Community-Based and -Driven Development: A Critical Review



                                    Ghazala Mansuri and Vijayendra Rao



Community-based (and driven) development (CBD) projects have become an important form of
development assistance, with the World Bank's portfolio alone approximating 7 billion dollars. This paper
reviews the conceptual foundations of CBD/CDD initiatives. Given the importance of the topic, there are,
unfortunately, a dearth of well-designed evaluations of such projects.                 There is, however, enough
quantitative and qualitative evidence, from studies that have either been published in peer-reviewed
publications or have been conducted by independent researchers, to gain some instructive lessons. We
find that projects that rely on community participation have not been particularly effective at targeting the
poor. There is some evidence that CBD/CDD projects create effective community infrastructure, but not a
single study establishes a causal relationship between any outcome and participatory elements of a CBD
project. Most CBD projects are dominated by elites and, in general, the targeting of poor communities as
well as project quality tend to be markedly worse in more unequal communities. However, a number of
studies find a U shaped relationship between inequality and project outcomes. We also find that a
distinction between potentially "benevolent" forms of elite domination and more pernicious types of
"capture" is likely to be important for understanding project dynamics and outcomes. Several qualitative
studies indicate that the sustainability of CBD initiatives depends crucially on an enabling institutional
environment, which requires upward commitment. Equally, the literature indicates that community leaders
need to be downwardly accountable to avoid a variant of `supply driven demand driven development'.
Qualitative evidence also suggests that external agents strongly influence project success. However,
facilitators are often poorly trained and inexperienced, particularly when programs are rapidly scaled up.
Overall, a naïve application of complex contextual concepts like "participation", "social capital" and
"empowerment"-is endemic among project implementers and contributes to poor design and
implementation. In sum, the evidence suggests that CBD/CDD is best done in a context-specific manner,
with a long time-horizon, and with careful and well designed monitoring and evaluation systems.




World Bank Policy Research Working Paper 3209, February 2004

The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of
ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations
are less than fully 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 countries they represent. Policy Research
Working Papers are available online at http://econ.worldbank.org.



Ghazala Mansuri is an economist in the Development Economics Research Group at the World Bank; her
email address is gmansuri@worldbank.org. Vijayendra Rao is a senior economist in the Development
Research Group at the World Bank; his email address is vrao@worldbank.org. The authors are grateful to
Evan Dennis and Lupin Rahman for excellent research assistance and Hans Binswanger, Rob Chase,
Monica Das Gupta, Gershon Feder, Emanuela Galasso, Keith McLean, David Mosse, Dan Owen, Berk
Ozler, Jean-Philippe Platteau, Menno Pradhan, Martin Ravallion, Jesse Ribot, Tara Vishwanath, David
Warren, Howard White, Michael Woolcock, three anonymous referees, and several colleagues at seminars
and informal discussions at the International Food Policy Research Institute and the World Bank for
valuable comments. The article was co-sponsored by the Development Economics Research Group and the
Social Development Anchor of the World Bank.

                                                               1

        Community-based development and its more recent variant, community-driven

development, are among the fastest growing mechanisms for channeling development

assistance. Community-based development is an umbrella term for projects that actively

include beneficiaries in their design and management, while community-driven

development refers to community-based development projects in which communities

have direct control over key project decisions, including management of investment

funds.


        By conservative calculations, the World Bank's lending for such projects has

risen from $325 million in 1996 to $2 billion in 2003--or from $3 billion in 1996 to $7

billion in 2003 when lending for an enabling environment for such projects is included.1

The World Bank's Poverty Reduction Strategy Paper Sourcebook (Dongier and others

2003) views community-driven development as a mechanism for enhancing

sustainability, improving efficiency and effectiveness, allowing poverty reduction efforts

to be taken to scale, making development more inclusive, empowering poor people,

building social capital, strengthening governance, and complementing market and public

sector activities. Community-driven development is said to achieve all this by reducing

information problems (by eliciting development priorities directly from target

communities and allowing communities to identify projects and eligible recipients of

private benefits), expanding the resources available to the poor (through credit, social




1There is considerable potential for measurement error in these numbers, and there is no

authoritative estimate of the World Bank's portfolio of such projects.


                                                    2

funds, capacity building, and occupational training), and strengthening the civic

capacities of communities by nurturing organizations that represent them.


        The potential gains from community-driven development are large. It has the

explicit objective of reversing power relations in a manner that creates agency and voice

for poor people, allowing them to have more control over development assistance. This is

expected to make the allocation of development funds more responsive to their needs,

improve the targeting of poverty programs, make government more responsive, improve

the delivery of public goods and services, and strengthen the capabilities of the citizenry

to undertake self-initiated development activities.


        This vision has become one of the cornerstones of the World Bank's

Comprehensive Development Framework, with its increasing emphasis on empowerment

(Dongier and others 2003; Mclean and others 2001; Narayan 2002). This vision is not

universally shared, however. Skeptics have misgivings about the basic precepts of the

approach and more practical concerns with the challenges of implementing such projects.

Summers (2001), for example, is concerned that local institutions promoted under the

aegis of such projects could undermine democratically elected governments. Harriss

(2001), , Mosse (2001), Cooke and Kothari (2001), and others have focused on what

happens when complex and contextual concepts like community, empowerment and

capacity for collective action are applied to the needs of large development projects on

tight timelines. Project implementers, whose incentives are often poorly aligned with the

needs of the project, may choose to gloss over differences within target groups that

underscore local power structures and to short-change the more difficult task of



                                                  3

institution building in favor of more easily deliverable and measurable outcomes.


        Many critics note that evidence on community-driven development initiatives lags

well behind the rate at which projects are being implemented and scaled up. However, the

diversity of views and the intensity of their expression make a review of the available

evidence both necessary and timely. Because of the considerable overlap between

community-based and community-driven development projects, evaluation evidence is

reviewed for any project with community participation as a crucial element of its design.

Using this broader definition, there is enough credible research to glean some useful

insights about specific facets of these programs. That, is the purpose of this review.


        The review focuses on several questions for which reliable information could be

found: Does community participation improve the targeting of private benefits, like

welfare or relief? Are the public goods created by community participation projects better

targeted to the poor? Are they of higher quality, or better managed, than similar public

goods provided by the government? Does participation lead to the empowerment of

marginalized groups--does it lessen exclusion, increase the capacity for collective action,

or reduce the possibility that locally powerful elites will capture project benefits? Do the

characteristics of external agents--donors, governments, nongovernmental organizations

(NGOs), and project facilitators--affect the quality of participation or project success or

failure? And finally, can community participation projects be sustainably scaled up?


        The literature reviewed includes two main types of studies: impact evaluations,

which use statistical or econometric techniques to assess the causal impact of specific

project outcomes, and ethnographic or case studies, which use anthropological methods

such as participant observation, in-depth interviews, and focus group discussions. While


                                                  4

case studies cannot attribute impact, they often provide a more nuanced picture of

projects in particular contexts and yield insights that can be difficult to generate with

quantitative techniques.

        We focus on studies that have either undergone the test of peer-reviewed

publication, or have been conducted by independent researchers. This allows us to use an

exogenous rule that improves the quality and reduces the level of potential bias while

casting a wide-enough net to let in research from a variety of disciplinary perspectives on

different types of CBD projects.

        The article looks first at the history of participatory development and the move

toward community-based development as a key mechanism for channeling development

assistance. The second section examines the literature on participatory development and

collective action to clarify what is meant by participation and social capital, what

constitutes a community, and what are the likely constraints on community participation.

The third section reviews the evidence on the effectiveness of community participation

and attempts to answer some of the questions raised above. The fourth section focuses on

the feasibility of sustainably scaling up projects, and the final section highlights the main

findings and identifies the gaps in knowledge.



Participatory Development and Development Assistance


        There is a long history of community-based forms of development. Clearly

significant were the cooperative movement and Gandhian (Gandhi 1962) notions of

village self-reliance and small-scale development, which Gandhi saw as an antidote to the

corrosive effects of modernization and colonial rule. Another influential perspective was

Paulo Freire's (1970), whose Pedagogy of the Oppressed argues that the "oppressed"



                                                   5

needed to unite to find a way to improve their own destinies. These ideas led to a first-

wave of participatory development in the 1950s, that by 1960 had spread to more than 60

countries in Africa, Asia, and Latin America, largely through the efforts of the U.S.

Agency for International Development (USAID; White 1999). Funding for these

programs dried up in the early 1960s, and within a decade most were shut down. White

(1999, p. 111) notes that one important lesson from the first-wave was "fadism" among

development agencies, which "build up an approach--to the extent that it is virtually a

requirement that a country have such a program in order to receive aid--and then lose

interest, leaving the program to collapse."


        Economists long remained skeptical of the approach. The early literature on

development policy was strongly influenced by the work of Olson (1973) and Russell

Hardin (1982) on collective action to achieve a common goal or pursue a common

interest.2 Olson argued that without coercion or some other special device to make

individuals act in their common interest, "rational self-interested individuals will not act

to achieve their common or group interests". Olson was concerned with "exploitation of

the great by the small," because those with smaller interests in a public good would tend

to free-ride on the efforts of those with greater interests. Garrett Hardin's (1992) "tragedy

of the commons" also acquired metaphorical power--particularly as environmental

concerns became more important--but has much broader implications that impinge on a

range of economic problems, including the domain of the public and the private,



2A key issue in the literature is that the goods or services produced through collective action

must be nonexcludable, jointly produced, and costly.


                                                       6

decentralization of power to local governments, and the provision and management of a

host of goods and services that are to some degree public or are common-pool resources.

Like Hardin, property rights theorists such as Demsetz (1970) and North (1990) argued

that common property resources would be overexploited as demand rose unless the

commons were enclosed or protected by strong state regulation. This view generated a

great deal of pessimism in multilateral development institutions about the viability of

local collective action in the provision of public goods and created a strong impetus for

state provision of public goods, state regulation of common-pool resources, and an

emphasis on development of private property rights.


        By the mid-1980s, critics of "big development" were complaining that many

large-scale, government-initiated development programs, from schooling to health to

credit to irrigation systems, were performing poorly, while rapidly degrading common-

pool resources were creating significant negative environmental and poverty impacts.

These complaints reawakened interest in the local management of resources and

decisions. The participatory development movement led by Chambers (1983) and others

was important in applying these ideas to small-scale development in ways that would

allow the poor to be informed participants in development, with external agents acting

mainly as facilitators and sources of funds. Further support came from the increasingly

strong critique of development from academic social scientists such as Escobar (1995)

and Scott (1998), who argued that top-down perspectives were both disempowering and

ineffective. Meanwhile, projects like the Self-Employed Women's Association in India,

the Orangi slum improvement project in Pakistan, and the Iringa Nutrition project in

Tanzania were generating interest as highly successful instances of community-driven

                                                  7

development (Krishna, Uphoff, and Esman 1997) that could provide important lessons

for large donors.


