land reform period to the democratic transition. Amsterdam: Centre for Latin American Research and Documentation. Khaleghian, Peyvand. 2002. "Decentralization and public services: the case of inmmunization." Part II, Dr.P.H. Thesis. Department of International Health, The Johns Hopkins University. Klitgaard, R. 1997. "Unanticipated Consequences in Anti-Poverty Programs." World Development 25(12): 1963-72. Kohli, Atul. 1987. The State and Poverty in India: the politics of reform. Cambridge: Cambridge University Press. Laffont, Jean-Jacques, and Mohamed Salah Matoussi. 1995. "Moral hazard, financial constraints and sharecropping in El Oulja." Review of Economic Studies 62(3): 381-99. Lanjouw, Peter, and Jean 0. Lanjouw. 1995. "Rural non-farm employment: a survey." Policy Research Working Paper 1463. Development Research Group, World Bank, Washington, D.C. 27 Lanjouw, Peter, and Nicholas Stem, eds., 1998. Economic Development in Palanpur over Five Decades. Oxford: Clarendon Press. Lin, Justin Yifu. 1992. "Rural reforms and agricultural growth in China." American Economic Review 82(1): 34-52. Manor, James. 1999. The Political Economy of Democratic Decentralization. Washington, D.C.: World Bank. Mitra, Suprata. 1997. "Making local politics work: rural elites political conflicts and legitimacy in India." South Asia Institute. University of Heidelberg. Processed. Newman, Constance, and Sudharshan Canagarajah. 2000. "Non-Farm Employment, Poverty, and Gender Linkages: Evidence from Ghana and Uganda." World Bank, Washington, D.C. Processed. North, Douglass C. 1989. "Institutions and Economic Growth: An Historical Introduction." World Development 17(9): 1319-32. North, Douglass. 1991. "Institutions." Journal of Economic Perspectives 5(Winter): 7-112. Oi, Jean C. 1999. Rural China Takes Off: institutionalfoundations of economic reform., Berkeley: University of California Press. Olson, Mancur. 1971. The logic of collective action; public goods and the theory of groups. Cambridge, Mass.: Harvard University Press Ostrom, Elinor, Larry Schroeder, and Susan Wynne. 1993. Institutional Incentives and Sustainable Development: infrastructure policies in perspective. Boulder Colorado: Westview Press. Ostrom, Elinor, and James Walker. 1997. "Neither markets nor states: Linking transformation processes in collective action arenas." In Dennis Mueller, ed., Perspectives on Public Choice. New York: Cambridge University Press. Owen, Daniel, and Julie van Domelen. 1998. Getting an earful: a review of beneficiary assessments of social funds. Social Protection Team, Human Development Network. Washington, D.C.: World Bank Pei, Xiaolin. 1996. "Township-village enterprises, local governments and rural communities: the Chinese village as a firmn during economic transition." Economics of Transition 4(1): 43-66. Platteau, Jean-Philippe. 2000. "Community Imperfections." Paper presented at the Annual Bank Conference on Development Economics, Paris June 2000. Probe Team. 1999. Public Report on Basic Education in India. Bombay: Oxford University Press. Prosterman, Roy, Mary Temple, and Timothy Hanstad. 1990. Agrarian Reform and Grassroots Development: Ten Case Studies. Lynne Rienner Publishers. Boulder and London. Putnam, Robert. 1993. Making democracy work: civic traditions in modern Italy. Princeton, N.J.: Princeton University Press. Qian, Yingyi, and Barry R. Weingast. 1996. "China's transition to markets: market-preserving preserving federalism, Chinese style." Journal of Policy Reform 2: 149-85. Ramachandran, Vimala K. 1996. "On Kerala's Development Achievements." In Jean Dreze and Amartya Sen, eds., Indian Development: selected regional perspectives. Delhi: Oxford University Press. Ranis, Gustav, and Frances Stewart. 1993. "Rural nonagricultural activities in development: theory and application." Journal ofDevelopment Economics 40: 75-101. Salamon, Lester M. 1979. "The time dimension in policy evaluation: the case of the New Deal land-reform experiments." Public Policy XXVI(2): 129-83. Sampark. 1997. Impact study of milk co-operatives at village level (Operation Flood III). Operations Evaluation Department. Washington DC: World Bank. Sidel, 1996. "Siam and its twin? Democratization and Bossism in contemporary Thailand and the Philippines." IDS 27(2, April). Sidel, 1997. "Philippine Politics in Town, District and Province: Bossism in Cavite and Cebu." The Journal of Asian Studies 56(4, November). 