        Thinking in mainstream development circles was also significantly affected by the

work of Hirschman (1970, 1984), Cernea (1985), and Ostrom (1990). Hirschman's

(1970) notions of "voice" and "exit" provided a way for development practitioners to

understand how collective agency could improve well-being, views that were bolstered

by Hirschman's (1984) own attempts to apply these ideas to participatory development.

Cernea (1985) showed how large organizations like the World Bank could "put people

first" by working systematically at the local level.


        Ostrom's (1990) work on the management of common-pool resources shifted

perceptions on the potential for collective action in poor communities. Ostrom argued

that what made Olson's and Hardin's work most powerful also made it dangerous. She

considered the use of these models as a foundation for policy troubling since their results

depended on a set of constraints imposed for purposes of analysis. Their relevance in

empirical settings, she argued, was an open question rather than a foregone conclusion. In

the real world, after all, the capabilities of those involved can be changed, thus altering

the constraints themselves. Ostrom and others assembled considerable evidence from

case studies to show that endogenous institutions often managed common-pool

resources--often very successfully. Thus Hardin's "open access" was not the universal

model for managing common-pool resources and "remorseless tragedies" were not an

inevitable outcome.


        Sen's (1985, 1999) influential effort to shift the focus of development from


                                                    8

material well-being to a broad-based "capability" approach also deeply influenced the

development community. Central to this approach were strategies to "empower" poor

people, an agenda taken on by the World Bank and other donors as part of their response

to critiques of top-down development. Arguments for "participatory development," as

advocated by Chambers (1983) and others, led to the inclusion of participation as a

crucial means of allowing the poor to have control over decisions.


        The inclusion of participatory elements in large-scale development assistance

came quickly at the World Bank, in social investment funds (Narayan and Ebbe 1997)

and other forms of assistance. Initially focusing on targeting, these projects have moved

toward a more holistic attempt at inducing participation through institutions that organize

the poor and build their capabilities to act collectively in their own interest (Narayan

2002). The World Bank's (2001) World Development Report 2000/2001: Attacking

Poverty focused on empowerment as a key priority of development policy. This has led

to a broad based effort at the World Bank to scale up community-based development and

to make it an important element of programs that seek to improve the delivery of public

services (World Bank 2003).




                                                  9

        With this second wave of interest in community-based approaches to

development,3 a movement that originated in revolutionary goals that were anticolonial

and antimodernization has been absorbed into the mainstream of development. This

transformation has required a new vocabulary to describe its objectives and processes.

The next section reviews the key terms and ideas briefly.



Participation, Community, and Social Capital


        Community-based development relies on communities to use their social capital

to organize themselves and participate in development processes. Thus, concepts such as

participation, community, and social capital are critical to how community participation

is conceptualized and implemented. Handbooks, guidelines, and terms of reference all

use the concepts uncritically, assuming that they are widely and uniformly understood.

What each of these concepts implies, however, is quite controversial.


Participation


        The cornerstone of community-based development initiatives is the active

involvement of members of a defined community in at least some aspects of project

design and implementation. While participation can occur at many levels, a key objective




3White (1999) identifies a second wave in the 1970s and 1980s, initiated by the UN system, but it

seems more a ripple than a wave because it had little influence on large lending agencies. He calls

the current interest in community-driven development "a third-wave which has engulfed the

World Bank."


                                                    10

is the incorporation of local knowledge into the project's decision making processes.

When potential beneficiaries also make key project decisions, participation becomes self-

initiated action--what has come to be known as the exercise of voice and choice or

empowerment. Participation is expected to lead to better designed projects, better

targeted benefits, more cost-effective and timely delivery of project inputs, and more

equitably distributed project benefits with less corruption and other rent-seeking activity.


        This idealized transformatory capacity of participation has been challenged on

several grounds. First, the exercise of voice and choice can be costly under certain

conditions. At the most basic level, it may involve real or imputed financial losses due to

the time commitments required for adequate participation. In addition, participation may

lead to psychological or physical duress for the most socially and economically

disadvantaged, since genuine participation may require taking positions that are contrary

to the interests of powerful groups. While the premise of participatory approaches is that

the potential benefits outweigh such costs, this is by no means certain.


        Second, mainstreaming participation has made it an instrument for promoting

pragmatic policy interests, such as cost-effective delivery or low-cost maintenance, rather

than a vehicle for radical social transformation. This may simply shift some of the costs

of service delivery to potential beneficiaries. Indeed, in both Asia (Bowen 1986) and

Africa (Ribot 1995), participation has been described as a form of forced or corvée labor,

with the poor pressured into making far more substantial contributions than the rich.


        Third, the belief that exposure to participatory experiences will transform the

attitudes and implementation styles of authoritarian bureaucracies (governments or


                                                  11

donors) may be naive. The routinization of participatory planning exercises into the work

of public sector implementation agencies puts new pressures on resources while leaving

implementers unclear about the potential gain to themselves from this new accountability.


        In an examination of several participatory projects, Mosse (2001) finds that even

in projects with high levels of participation, what was labeled "local knowledge" was

often a construct of the planning context and concealed the underlying politics of

knowledge production and use. He identifies four aspects of this. First, participatory

exercises are often public events and open-ended regarding target groups and program

activities. This makes such events inherently political, and what is reflected is often

strongly shaped by local relations of power, authority, and gender. Second, outside

agendas get expressed as local knowledge. Project facilitators shape and direct

participatory exercises, and villager "needs" are often shaped by perceptions of what the

project can deliver. Third, there is local collusion in the planning exercise. People concur

in the process of problem definition and planning because it creates space within which

they can manipulate the program to serve their own interests. While this can benefit both

project staff and project beneficiaries, it clearly places consensus and action above

detailed planning. Fourth, the concept of participation is used to legitimize the project's

previously established priorities and the needs of donors to include such processes in

their projects. Since such projects have little real support from the community or the

project staff, operational demands eventually take over and participatory objectives and

goals are sidelined.


        The upshot seems to be that a project may deliver many things that both recipients



                                                   12

and project implementers consider beneficial, but what can actually be ascribed to

participation is unclear. A convincing evaluation would need to validate (or invalidate)

the participatory model itself and the relationship between project process and impact.


Community


        Participatory projects are typically implemented in a unit referred to as a

community. Most of the literature on development policy uses the term community

without much qualification to denote a culturally and politically homogeneous social

system or one that, at least implicitly, is internally cohesive and more or less harmonious,

such as an administratively defined locale (tribal area or neighborhood) or a common

interest group (community of weavers or potters). This notion of community is

problematic at two levels. First, defining the geographic or conceptual boundaries of a

community is not always straightforward. Administrative boundaries can be meaningless

where settlement patterns are distinct from such boundaries or where increasing mobility

or temporary migrations have transformed community boundaries. In many cases,

factional, ethnic, or religious identities may further complicate the picture. Second, an

unqualified use of the term often obscures local structures of economic and social power

that are likely to strongly influence project outcomes.


        Recent studies have shown that the uncritical adoption of the term community is

particularly problematic for participatory projects that seek to empower people who are

excluded or without voice (see, for example, Guijit and Shah 1998; Cooke and Kothari

2001; and Mosse 2001). What is labelled a community is often an endogenous construct

defined by the parameters of a project, by project facilitators, or by the nature of

                                                   13

administrative or identity boundaries rather than an organic form.4 Also, the effectiveness

of participatory strategies may hinge on an explicit understanding of local structures of

power, which both limit and enhance prospects for participatory development.


Social Capital


        The third key concept in the literature on implementation of community

participation projects is social capital. This term, which entered the literature on

participatory development in Robert Putnam's (1993) work on Northern Italian

communities, has strongly affected mainstream thinking on community-based and -driven

development.5 It has become commonplace, for example, to refer to such projects as

building social capital (Dongier and others 2003), or creating "assets for poor people"

(World Bank 2000). Putnam (1993, p. 167) describes social capital as the "features of

organization, such as trust, norms, and networks, that can improve the efficiency of

society by facilitating coordinated actions." It refers to the ability of individuals to build

"bonds" within their own group and "bridges" to other groups and is deeply tied to the

belief that the quality and quantity of group activity are key sources of a community's

strength and its ability to work for its own betterment. Social capital is thus a stock from




4In essence, the word community need not describe an object as much as it is describes a concept.

It may thus be more useful to see it as an analytical rather than an empirical concept (Gusfield

1975).


5See Woolcock (1998) for a review.


                                                      14

which people can draw to improve their incomes and which can be "built" to facilitate

economic growth and development (Grooteart 1998).


        This version of social capital has been criticized on many grounds, among them

for not being concerned enough with issues of class distinction and power (Fine 2001;

Harriss 2001); for ignoring reverse causality, with the link going from wealth to more

group activity rather than the other way (Portes 1998; Durlauf 2001); and for not

recognizing that it can be destructive as well as constructive (Portes 1998). The

operationalization of Putnam's ideas has recognized neither the complex strategic,

informational, and relational choices that underpin the endogeneity of community

formation nor the fact that community is itself an abstract social construct. Policy

recommendations thus focus on building communities to facilitate processes of

community participation, without recognizing that investing in social capital may not be

as straightforward as investing in physical capital and may be considerably more complex

than investing in human capital. Returns to investments in social capital can be realized

only in concert with other members of a group and are thus likely to be subject to all the

concerns about collective action.


        The delinking of power and social relationships has been criticized as one of the

main problems with the World Bank's application of Putnamian social capital (Harriss

2001). A more nuanced understanding sees social capital as part of the relations of power

within a social system and recognizes that different groups within a social system can

have different types of social capital. It also recognizes that social capital must be viewed

contextually since it is embedded within structures of power and can be used to facilitate



                                                   15

collective action for the common good or to perpetuate symbolic or actual violence

against others. The rich may have better internal and external networks than the poor and

may use these networks to reproduce unequal systems of domination, as explored in the

work of the sociologist Bourdieu (1984, 1990, 1998), whose ideas precede Putnam's by

at least two decades. The poor have less powerful networks that may help them cope with

the vicissitudes of life but also restrict their chances for mobility. And to the extent that

such capital can be bequeathed, it can contribute to the perpetuation of inequality.


        Lately social scientists associated with the World Bank have argued that social

capital is less an original theoretical concept and more an umbrella term that has

facilitated the insertion of social relations into the thinking of development institutions

dominated by economists (Bebbington and others forthcoming). However, now that

social capital has made such powerful inroads into development thinking, its value as a

"Trojan Umbrella" may have passed and it may be time to bring in all the complexity of

thought on society and culture to inform the design of development practice--as the best

community participation projects already do. Notions such as trust and norms are not

generalizable, and that means that social capital has to be understood within its cultural

and political context (Krishna 2002; Rao 2001), with Bourdieu supplanting Putnam as the

main theorist in the area. Building the capacity for collective action cannot be divorced

from a deep sense of the structures of power within which the poor attempt to cope

(Harriss 2001; Appadurai 2004; Rao and Walton 2004). The question that remains is

whether large development organizations can apply these complex notions of the capacity

for collective action in the everyday practice of project implementation.




                                                    16

         In sum, precisely because community-based and -driven development turns the

pyramid of development mechanisms upside down, by giving beneficiaries voice and

choice, it cannot ignore the social and cultural context within which beneficiaries live and

organize themselves. One possible consequence is that universalistic notions such as

social capital or community may have to be viewed as deeply contextual and endogenous

constructs. This implies that terms such as best practice should be retired to the archives

of development, and much greater emphasis should be placed on contextualized project

design.



Evidence on the Impact of Community-Driven Development Initiatives


         This section examines the impact of community participation projects, looking at

whether they improve targeting, public service delivery, and project sustainability and

reduce the risk of capture by elites. It also looks at the role of external agents and the

state.


Does Community Involvement Improve Targeting?


         Selecting beneficiaries. Community involvement in selecting the beneficiaries of

antipoverty programs has become increasingly common.6 Incorporating local knowledge,

it is argued, can improve targeting, lower the informational costs of delivering




6The word community is typically used in the evaluation literature to indicate different types of

target groups or the larger context in which they live. This section often uses the word in an

uncritical way because it follows the literature being reviewed.


                                                     17

antipoverty programs, and ensure higher quality monitoring of program implementation

(Chambers 1993; Ostrom, Lam, and Lee 1994; Uphoff 1986; Narayan 1998). But such

informational advantages are likely to be realized only when there are institutions and

mechanisms to ensure local accountability. Some argue that such institutions are more

likely to emerge in societies that are highly mobile, with a tendency toward homogeneous

neighborhoods (see, for example, Seabright's 1996 analysis in the context of political

decentralization). Where mobility is low, communities are more likely to reflect social

orderings with long histories and deeply entrenched power hierarchies--just where

poverty programs are most needed.7 Consequently, local inequality in relations of power

and authority may well allow program benefits to be captured by nontarget groups. In the

extreme, the decentralization of poverty programs in such contexts could worsen local

inequality and perpetuate local power relations.


        Conning and Kevane's (2002) recent review of community-based targeting

highlights this potential tradeoff. They find that while community groups are likely to

have better information on who the poor are, only communities that have relatively

egalitarian preferences, relatively open and transparent systems of decisionmaking, or

clear rules for determining who is poor will tend to be more effective than outside

agencies in targeting programs to the poor within those communities. Heterogeneous

communities, where people have multiple and conflicting identities, may pose a

challenge because of competing incentives. They also note that communities vary in their



7See also Agarwal and Gibson (1999), in the context of the literature on natural resource

mobilization.


                                                 18

ability to mobilize information and monitor disbursements. This could affect cost-

efficiency and create further opportunities for corruption and capture by elites.


        What do evaluations of community-based targeting mechanisms say on this issue?

Galasso and Ravallion (forthcoming_ examine the targeting performance of a

decentralized poverty program in Bangladesh, the Food for Education program.8 First

participating Union Parishads (which typically have about 15 villages) were identified by

the center and then eligible households were identified at the community level. Using

both household and community data, Galasso and Ravallion examine how much of the

program's performance in reaching poor families was due to the center's efforts to reach

poor communities and how much to the communities' efforts to reach their own poor.

They also look at the factors that influenced targeting at both levels. They are specifically

interested in the role of village institutions and land inequality within villages as

determinants of targeting performance.


        While they find a positive targeting differential (the difference in optimal

spending on the poor and the non-poor), with more poor people than non-poor people

receiving the program, the size of the effect is small and most of the targeting differential

arises from targeting within villages. They find no evidence that the center is targeting

villages at all. The program's eligibility criteria for selecting participating Union

Parishads, for example, cannot account for any of the variation in the targeting




8Ravallion and Wodon (2000) show that the program accounted for some 13 percent of primary

school enrollment.


                                                   19

differential or in allocations to the poor. If anything, the inter-village component of the

targeting differential tends to worsen the overall differential in participating villages.9 In

contrast, they find that structural features of the village are significant predictors of

targeting performance across villages. Villages that were more isolated or had higher

levels of land inequality tended to have poorer targeting performance. The evidence thus

indicates that the performance of decentralized targeting programs can be substantially

constrained by local inequality.


         Alderman (2002) evaluates an Albanian economic support program (the Ndihme

Ekonomika) that provided social assistance to some 20 percent of the population. The

program provided a block grant to communes and allowed local officials to determine

eligibility and the amount of the transfer to beneficiary households. Alderman finds that

local officials were able to target recipients better than the center could have done using

proxy entitlement indicators. He surmises that communities were able to improve

targeting by using specialized information unavailable to the center.


         Other studies suggest that the center can identify poor communities reasonably

well, but does much less well at identifying the poor within communities. For example,

Coady (2001) finds that Progresa, an antipoverty program in Mexico that selected poor




9Note, however, that since the center identified Union Parishads while the analysis is done at the

village level, variation in eligibility criteria across villages within Union Parishads may be partly

responsible for this result.


                                                         20

households on the basis of census data without any community involvement, was more

effective at targeting poor communities than at targeting poor households within them.


        A series of studies on social funds, which have become a popular mechanism for

public service delivery, also provide evidence on the targeting performance of

antipoverty programs. In a typical social fund, community infrastructure is built with

local participation in the selection and management of facilities. In most cases,

community representatives send project proposals to a central public agency, which

allocates funds on the basis of such criteria as extent of community involvement,

community capacity for collective action, and other factors affecting the feasibility of the

proposed project. Social fund rules almost always require that funds be targeted to poor

communities, in accordance with specific targeting criteria, and often require some

cofinancing from fund recipients. Most social funds restrict the menu of feasible projects

to a limited number of public goods, usually schools, clinics, roads, and water and

sanitation facilities.


        Paxson and Schady (2002) assess poverty targeting in the Peruvian social fund

Foncodes using district-level data on expenditures and poverty. They find that the fund,

which emphasized geographic targeting, successfully reached the poorest districts but not

the poorest households in those districts--better-off households were slightly more likely

than poor households to benefit. Using propensity score matching, Chase (2002)

compares targeted communities in the Armenia Social Fund with communities that had

not received projects but were in the pipeline for them. He finds that while the social fund

was targeted to areas with the poorest infrastructure, these were not always the poorest



                                                  21

areas. The fund was also slightly regressive in targeting households in rural areas. Like

other social funds, the Armenia fund required a community contribution. Chase reports

some anecdotal evidence that this may have led to a selection bias against the poorest

communities, which are often unwilling or unable to contribute. Using similar techniques

to evaluate the Nicaragua Social Fund, Pradhan and Rawlings (2002) find that some

social fund investments were well targeted to poor communities and households. The

World Bank's Operations Evaluation Department, in reviewing these and other

evaluations of social funds, concludes that while social funds' geographic targeting has

been mildly progressive, social funds were less effective in targeting poor households

(World Bank 2002).


        In evaluating the success of Argentina's Trabajar 2 program, introduced in 1997

with World Bank support, Jalan and Ravallion (2003) and Ravallion (2000) find that it

significantly expanded an earlier workfare program in order to provide short-term work

to poor households and to locate socially useful projects in poor areas. The central

government allocated funds to the provinces, making an effort to provide more program

funding to poorer provinces. Provincial governments then allocated funds to projects

within the provinces. Projects were proposed by local governments and NGOs, which

bore the non-wage costs of the projects.


        Jalan and Ravallion (2003)show that program participants were overwhelmingly

drawn from among the poorest households, reflecting successful self-targeting based on

the low program wage. Ravallion (2000) finds a marked improvement in central

government targeting of poorer provinces. He also finds some improvement in reaching



                                                 22

poorer areas within provinces. About a third of the overall improvement in targeting poor

areas came from better targeting of provinces, and the rest from better targeting of poor

areas within provinces. However, despite higher provincial allocations to poorer

provinces, provinces were initially less able to target their poor areas, possibly because

wealthier areas were better able to propose and cofinance good projects. In response to

this, project monitoring tools were set up which continuously updated targeting

performance at the intra-district level. Ravallion (2000) has shown that this simple but

powerful tool, which could be put in place within the context of regular project M&E,

was able to substantially improve the intra-provincial targeting of the poor. This project

demonstrates how targeting of poor areas within a decentralized system can be improved

by providing stronger incentives from the center for pro-poor targeting by local

governments.


        Preference targeting. Since one of the theorized advantages of participation is that

it makes development demand driven, participation should improve the match between

what a community needs and what it obtains--what can be called preference targeting

(see, for example, McLean and others 2001). There is little reliable evidence on

preference targeting, which ideally requires panel data with baseline information on

major problems faced by the community and post-project follow up in experimental and

control communities. Rao and Ibanez (2003) present evidence on the Jamaica Social

Investment Fund. Sampled communities were asked to answer retrospective questions on

their assessment of major problems in the community before the social fund had been

introduced. Rao and Ibanez find that the overall match between the project and the

problems identified was poor. Only in two of the five communities studied did the project

                                                 23

match the preferences of a majority in that community. Better educated and better

networked people were more likely to obtain projects that matched their preferences.

Thus overall preference targeting was poor, and preference targeting was worse for the

most deprived within the community. Despite that, some 80 percent of respondents were

satisfied with the chosen projects. This suggests that despite weak initial targeting, a

broad majority of beneficiaries were satisfied with the public good provided.


        Mechanisms for identifying who benefits. The literature also assumes that a

community's definition of who is poor and vulnerable would be similar to that of social

planners. This may not always be the case. Harragin (forthcoming), in a case study of

famine relief efforts in Southern Sudan, finds that local notions of how food should be

distributed differed from those of aid workers, resulting in a poorly designed project.

Since what the community perceives as fair and just may be at variance with the

preferences of project supervisors, how should the effectiveness of targeting in

community-based projects be evaluated--whose preferences should count?


        Political economy considerations and perverse incentives created by project

performance requirements can also result in poor targeting. Galasso and Ravallion

(forthcoming) and Ravallion (2000) note that a desire to ensure a broad geographic

spread of participants can weaken pro-poor geographic targeting by the center. Jalan and

Ravallion (2003), in their evaluation of the Trabajar 2 program, also note that social

networks were a crucial determinant of who benefited from the workfare program. They

argue that this can be corrected in the design of the program by offering a wage low

enough to discourage wealthier members of the community from participating.



                                                 24

        Schady (2000) shows that community support for then-President Alberto Fujimori

affected the allocation of projects in the Peruvian social fund, Foncodes. Specifically,

districts with a smaller share of pro-Fujimoro votes than in the previous election received

more funding--presumably to increase Fujimoro's popularity. This association was

strongest in wealthier districts--again suggestive of a political link. Thus programs that

attempt to elicit community participation and meet community demand may be no less

immune to political manipulation than strictly top-down programs.