28 Srmivasan, T.N. 1979. "Bonded Labor Contracts and Incentives to Adopt Yield-Raismg Innovations in Semi- feudal Agriculture." Economic Journal 89. Stavis, B. 1982-83. "Rural Local Governance and Agricultural Development in Taiwan." In N. T. Uphoff, ed., Rural Development and Local Organization in Asia. New Delhi : Macmillan, Vol. II, pp. 166- 271. Tannenbaum, Frank. 1962. Ten Keys to Latin America. New York: Alfred A. Knopf Tendler, Judith. 1997. Good Government in the Tropics. Baltimore: Johns Hopkins Press. Tendler, Judith. 2000. "Why are Social Funds so Popular?" In Shahid Yusuf, W. Wu, and S. Everett, eds., Local Dynamics in the Era of Globalization. Oxford: Oxford University Press. Thiesenhusen, William C. 1995a. "Land Reform Lives! Review essay." The European Journal of Development Research 7: 193-209. . 1995b. Broken Promises: agrarian reform and the Latin American campesino. Boulder, Colorado: Westview Press. Wade, Robert. 1982 . Irrigation and agricultural politics In South Korea. Boulder, Colorado: Westview Press. . 1985. "The market for public office: why the Indian state is not better at development." World Development 13: 467-97. .1997. "How mfrastructure agencies motivate staff: canal irrigation in India and the Republic of Korea." In Ashoka Mody, ed., Infrastructure Strategies in East Asia: the untold story. Washington DC: World Bank. Whang, In-Joun. 1981. Management of Rural Change in Korea. Seoul: Seoul National University Press. World Bank. 1997. Latin America and the Caribbean Region, Staff Appraisal Report: Brazil Northeast Rural Poverty Alleviation Program, Rural Poverty Alleviation Project - Paraiba. World Bank, Washington, D.C. World Bank. 1998. Assessing Aid-What Works, What Doesn 't, and Why. Washington DC: Oxford University Press. World Bank. 2000a. Africa Region Community Action Program Team 2000. The vision that conquered blindness. World Bank, Washington, D.C. Draft. World Bank. 2000b. Sourcebook on Community-Driven Development in the Africa Region: Community Action Programs.- Africa Region Comrnunity Action Program Team 2000: Washington DC: World Bank. World Bank. 2000c. Can Africa Claim the 21" Century? Washington, D.C.: World Bank. 29 Policy Research Working Paper Series Contact Title Author Date for paper WPS2952 The Effects of a Fee-Waiver Program Nazmul Chaudhury January 2003 N Chaudhury on Health Care Utilization among the Jeffrey Hammer 84230 Poor. Evidence from Armenia Edmundo Murrugarra WPS2953 Health Facility Surveys An Magnus Lindelow January 2003 H Sladovich Introduction Adam Wagstaff 37698 WPS2954 Never Too Late to Get Together Bartlomiej Kaminski January 2003 P Flewitt Again Turning the Czech and Slovak Beata Smarzynska 32724 Customs Union into a Stepping Stone to EU Integration WPS2955 The Perversity of Preferences. The C,aglar Ozden January 2003 P Flewitt Generalized System of Preferences Eric Reinhardt 32724 and Developing Country Trade Policies, 1976-2000 WPS2956 Survey Compliance and the Johan A Mistiaen January 2003 P Sader Distribution of Income Martin Ravallion 33902 WPS2957 Mexico In-Firm Training for the Hong Tan January 2003 H Tan Knowledge Economy Gladys Lopez-Acevedo 33206 WPS2958 Globalization and Workers in Martin Rama January 2003 H Sladovich Developing Countries 37698 WPS2959 Wage Differentials and State- Michael M Lokshin January 2003 P Sader Private Sector Employment Choice Branko Jovanovic 33902 in the Federal Republic of Yugoslavia WPS2960 The Poverty/Environment Nexus in Susmita Dasgupta January 2003 Y D'Souza Cambodia and Lao People's Uwe Deichmann 31449 Democratic Republic Craig Meisner David Wheeler WPS2961 Strategic Planning for Poverty Rob Swinkels January 2003 N Lopez Reduction in Vietnam Progress and Carrie Turk 88032 Challenges for Meeting the Localized Millennium Development Goals WPS2962 High Consumption Volatility Philippe Auffret January 2003 K Tomlinson The Impact of Natural Disasters? 39763 WPS2963 Catastrophe Insurance Market in the Philippe Auffret January 2003 K Tomlinson Caribbean Region Market Failures 39763 and Recommendations for Public Sector Interventions WPS2964 Wages and Productivity in Mexican Gladys Lopez-Acevedo January 2003 M Geller Manufacturing 85155 WPS2965 Informality Revisited William F Maloney January 2003 P Soto 37892 WPS2966 Health and Poverty in Guatemala Michele Gragnolati January 2003 M Gragnolati Alessandra Marini 85287 WPS2967 Malnutrition and Poverty in Alessandra Marini January 2003 M Gragnolati Guatemala Michele Gragnolati 85287 Policy Research Working Paper Series Contact Title Author Date for paper WPS2968 Refining Policy with the Poor Local Edwin Shanks January 2003 N Lopez Consultations on the Draft Carrie Turk 88032 Comprehensive Poverty Reduction and Growth Strategy in Vietnam