        Morris and others (1999), in their study of the Plandero antipoverty program to

increase the incomes of the rural poor in western Honduras, find that better-off areas

were the most likely to receive program assistance while the most deprived areas were

the least likely. They argue that the weak targeting was due to the project's

implementation schedule, its rate of return criteria, and an evaluation strategy that

emphasized economic results for beneficiary farmers. These jointly created an incentive

to select areas that were easily reached--which typically tended to be better-off areas--

and to target project benefits to better-off households within these areas since they tended

to be the most creditworthy and most able to absorb project funds.


        These ethnographic and case studies suggest that the mechanisms used to identify

beneficiaries are crucial in determining how pro-poor decentralized targeting will be,

especially when community members have unequal access to project implementers. The

center's ability to target benefits to poor households may be constrained by information

gaps or political economy concerns. While decentralized targeting can improve

outcomes, it does not automatically solve the targeting problem. Under certain



                                                 25

conditions, local inequality can even worsen when targeting is decentralized.



Does Community Involvement Improve Public Service Delivery?


        Most large-scale community-based development projects supported by aid

organizations aim to improve access to public services in one of two ways. The project

either produces the public good itself or enhances a community's capacity to act

collectively to obtain the public good from other providers, usually through social

mobilization activities.


        Assessing whether projects that produce public goods with community

involvement are more effective than those that supply public goods produced by others,

typically by governments, would focus on whether such projects were more consistent

with the preferences of target groups, better designed, more sustainable, and deliver

public services better. Assessments would ideally compare similar projects developed in

these two ways. But few evaluations do this in a completely satisfactory way. Most of the

evidence reviewed here compares community-driven development project areas with

communities that are otherwise similar but have no projects or have projects of unclear

method and provenance. One exception is Khwaja's (2001) evaluation of community-

driven projects funded by the Agha Khan Rural Support Program in Northern Pakistan.

Khwaja compares a random sample of these projects with other projects in the same

village that involved no community participation. Khwaja finds that community-managed

projects are better maintained than projects managed by the local government. He also

finds that community participation in technical decisions reduced the quality of



                                                26

maintenance , while community participation in non-technical decisions significantly

improved maintenance.


       In a meta-analysis of reports on 52 USAID projects with participatory elements,

Finsterbusch and Van Wincklin (1989) conclude that less technically complex and

smaller projects were more effective than others. They also find that participatory

projects were more effective in more developed countries than in less developed

countries--suggesting that the broader institutional environment may play an important

supportive role. While this appears to suggest that institutional reform is required before

participatory projects are introduced, the question is actually quite difficult to answer

because institutional reform probably makes all types of development initiatives more

effective.


       Facilities constructed with community involvement tend to improve access to

public services. Paxson and Schady (2002) find that the Peruvian social fund Foncodes

increased school attendance, particularly for younger children. Chase and Sherburne-

Benz (2001) reach a similar conclusion for the Zambia Social Fund. They also find that

household education expenditures rose where there was a school constructed by the social

fund and that use of primary care services and the prevalence of child vaccinations

increased where there was a health facility constructed by the social fundNewman and

others (2002), using both panel data and random assignment approaches to evaluate the

Bolivian Social Fund, find a significant reduction in under-five mortality due to the

provision of health clinics but little change in education outcomes with the provision of

education projects. The reason, they surmise, is that investments in health went beyond



                                                  27

providing infrastructure to providing medicine, furniture, and other necessary inputs.

Newman and others (2002) also find that water projects improved water quality and

access to water only when community-level training was also provided. These findings

suggest that to succeed, participatory projects need to go beyond construction of facilities

and may require the continuing involvement of external agencies in providing marginal

inputs and training.


       King and Ozler (2000) evaluate the impact of school autonomy on student

performance in Nicaragua, comparing participating and nonparticipating schools.

Participating public schools acquired de jure responsibility for a range of functions

previously managed centrally, including the right to hire and fire the director and

maintain the school's physical and academic quality. School management boards were set

up to handle these tasks, drawing their membership from among school staff, parents, and

students. King and Ozler find that students in schools with both de jure and de facto

autonomy performed better on standardized tests than students in schools that did not

have autonomy or that had only de jure autonomy. Their work suggests that school

effectiveness could be increased by decentralizing certain aspects of school management.


       Jimenez and Sawada (1999) report that community-managed schools in El

Salvador had fewer absences than comparable centrally managed schools. Jimenez and

Paqueo (1996) find that schools in the Philippines that relied more on community

contributions used their resources more efficiently. This literature, while not about

community-driven development in the strict sense but about the impact of participation

on the provision of schooling services, unambiguously supports the idea of increasing



                                                 28

school autonomy to improve the quality of public schools.


        While these studies provide a positive impression of the impact of participation on

project effectiveness, other studies are more ambiguous. Gugerty and Kremer (2000) use

random assignments to evaluate the impact of a participatory program to provide

agricultural inputs that was designed to build social capital among women in Western

Kenya. They find that the costs of inputs were greater than the benefits (increased

agricultural output).


        Hoddinott and others (2001) look at the relationship between participation and

project employment outcomes using data on public works programs from South Africa's

Western Cape Province and attempt to isolate the impact of de jure and de facto

participation on each outcome using an instrumental variable strategy. Their instruments

for participation are measures of community fractionalization (percentage of adults who

are divorced, a measure of racial fractionalization, an index of political fractionalization).

They find that participation has no effect on any of their outcome variables. When they

interact the participation variable with project type, the effects of participation are

somewhat stronger but still quite weak. However, it is unclear whether their instrument

set is valid for the outcome variables they analyze, and their results are also based on a

small set of projects.


        In addition to impact evaluations, there are other types of studies that define the

causal effects less clearly but that are nevertheless interesting. Katz and Sara (1997), who

analyze the performance of water systems in a variety of countries, find that performance

was markedly better in communities where households were able to make informed


                                                  29

choices about the type of system and the level of service they required and where

decisionmaking was democratic and inclusive. Projects constructed without community

supervision or management tended to be poorly constructed by private contractors.

Where construction was of poor quality despite community participation, the reason was

often inadequate technical support--similar to Khwaja's (2001) findings on communities

faring worse on technical decisions. Katz and Sara also report that community members

were more willing to pay for investment costs when they had control over the funds.

When government staff or contractors controlled the funds, communities viewed their

contribution as a tax rather than a fee for service.10


        Isham and Kahkonen, in analyses of water projects in Indonesia (1999b) and India

and Sri Lanka (1999a), confirm that greater community participation is associated with

better water supply and that well designed community-based water services lead to

improvements in health outcomes. Differences in project effectiveness are explained

largely by a community's ability to engage in collective action, with high levels of

"social capital" improving participation in design and monitoring. This is also the




10A long report by Zyl, Sonn, and Costa (2000) on the Northeast Rural Development Program

and the Rural Poverty Alleviation Program in Northeast Brazil summarizes the results of several

studies and find that the combined impact of these projects was "impressive" and that the projects

were well targeted and extremely beneficial to poor communities. While their methodology and

analysis are never clearly outlined, the paper suggests that none of the studies had an adequate

counterfactual or control group, though they sometimes had a baseline. Thus it is difficult to

judge how "impact" was calculated.


                                                    30

conclusion of Rao and Ibanez (2003), who find that a community's capacity for collective

action influences its ability to successfully apply for funds from the Jamaica Social

Investment Fund. Poorly organized communities are not only less likely to obtain

projects, but are also more likely to mismanage projects that are allocated to them.


Does Community Involvement Improve Project Sustainability or the Capacity for

Collective Action?


        Few studies have examined the relationship between community-based

development projects and capacity for collective action. Finsterbusch and Van Wincklin

(1989), in their review of USAID projects, claim without ambiguity that projects with

participatory elements increased the overall effectiveness of projects particularly in

building capacity for collective action. They make no attempt, however, to identify the

causal direction of this claim, which they assert largely on the basis of an anecdotal

assessment of project reviews. In a study of incentives for collective action in a district in

South India, Wade (1987) reports that some villages had what he calls a "public realm,"

with well functioning institutional arrangements for managing common-pool resources,

while others had virtually no public realm. Accounting for the difference, he argues, are

economic factors that gave rise to differential collective benefits from organization. This

suggests that stimulating participation by forming community organizations such as water

user groups may require more than educating people about their common interests or

promoting communal values. It may also require helping them to understand the

collective benefits of participation.


        Rao and Ibanez (2003) use propensity scores and qualitative methods in

                                                  31

constructing a survey with a contextualized, project-specific set of variables to explain

selection into the program, thereby minimizing unobserved heterogeneity. They then

match on observable differences to examine whether the Jamaica Social Investment Fund

increased the ability to engage in collective action and build trust. The answer is yes, but

the effect was greater for more educated, better networked members of the community.

Gugerty and Kremer (2000), in their random assignment study, also find that bringing in

outside assistance may change the composition of beneficiary groups. The formation and

training of village groups attracted wealthier and more educated men and women into

leadership positions within the group because of the outside funding.


        Does participation increase project sustainability? Here again the evidence is

limited but instructive. Khwaja's (2001) study suggests that projects managed by

communities are more sustainable than projects managed by local governments because

of better maintenance. Katz and Sara (1997) and Isham and Kahkohnen (1999a,b) also

find strong associations between participation and sustainability, although they do not

establish causal direction. In an in-depth anthropological study of the sustainability of

participatory projects, Kleemeier (2000) examines a Malawi rural piped water project and

finds that half the schemes are performing poorly, with the newest ones performing best.

The weak sustainability, she argues, stems largely from the lack of institutional support

from external agencies--echoing the conclusions of Katz and Sara (1997) and Newman

and others (2002).


        For water projects in Sub-Saharan Africa, Cleaver (1999) finds that even if

communities are initially successful in creating the project, they may lack the material



                                                 32

resources and connections to sustain their efforts. Mosse (1997a comes to similar

conclusions in an in-depth examination of tank management in South India. He finds that

maintenance of community infrastructure is often crucially dependent on external agents.

Thus the need for a well functioning state apparatus does not disappear with active

community involvement.


        While community participation projects have the potential to be more sustainable

than top-down projects, they also appear to suffer from neglect by line ministries once

they are completed. Several studies suggest that unless communities can lobby for

continuing support for marginal inputs and training, their ability to sustain such projects

may be limited.


Economic and Social Heterogeneity and the Risk of Capture by Elites


        Previous discussion suggested that a naive understanding of the concepts of

community, participation, and social capital can obscure differences that critically

influence outcomes. Such differences may be driven by local structures of power; by

social divides based on gender, caste, race, or ethnic identities; by uneven interest in the

particular public good; or by economic inequality and uneven resource distribution. How

do such differences shape outcomes? When is heterogeneity good for collective action,

and when does it constrain true participation?


        Heterogeneity and inequality. There is a substantial and varied theoretical

literature on collective action and coordination by economists, sociologists, and

anthropologists that examines the relationship between heterogeneity and the capacity for



                                                 33

collective action. The literature identifies several constraints to collective action and the

types of environments that favor or disfavor coordination, offering a broad set of

hypotheses for empirical work.11


         In a classic study, Olson (1973) argues that certain types of inequality might

favor the provision of public goods. In particular, Olson holds that collective action is

difficult to mobilize in large homogeneous groups in which no individual could make a

significant difference in the level of provision of the good. In contrast, smaller and more

unequal groups might do better, although there would be free-riding by those with a

smaller interest in the public good.


         Several other theoretical studies also suggest that inequality can be conducive to

the provision of pure public goods. For example, Bergstrom, Blume, and Varian (1986)

demonstrate that, when individual preferences over private goods are identical and

distributional changes mean preserving, a redistribution of wealth can leave the supply of

the public good unchanged or even increase it where individual preferences over private

goods are identical and distributional changes are mean preserving. Itaya, Meza, and

Myles (1997) show that income inequality that is so great that only the rich contribute to

the public good increases welfare relative to the case in which all individuals contribute.

Baland and Platteau (1998) show that inequality in entitlements to resource use is




11It is impossible to do justice here to the rich literature on heterogeneity and collective action.

For a review of the literature in economics, in the context of environmental resources, see Baland

and Platteau (2003).


                                                        34

associated with greater conservation when there are decreasing returns to effort for the

resource-use technology.


         There have been several critiques of Olson's group-size paradox. Oliver, Marwell,

and Teixeira (1985) argue that the probability, extent, and effectiveness of collective

action depend on the relation between contributions and levels of the public good and on

the extent and type of heterogeneity in the population. They show that inequality of

interest can increase the level of collective action. Since inequality of interest is more

likely in larger groups, public goods may be provided at higher levels in large groups.

However, free-riding remains a problem, and sub-optimal levels of collective action and

provision of public goods are possible in large groups. Oliver, Marwell, and Teixeira also

show that a strong positive correlation between interest and resource heterogeneity

increases the probability of collective action.


         Heckathorn (1993) argues that the polarizing effect of heterogeneity depends on

how collective action is organized. The key issue is that collective action can impose

differential costs and benefits on members of a heterogeneous community. If all costs are

covered by voluntary contributions, there is no polarization. If the collective action

imposes negative net benefits on some members, polarization and conflict are likely.


         The literature on collective action related to common-pool resources has also

focused considerable attention on heterogeneity.12 Russel Hardin's (1982) prisoner's




12All public goods have the property that many people can use them at once, thus exclusion is

difficult. Some public goods yield infinite benefits, but not all. Common-pool resources are

                                                      35

dilemma outcome for common-pool resources with open access argues that individually

rational strategies can lead to collectively irrational outcomes. More recently, Dayton-

Johnson and Bardhan (2002) show that the relationship between inequality and levels of

collective action in conservation can be U-shaped. In theoretical work based on fisheries,

they show that if fishers have earnings opportunities outside the commons that are

concave functions of wealth, increased inequality has a negative effect on conservation.

However, as inequality increases, levels of conservation can rise and perfect conservation

is possible under perfect inequality.


        The dominant economic approach to collective action on a common-property

resource focuses on individual incentives to contribute, as determined by the private

benefits and costs of participation. Baland and Platteau (forthcoming) show that when

collective action involves conservation, users' time preferences are also likely to be

important. Specifically, people with a shorter time horizon are more likely to adopt

strategies that yield immediate results and to neglect longer term considerations.

However, variations in time horizons are likely to be related to the initial distribution of

wealth. The poorest users may have wealth levels that are so low that collective action

may violate their survival constraint. Thus, the poor may face particular barriers to

collective action.




public goods with finite or subtractive benefits--they are thus potentially subject to congestion,

depletion, or degradation (Blomquist and Ostrom 1985).


                                                    36

       The role of economic inequality is similarly conditioned by other factors. Baland

and Platteau (1999) argue that the interactions among multiple dimensions of inequality

need to be considered to understand how they affect collective action. For instance,

economic inequality can have very different effects when it is combined with caste or

ethnic polarization than when it occurs within a more mobile social structure. When

regulatory agencies have the power to enforce fines and sanctions, collect fees, and

impose rules, inequality is less likely to constrain collective action (Baland and Platteau

(1998).


       Wade (1987) notes that cultural forces often shape collective action by providing

conventions or norms that act as implicit or informal equivalents of all-or-nothing

contracts (or assurance games) in resolving collective action dilemmas. He reports that

effective village bodies are often composed of elites who can exercise authority with little

or no pretense at representation. Thus, he argues, it may be necessary to organize around

existing structures of authority, with a major role for village elites.


       More recently, Chwe (1999, 2001) builds on this idea of regulatory authority. He

argues that most models of collective action implicitly assume the pre-existence of

"common knowledge" (see also Bardhan 1993 on this point). This common knowledge

permits games of strategy to be played with a common understanding of the rules of the

game--everyone knows what everyone else is playing. The common knowledge

assumption is arguably the core concept behind amorphous notions such as trust and

social capital that figure prominently in the discourse on collective action. Rao (2003)

builds on this to argue that mechanisms for generating common knowledge, which he



                                                  37

calls "symbolic public goods," are a precursor to the possibility of collective action in

community development because they may mitigate the effects of heterogeneity. People

who belong to the community abide by the rules of the community both because they

internalize its ideology and because they face sanctions if they violate the rules. The

incentives work not just as exogenous constraints but as what can be called "constraining

preferences" (Rao and Walton 2004). Thus studying the ideology and beliefs underlying

collective action is key to understanding how it interacts with inequality and

heterogeneity.


        Empirical work on heterogeneity and collective action has focused mostly on

economic or social (race, caste, ethnicity) heterogeneity. Social heterogeneity is the more

difficult concept to measure because it is usually more than a matter of creating an index

that weighs the number of identifiable groups in the community by the size of their

populations--though most of the economics literature does precisely this.


        Alesina and La Ferrara (2000) examine the influence of heterogeneity on

participation using survey data on group membership in the United States and data on

U.S. localities. They find, after controlling for many individual characteristics, that

participation in social activities is significantly lower in more economically unequal or

more racially or ethnically fragmented communities. They also show that heterogeneity

has the most significant impact on participation in groups where excludability is low and

significant interaction among members is necessary. The study attempts to deal with the

possible endogeneity of the income inequality measure. It is unclear from their study,

however, what the marginal impact of any given heterogeneity measure is. In particular,



                                                    38

it is unclear how much racial or ethnic heterogeneity matters once economic inequality

has been controlled for.


        In a similar study, La Ferrara (2002) looks at the influence of inequality on group

participation using data from rural Tanzania. She also finds that higher levels of village

inequality reduce the probability of participation in any group. She reports that groups in

more unequal communities were less likely to take decisions by vote, were more likely to

report misuse of funds and poor group performance, and were more likely to split up into

homogeneous ethnic and income groups, to interact less frequently, and to be less

motivated to participate.


        In a random assignment evaluation of community-managed schools in Kenya,

Gugerty and Miguel (2000) show that ethnic diversity results in fewer social sanctions

against non-participants. This leads to low parental participation in school activities and

poor attendance by teachers, which in turn reduce the funds available to the school and

worsen school quality.


        Bardhan and Dayton-Johnson (2001), in a survey of the evidence on community-

driven water projects in India, Mexico, and Nepal, conclude that heterogeneity has a

negative impact on cooperation and commons management and weakens the cohesive

effect of social norms and the application of sanctions for violating cooperative behavior

and collective agreements. Bardhan (2000), examining the factors affecting irrigation

maintenance in South India, also finds that inequality has a negative effect. Work by

Ostrom, Lam, and Lee (1994) and Ostrom (1990) shows that farmer-managed irrigation

schemes had more equitable water distribution. However, these studies can say little


                                                 39

about the impact of community heterogeneity on project choice or on the distribution of

benefits since they have little to say about non-participants.


        Not all quantitative studies, however, find that economic inequality has a

monotonically negative impact. Somanathan, Prabhakar, and Mehta (2002) analyze the

effect of collective action on forest conservation in India using matched-comparison

difference-in-difference techniques. They find that communities with more equal land

ownership conserved pine forests somewhat better than other communities. However,

they find no impact of caste heterogeneity on conservation efforts. They also find that

more equal land ownership does not translate into better conservation of broadleaf

forests, which are of much greater importance to villagers for firewood and fodder. Some

in-depth case studies of participatory projects using qualitative methods also suggest a

more complex role for social heterogeneity (see, for example, Vedeld 2000.


        Dayton-Johnson (2000) develops a model of cooperation in small irrigation

systems and tests the model with data from a survey of Mexican irrigation societies. He

finds that social heterogeneity is consistently and significantly associated with lower

levels of maintenance. Landholding inequality and the proportional sharing of water in

accordance with landholding size also tend to decrease maintenance, but in a

nonmonotonic way. Khwaja (2001) finds empirical evidence for a U-shaped relationship

between land inequality and project maintenance, with both highly equal and highly

unequal communities performing better than moderately equal ones.


        Capture by local elites. The frequent tendency for participatory projects to be

dominated if not captured by local elites is highlighted by several case studies. Katz and


                                                 40

Sara (1997), in a global review of water projects, find numerous cases of project benefits

being appropriated by community leaders and little attempt to include households at any

stage. They find that training community members was crucial for informed choice and

for maintenance of projects and willingness to pay for them. And while well trained

project staff were critical for ensuring inclusiveness and providing information on

options, Katz and Sara note that even well trained staff are not always effective in

overcoming entrenched norms of exclusion. In a study of community forestry projects in

India and Nepal that worked reasonably well, Agarwal (2001) reports that women were

systematically excluded from the participatory process because of their weak bargaining

power. Rao and Ibanez (2003) find that in the participatory projects in their Jamaican

case study, wealthier and better networked individuals dominated decision making. In a

similar case-based evaluation of social funds in Jamaica, Malawi, Nicaragua, and

Zambia, the World Bank (2002) Operations Evaluation Department concludes that the

process was dominated by "prime movers."


       Abraham and Platteau (2004) present evidence on community participation

processes in Sub-Saharan Africa based largely on anecdotal evidence from their work in

community-based development and on secondary sources. They argue that rural African

communities are often dominated by dictatorial leaders who can shape the participation

process to benefit themselves because of the poor flow of information. Thus participatory

development can be very difficult and may require slow, careful efforts to make

communities more receptive.


       The related literature on decentralization also provides insights. Bardhan and



                                                  41

Mookherjee (2000) theoretically evaluate the hypothesis that local governments have

better information but are less accountable and thus are more prone to capture. They

show that the probability of capture increases with local inequality and it is unlikely that

local governments are universally more prone to capture..


        Some degree of elite domination may be inevitable in community participation

projects, particularly in rural areas, where the elites are often leaders who embody moral

and political authority. Often these elites are the only ones who can effectively

communicate with outsiders, read project documents, keep accounts and records, and

write proposals. This domination may, however, work against the kind of broad-based

democratic participation envisioned by advocates of community-driven development

because an awareness that project rules have been crafted by the elite may discourage

community participation in the project (Bardhan 2000).


        Rao and Ibanez (2003) argue, however that elite domination is not always elite

capture. In their case study, they find a potentially more benevolent form of elite

domination, with more than 80 percent of beneficiaries ultimately expressing satisfaction

with the project. Substantiating this point, Khwaja (2001) finds that participation by

hereditary leaders tends to improve maintenance.


        Finally, social cohesion may facilitate collective action, but community

participation projects do not necessarily perform better in communities with a high score

on some index of social capital. Mosse (1997b), in a study of tank management in South

India, shows that a traditionally cohesive village will not necessarily have good

development projects, while communities with low levels of cohesion could have good


                                                  42

development projects. Indigenous systems of social organization have very different

"moral claims," depending on the public good around which they are organized. A

festival or a temple embodies motives and constraints toward collective action that are

considerably different from those of a development project, and the logic of social

relations in indigenous societies may be quite contrary to the egalitarian principles that a

community-based or -driven project is trying to propagate. Abraham and Platteau (2004)

make a similar argument for Sub-Saharan Africa. External agencies and programs change

incentives and can change the political and social dynamic. Done naively, this can disrupt

the social equilibrium because they introduce a new element into already contested

domains of power and meaning (Mosse, 1997b).


        Summary. The social role of heterogeneity within a group is complex.

Heterogeneity can increase or decrease social cooperation and can polarize or strengthen

group identity. The success of community-driven development may also be affected by

how well heterogeneity is "managed"--by the resources and strategies that are used to

bring communities together and by how effectively differences are discussed and

solutions arrived at. Every time an external agent interacts with a group of people, it

creates competition between different interests and incentives. The success of a project

may thus depend on how those incentives are aligned--whether by persuasion, ideology,

consensus, good governance, domination by greedy elites, or sheer hard work by a group

of altruistic individuals.


The Role of External Agents and the State


        Community initiatives are usually designed by a central authority that sets the

                                                43

basic parameters and the mechanisms for disbursing funds. Project implementers are also

extremely important to the effectiveness of projects. Frontline staff who work directly

with beneficiary groups are especially critical actors in building participatory processes.

They are expected to mobilize communities, build the capacity for collective action,

ensure adequate representation and participation, and, where necessary, break through

elite domination. They must be culturally and politically sensitive, charismatic leaders,

trainers, anthropologists, engineers, economists, and accountants. Despite their centrality,

however, there is virtually no generalizable evidence on their role. What evidence there is

comes from case studies, which present a somewhat dismal picture.


        Jackson's (1997) analysis of field worker diaries in India indicates that field staff

tend to be driven by the incentives they face, which are often not well aligned with the

needs of the project. In particular, Jackson notes that field workers tend to gloss over

local power relations in a rush to show results. Vasan (2002) shows how facilitators in the

forest management projects she examined overlooked project goals when their personal

incentives were not aligned with them. Botchway (2001), studying participatory projects

in Ghana, notes that project facilitators, who are typically young, inexperienced, and

poorly paid, are often vulnerable to manipulation by locally powerful elites. Michener

(1998) echoes this and points out that senior members of the implementing agency are

often former bureaucrats with little experience in community empowerment. Their vision

of development, guided by a lifetime of work with line ministries, is closer to "supply

driven demand driven development" (Serrano 1996; Tendler 2000; Tendler and Serrano

1999).




                                                 44

        Both beneficiaries and facilitators have an incentive to present the impression of a

successful project to outsiders and may collude for this purpose. NGOs often avoid

working in difficult communities, where quick results may be harder to demonstrate. The

community may also use participation as a bargaining chip to extract resources from the

outside agency--and may even construct a version of events that gets them the maximum

benefit. All this adversely effects project quality and sustainability.


        The state is another strategic actor in this context. Beneficiary communities, often

too poor to fund their own teachers, doctors, desks, and medicine (Cleaver 1999;

Kleemeier 2000), remain in need of government support for inputs, maintenance

investment, and trained staff in order to sustain project benefits. Thus, the need for a

responsive state apparatus may increase when community participation projects are

implemented. But some states may manipulate project allocations to satisfy political ends

(Schady 2000]) or may try to use community participation programs to shift certain costs

to community groups. Mosse (1997a) has an extended discussion of this motivation in the

context of irrigation infrastructure.


        Some studies claim that centralized bureaucracies tend to inhibit project

effectiveness, although the evidence is generally weak. Finsterbusch and Van Wincklin

(1989), in their review of community-managed USAID projects, hold that for such

projects to be effective, the organizations that implement them also need to be

decentralized and nonauthoritarian. White (1996) notes that power relations in the wider

society within which participation occurs have to be taken into consideration. In some

cases, the state may have to support broad-based redistributions of power for community



                                                  45

projects to be really successful. This suggests that community-based and -driven

development projects must be seen as part of a shift toward a broad-based participatory

and decentralized system of governance.


       It is unclear, however, how this is to be achieved. Several writers have pointed out

the potential for conflict between local political interests and community organizations.

Thomass-Slayter (1994) notes that as communities get stronger, they often pose a

challenge to local political interests, leading to competitive relations between the state

and community organizations and a withdrawal of state support. Das Gupta and others

(2000) find that community development efforts can be fragile in such circumstances.


       What kind of state creates the right kind of enabling environment for community

development efforts? An extreme example is the Indonesian government's appropriation

of the concepts of mutual assistance (gotong royong) and self-reliance (svadaya) as

centrals tenets of its nationalist ideology. During the long period of military rule,

participation was imposed by a strong state on the local population (Bowen 1986).

Sukarno tried to use the notions of mutual assistance and self-reliance to unify the diverse

groups in the new country and to provide a form of cultural legitimacy to state control.

The state had to be strongly authoritarian and development had to proceed in a

cooperative and collaborative manner.


       Sullivan (1992) in his detailed ethnography of local development in a Javanese

community, demonstrates that the combination of an autocratic state and the principles of

mutual assistance and self-reliance resulted in a form of forced labor: being a good

Indonesian meant contributing labor and cash for development projects. Collective action


                                                   46

was the norm. Grants received by the village headman assumed, in the mismatch between

the size of the funds and expected cost of the project, that most funds would be locally

mobilized. Contributions from the community were mobilized by the ward leaders.

Everyone was expected to contribute free labor--or face social, political, material, and

even physical sanctions. There was no choice but to participate.


       Ribot (1999) makes a similar argument for how participatory ideologies were

implemented in Sahelian forestry. Powers were devolved to village chiefs who were

accountable upward to admininstrative authorities rather than downward to their

constituents. Participation was essentially another form of enforcing central rule. Another

mode was the Indian experiment with local democracy, instituted through a series of

constitutional amendments enfranchising and enabling the Panchayat system of regularly

elected village councils. But while many Panchayats have strong downward

accountability, they often lack the resources and the links with higher levels of the state

to effectively deliver public services (Matthew and Buch 2000).


       Thus, the key to making participation work is to create forms of downward

accountability and simultaneously to maintain close links between the higher levels of

government and the community--a kind of upward commitment. Participation in the

absence of state facilitation can result in a closed village economy, which limits the

possibility for improved public action.



Scaling Up Community-Based and -Driven Development


       Can community-based and -driven development projects be scaled up? This



                                                  47

question cannot be answered unequivocally. There is little generalizable evidence on the

optimal pace of scaling up or on the marginal benefits of increasing community

involvement. But there are stories of successful initiatives that offer some grounds for

optimism. One of the most inspiring is the experience of the Self-Employed Women's

Association (SEWA), inspired by Gandhian notions of self-reliance (Rose 1993). SEWA,

which provides inexpensive credit and insurance to poor women workers, has

successfully unionized its members, enabling them to negotiate for minimum wages,

legal protection, and other equitable treatment. While there has never been an impact

evaluation of SEWA, it is widely regarded as successful. Similarly, the Amul movement

in India has created a cooperative that has been remarkably successful in creating new

markets and achieving higher incomes for dairy farmers across India (Kurien 1997 ).13


Is Rapid Replication Possible--or Desirable?


        One question is whether such movements can be rapidly replicated through

external interventions led by large bureaucracies, as the World Bank and several other

donors are attempting to do. One problem is that it is difficult to replicate the success of a

highly motivated group of charismatic individuals who are able to sustain a long-term




13Uphoff, Esman, and Krishna (1998) document several such "reasons for hope." Binswanger

and Aiyer (2003) review the "gray literature," highlighting examples of successful community-

driven development interventions. Uphoff and Wijeratna (2000) also find evidence to argue that

there was a strong correlation between "social capital" building initiatives in improved efficiency

in water management in the Gal Oya irrigation project in Sri Lanka.


                                                    48

vision of structural transformation through dedication, patience, and creativity. When

tasks are handed over to salaried professionals, motivated by more mundane

preoccupations such as wages and promotions, incentives change.


        Given the pivotal role that facilitators play in the community-based development

process, the effort can be doomed from the start, especially when the facilitators work for

a large bureaucracy, such as a line ministry or the World Bank (Pritchett and Woolcock

2002). They begin "seeing like a state" (Scott 1999). The complicated process of building

community participation becomes routinized and subject to the imperatives of short time

horizons and the need for quick results.


        Failure is not inevitable. But the mechanical application of best practice

guidelines, without attention to local context and adequate monitoring, can easily result in

poor community-based projects. There are generalizable lessons from the successes and

failures of community-based projects (Binswanger and Aiyer 2003), but the lessons need

to be adapted to the historical, political, and social environment where the project is to be

implemented.14


        Social and cultural inequities create several challenges for the poor. Appadurai

(forthcoming) argues that to break structural inequities in social relations and achieve

equitable development it is important to build the "capacity [of the poor] to aspire." This




14The World Bank, for instance, tends to borrow the social fund design from activities in

Indonesia or Zambia and drop them into very different settings. This is particularly undesirable in

the Indonesian case, because of the historical role of a strong state in community development.


                                                     49

often requires organization to realize collectively what individuals cannot aspire to

alone--a way out of the culture of domination and poverty. Rao and Walton (2004)

describe this as building "equality of agency," creating environments to equalize the

relational and group-based structures that influence individual aspirations, capabilities,

and agency. This is the kind of empowerment that advocates of community-based

development envisage.


        Can a structurally transforming, aspiration-building, agency-equalizing,

community-empowering project result from a rapid scaling-up process? As Trawick

(2001) points out, this is unlikely. Community-based development requires the

transformation of a social equilibrium in which traditional systems of social organization

have evolved to manage resources in a manner that serves the purposes of entrenched

elites. Breaking this up is a slow and gradual process. If external agencies try to change

the political and social dynamic without fully understanding it, the social equilibrium can

be severely disrupted, with nothing to take its place (Mosse 1997b). This may sometimes

be necessary, but it needs to be done with full knowledge of the impact--and it cannot be

done quickly. Effective community-based development requires slow, gradual, persistent

learning by doing, with a project design that gradually adapts to local conditions by

learning from the false starts and mistakes that are endemic to all complex interventions.


        Effective learning by doing requires effective monitoring and evaluation. Phillips

and Edwards (2000) show that the culture and incentives of large bureaucracies make

this difficult. Task managers, facilitators, beneficiaries, and evaluation consultants all

have incentives aligned to present a favorable impression, a mindset that hampers



                                                  50

learning and encourages the replication of bad design and practice.15 The vast majority of

community-based interventions, including those funded by the World Bank, have not had

reliable evaluations, based on representative samples with treatment and control groups

and baseline and follow-up data. If the development community is going to embark on a

major change in development strategy, surely it should be concerned about how well it

might work.16


        Community-based approaches are particularly difficult to get right in the initial

stages. Thus successful community participation projects rely much more than other

types of projects on learning by doing, which requires careful evaluations coupled with

phased-in scaling up with constant adaptation. Therefore, evaluations should be a central

part of the project design from the start and not merely a method to judge the

effectiveness of projects after they are completed.


How to Scale Up Community Participation Projects


        Community participation projects can be scaled up in several ways. One is to test

different approaches and methodologies in a carefully selected set of pilots. Learning

from this experiment can then lead to improved project designs applied to a much larger

set of communities, in a process of "piloted scaling up." Another method is to apply a




15See Pritchett (2002) for an economists' version of this argument.


16Several evaluations of community-based development projects have recently begun at the

Bank, and much should be learned from them during the next few years.


                                                   51

design in a limited but large set of communities, say in one province in the country. The

lessons from this large test can be applied to projects expanded in a phased manner to the

entire country, in a process of "phased scaling up." A third possibility is to start with

wide national coverage, in a process of "untested scaling up." In ideal circumstances, the

evidence seems to favor a mix of piloted approaches followed by phased scaling up. But

sometimes, untested scaling up is called for--to deal with an economic emergency, for

instance. The evidence suggests that untested expansion should be applied with extreme

caution, particularly in countries with little experience with community-based

development.


        What, then, are the preconditions for successfully scaling up community-based

projects?17 As argued above, the process has to be gradual. Because of the contextual

complexities involved, initial designs based on best practices are bound to be imperfect.

Rapidly scaling these up, particularly in countries with little experience with community-

based projects, will likely result in failed projects. A historical, political, and social

analysis of the country is critical, in addition to the usual economic analysis, in order to

carefully assess whether the government has the capacity to support community-based

development in a manner consistent with both upward commitment and downward

accountability.


        Second, scaling up requires a strong ethic of learning by doing, with rigorous

evaluations and reliable monitoring systems to provide constant feedback. Ideally, both




17Also see Binswanger and Aiyer (2003) on this point.


                                                   52

qualitative and quantitative methods should be used to provide reliable estimates of

impact and an in-depth examination of context and process (Rao and Woolcock 2003).

Lessons should be incorporated into the next phases of project design to correct mistakes,

which are inevitable.


        Third, careful attention should be paid to the training of a core cadre of

facilitators, who are pivotal to successful community-based interventions. Inexperienced

facilitators should be given a chance to learn and grow, under the supervision and

leadership of more experienced individuals, as part of a gradual learning by doing

process.


        Fourth, there should be a commitment by the country to a cultural change in the

institutional environment, which has to be become more participatory, responsive, and

transparent, with downward accountability. These are not just buzzwords. As Uphoff,

Esman, and Krishna (1998 ; Page 202) argue, "If the expansion is occurring because the

government or donor sources are promoting it, this is less persuasive than if there is a

spontaneous joining of the program or if local governments take over responsibility."

There also has to be a strong upward connection to higher levels of government and line

ministries to facilitate the flow of resources.


        Finally, changing from top-down to bottom-up development in a manner that is

sensitive to local context and culture requires a long time horizon. This also means that

community-based development projects should not be judged hastily. Initial evaluations

may well be unfavorable. The key is to fix the problems and work toward incremental

improvements.


                                                 53

Conclusion


        One important question on the effectiveness of community-based and -driven

development initiatives is the extent to which they successfully target the poor. Evidence

suggests that decentralized targeting has not always been effective, especially in targeting

projects to the poor within communities. In addition, the evidence, while thin, suggests

poor preference targeting--the preferences of the poor have not been adequately

considered in project selection. Finally, political economy considerations and perverse

incentives created by project performance requirements also constrain targeting, although

evidence suggests that decentralized targeting can be made more effective by monitoring

projects to improve performance incentives.


        Another important question is the extent to which participatory development

initiatives improve project quality and performance. Here, there is some evidence that

participatory projects create effective community infrastructure and improve welfare

outcomes, but the evidence does not establish that it is the participatory elements that are

responsible for improving project outcomes. Few studies compare community-based

projects with centralized mechanisms of service delivery, so it is difficult to tell whether

alternate project designs would have produced better outcomes.


        Evidence on the impact of economic and social heterogeneity on project

outcomes, and on collective action capacity more broadly, suggests that the relationship

is complex. While theoretical work by economists has shown that economic inequality

need not constrain collective action, empirical work has shown mixed results. The

targeting of poor communities and poor households within communities is markedly


                                                 54

worse in more unequal communities, particularly when the distribution of power is

concentrated within elites. A number of studies find a U-shaped relationship between

inequality and project outcomes. The role of social heterogeneity is more complex to

measure. However, most econometric studies that have attempted to devise measures of

social fractionalization have shown that fractionalization tends to inhibit collective

activity, but there is also qualitative evidence in the opposite direction.


        Even in the most egalitarian societies, however, community involvement in

choosing, constructing, and managing a public good will almost always be dominated by

elites, who tend to be better educated, have fewer opportunity costs on their time, and

therefore have the greatest net benefit from participation. It is not clear, however, that this

always represents "capture," in the sense of elites appropriating all the benefits from the

public good. It may be useful to distinguish between extreme forms of capture, such as

outright theft and corruption, and what might be called "benevolent capture." However,

when local cultures and systems of social organization result in tight control of

community decisions by elites, malevolent forms of capture become likely. It is important

therefore to understand what types of checks and balances are most effective in reducing

capture and the systematic exclusion of the poor and of discriminated-against minorities.

The problem in assessing elite capture is that there are no studies that look at an

appropriate counterfactual. This remains an important area for future work.


        Several case studies suggest that the success of participatory projects may also be

affected by how well heterogeneity is managed, by what resources and strategies are used

to bring communities together, and by how effectively differences are debated. The



                                                  55

involvement of external agents creates competition among different interests and

incentives, and the success of projects may depend on how these incentives are aligned--

whether by persuasion, ideology, consensus, good governance, domination by greedy

elites, or sheer hard work by a group of altruistic individuals. This is another area where

more research would be useful.


        The level of community cohesion, or "social capital," is also expected to improve

the quality and sustainability of projects. Some studies have shown an association

between the level of some index of participation and project effectiveness, but the

direction of causality is unclear. While community-based development seems likely to be

more effective in more cohesive and better managed communities, evidence also

indicates that better networked, or better educated, groups within a community may be

better able to organize and thus to benefit most from projects. There is virtually no

reliable evidence on community participation projects actually increasing a community's

capacity for collective action. This is clearly an area for further research.


        Several qualitative studies indicate that the sustainability of community-based

initiatives depends crucially on an enabling institutional environment. Line ministries

need to be responsive to the needs of communities, and national governments need to be

committed to transparent, accountable, and democratic governance, through upward

commitment. To avoid "supply driven demand driven development" it is important that

community leaders also be downwardly accountable, answerable primarily to

beneficiaries rather than to political and bureaucratic superiors.


        Qualitative evidence also suggests the importance of external agents, such as


                                                  56

project facilitators, to project success. Projects often work with young, inexperienced

facilitators whose incentives may not be aligned with the best interests of the community.

Knowledge of their impact on the success of projects is limited and requires more

investigation. This lack of evidence also relates to the question of how rapidly

participatory projects can be scaled up, because rapid scaling up may rely on especially

inexperienced facilitators.


        Overall, since the success of community-based development is crucially

conditioned by local cultural and social systems, projects are best done with careful

learning by doing. While successful projects in any context provide a tremendous

learning opportunity, any wholesale application of best practices in unlikely to be useful.

In a similar vein, key concepts that underpin community-based initiatives, such as

participation, community, and social capital, must be adequately detailed in a context-

specific manner. Case study evidence indicates that any naive application of these notions

by project implementers can lead to poor project design and to outcomes that are at odds

with the stated intentions of projects.


        Finally, it is important to realize that community-based development is not

necessarily "empowering" in practice. A less fervent, and more analytical, approach by

both proponents and opponents would be extremely beneficial. This requires a long time

horizon and programs that are well monitored, to enable learning from mistakes, and

carefully evaluated. Little is known about the impact of community-based projects,

largely because most such projects lack careful evaluations with good treatment and

control groups and with baseline and follow-up data. This situation urgently needs to be



                                                   57

remedied.




             58

Bibliography


Abraham, Anita, and Jean-Phillippe Platteau. 2004. "Participatory Development: When

      Culture Creeps." In V. Rao and M. Walton, eds., Culture and Public Action: A

      Cross-Disciplinary Dialogue on Development Policy. Palo Alto, Calif.: Stanford

      University Press.


Agarwal, A., and C. Gibson. 1999. "Enchantment and Disenchantment: The Role of

      Community in Natural Resource Conservation." World Development 27(4): 629­

      50.


Agarwal, Bina. 2001. "Participatory Exclusions, Community Forestry, and Gender: An

      Analysis for South Asia and a Conceptual Framework." World Development

      29(10): 1623­48.


Alderman, Harold. 2002. "Do Local Officials Know Something We Don't?

      Decentralization of Targeted Transfers in Albania." Journal of Public Economics

      83(3): 375­404.


Alesina, Alberto, and Eliana La Ferrara. 2000. "Participation in Heterogeneous

      Communities." Quarterly Journal of Economics 115(3): 847­904.


Appadurai, Arjun. 2004 "The Capacity to Aspire: Culture and the Terms of

      Recognition." In V. Rao and M. Walton, eds., Culture and Public Action: A

      Cross-Disciplinary Dialogue on Development Policy. Palo Alto, Calif.: Stanford

      University Press.


Baland, Jean Marie, and Jean-Philippe Platteau. 1998. "Wealth Inequality and Effeciency


                                               59

      in the Commons, Part II: The Regulated Case." Oxford Economic Papers 50(1):

      1­22.


------. 1999. "The Ambiguous Impact of Inequality on Local Resource Management."

      World Development 27(5): 773­88.


------. 2003. "Institutions and the Efficient Management of Environmental Resources."

      In K.-G. Mähler and J. R. Vincent, eds., Handbook of Environmental Economics.

      Vol. 1. Amsterdam: Elsevier North-Holland.


------. Forthcoming. "Collective Action and the Commons: The Role of Inequality." In

      J. M. Baland, P. Bardhan and S. Bowles, eds., Inequality, Cooperation and

      Environmental Sustainability. Princeton, N.J.: Princeton University Press.


Bardhan, Pranab. 2000. "Irrigation and Cooperation: An Empirical Analysis of 48

      Irrigation Communities in South India." Economic Development and Cultural

      Change 48(4): 847­65.


Bardhan, Pranab, and Jeff Dayton-Johnson. 2001. "Unequal Irrigators: Heterogeneity and

      Commons Management in Large-Scale Multivariate Research." University of

      California, Berkeley. Processed.


Bardhan, Pranab, and Dilip Mookherjee. 2000. "Capture and Governance at Local and

      National Levels." American Economic Review 90(2): 135­39.


Bebbington, Anthony, Scott Guggenheim, Elizabeth Olson, and Michael Woolcock.

      Forthcoming. "Exploring Social Capital Debates at the World Bank." Journal of

      Development Studies.



                                               60

Bergstrom, T. C., L. Blume, and H. Varian. 1986. "On the Private Provision of Public

       Goods." Journal of Public Economics 29 (1): 25­49.


Binswanger, Hans, and Swaminathan Aiyer. 2003. "Scaling Up Community Driven

       Development: Theoretical Underpinnings and Program Design Implications."

       World Bank, Washington, D.C. Processed.


Blomquist, W., and E. Ostrom. 1985. "Institutional Capacity and the Resolution of a

       Commons Dilemma." Policy Studies Review 5(2): 383­93.


Botchway, Karl. 2001. "Paradox of Empowerment: Reflections on a Case Study from

       Northern Ghana." World Development 29(1): 135­53.


Bourdieu, Pierre. 1984. Distinction: A Social Critique of the Judgement of Taste. Trans.

       Richard Nice. Cambridge, Mass.: Harvard University Press.


------. 1990. The Logic of Practice. Trans. Richard Nice. Palo Alto, Calif: Stanford

       University Press.


------. 1998. Practical Reason: On the Theory of Action. Trans. Randal Johnson. Palo

       Alto, Calif.: Stanford University Press.


Bowen, John R. 1986. "On the Political Construction of Tradition: Gotong Royong in

       Indonesia." Journal of Asian Studies 45(3): 545­61.


Calderon, Fernando, and Alicia Szmuckler. 2004. "Political Culture and Development."

       In V. Rao and M. Walton, eds., Culture and Public Action: A Cross-Disciplinary

       Dialogue on Development Policy. Palo Alto, Calif.: Stanford University Press.


Cernea, Michael, ed. 1985. Putting People First: Sociological Variables in Rural


                                               61

       Development. New York: Oxford University Press and World Bank.


Chambers, Robert. 1983. Rural Development: Putting the First Last. London: Longman.


Chambers, Robert. 1997. Whose Reality Counts? Putting the First Last. London:

       Intermediate Technology Publications


Chase, Robert S. 2002. "Supporting Communities in Transition: The Impact of the

       Armenian Social Investment Fund." World Bank Economic Review 16(2): 219­

       40.


Chase, Robert S., and Lynn Sherburne Benz. 2001. "Household Effects of Community

       Education and Health Initiatives: Evaluating the Impact of the Zambia Social

       Fund." World Bank, Washington, D.C. Processed.


Chwe, Michael Suk-Young. 1999. "Structure and Strategy in Collective Action."

       American Journal of Sociology 105(1): 128­56.


------. 2001. Rational Ritual: Culture, Coordination and Common Knowledge.

       Princeton, N.J.: Princeton University Press.


Cleaver, Frances. 1999. "Paradoxes of Participation: Questioning Participatory

       Approaches to Development." Journal of International Development 11(4): 597­

       612.


Coady, David. 2001. "An Evaluation of the Distributional Power of PROGRESA's Cash

       Transfers in Mexico." FCND Discussion Paper 117. International Food Policy

       Research Institute, Food Consumption and Nutrition Division, Washington, D.C.


Coirolo, Luis, Keith McLean, Mondonga Mokoli, Andrea Ryan, Melissa Williams. 2001.


                                               62

      "Community Based Rural Development: Reducing Rural Poverty from the

      Ground Up." World Bank Rural Strategy Discussion Paper 6. Washington, D.C.


Conning, Jonathan, and Michael Kevane. 2002. "Community Based Targeting

      Mechanisms for Social Safety Nets: A Critical Review." World Development

      30(3): 375­94


Cooke, Bill, and Uma Kothari. 2001. Participation: The New Tyranny? London: Zed

      Books.


Das Gupta, Monica, Helene Grandvoinnet, and Mattia Romani. 2000. "State-Community

      Synergies in Development: Laying the Basis for Collective Action," Policy

      Research Working Paper 2439. World Bank, Development Research Group,

      Washington, D.C.


Dayton-Johnson, Jeff. 2000. "Determinants of Collective Action on the Local Commons:

      A Model with Evidence from Mexico." Journal of Development Economics

      62(1): 181­208.


Dayton-Johnson, Jeff, and Pranab Bardhan. 2002. "Inequality and Conservation on the

      Local Commons: A Theoretical Exercise." Economic Journal 112(481): 577­602.


Demombynes, Gabriel, Chris Elbers, Jean O. Lanjouw, Peter Lanjouw, Johan Mistiaen,

      and Berk Özler. 2002. "Producing an Improved Geographic Profile of Poverty:

      Methodology and Evidence from Three Developing Countries." WIDER

      Discussion Paper 2002/39. United Nations University and World Institute for

      Development Economics Research, Helsinki.




                                             63

Demsetz, Harold. 1970. "The Private Production of Public Goods." Journal of Law and

        Economics 13(2): 293­306.


Dongier, Philip, Julie Van Domelen, Elinor Ostrom, Andrea Ryan, Wendy Wakeman,

        Anthony Bebbington, Sabina Alkire, Talib Esmail, and Margatet Polski. 2001.

        "Community Driven Development." In World Bank, PRSP Sourcebook. Vol. 1.

        Washington, D.C.


Durlauf, Steven N. 2001. "On the Empirics of Social Capital." University of Wisconsin­

        Madison, Department of Economics. Processed.


Elbers, Chris, Peter Lanjouw, Johan Mistiaen, Berk Özler, and Ken Simler. 2003. "Are

        Neighbors Equal? Estimating Local Inequality in Three Developing Countries."

        FCND Discussion Paper 147. International Food Policy Research Institute, Food

        Consumption and Nutrition Division, Washington, D.C.


Escobar, Arturo. 1995. Encountering Development: The Making and Unmaking of the

        Third World. Princeton, N.J.: Princeton University Press.


Fine, Ben. 2001. Social Capital versus Social Theory: Political Economy and Social

        Science at the Turn of the Millennium. London: Routledge.


Finsterbusch, Kurt, and Warren Van Wicklin III. 1989. "Beneficiary Participation in

        Development Projects: Empirical Tests of Popular Theories." Economic

        Development and Cultural Change 37(3): 573­93.


Freire, Paulo. 2000. Pedagogy of the Oppressed. Trans. Myra Bergman Ramos. New

        York: Herder and Herder; 1970. Reprint, New York: Continuum.



                                               64

Galasso, Emanuela, and Martin Ravallion. (forthcoming). "Decentralized Targeting of an

        Anti-Poverty Program." Journal of Public Economics, in press.


Gandhi, Mohandas Karamchand. 1962. Village Swaraj. Ahmedabad, India: Navjivan

        Press.


Grootaert, Christiaan. 1998. "Social Capital: The Missing Link?" SCI Working Paper 3.

        World Bank, Environmentally and Socially Sustainable Development Network,

        Washington, D.C.


Gugerty, Mary Kay, and Michael Kremer. 2000. "Outside Funding of Community

        Organizations: Benefiting or Displacing the Poor?" NBER Working Paper 7896.

        National Bureau for Economic Research, Cambridge, Mass.


Gugerty, Mary Kay, and Edward Miguel. 2000. "Community Participation and Social

        Sanctions in Kenyan Schools." Harvard University, Program of Political

        Economy and Government, Cambridge, Mass. Processed.


Guijit, Irene and Meera Kaul Shah. 1998. The Myth of Community: Gender Issues in

        Participatory Development, Colchester, Intermediate Technology Development

        Group Publishing.


Gusfield, Joseph R. 1975. Community: A Critical Response. Oxford: Basil Blackwell.


Hardin, Garrett. 1992. "Tragedy of the Commons." In Anil Markandya and Julie

        Richardson, eds., Environmental Economics: A Reader. New York: St. Martin's

        Press.


Hardin, Russell. 1982. Collective Action. Baltimore, Md.: Johns Hopkins University



                                               65

       Press.


Harragin, Simon. 2004. "Relief and an Understanding of Local Knowledge: The Case of

       Southern Sudan." In V. Rao and M. Walton, eds., Culture and Public Action: A

       Cross-Disciplinary Dialogue on Development Policy. Palo Alto, Calif.: Stanford

       University Press.


Harriss, John. 2001. Depoliticizing Development: The World Bank and Social Capital.

       New Delhi: LeftWord Books.


Heckathorn, D. 1993. "Collective Action and Group Heterogeneity: Voluntary Provision

       versus Selective Incentives." American Sociological Review 58(3): 329­50.


Hirschman, Albert O. 1970. Exit, Voice, and Loyalty: Responses to Decline in Firms,

       Organizations, and States. Cambridge, Mass.: Harvard University Press.


------. 1984. Getting Ahead Collectively: Grassroots Experiences in Latin America.

       New York: Pergamon Press.


Hoddinott, John, M. Adato, T. Besley, and L. Haddad. 2001. "Participation and Poverty

       Reduction: Issues, Theory, and New Evidence from South Africa." FCND

       Discussion Paper 98. International Food Policy Research Institute, Food

       Consumption and Nutrition Division, Washington, D.C.


Isham, Jonathan, and Satu Kähkönen. 1999a. "Institutional Determinants of the Impact of

       Community-Based Water Services: Evidence from Sri-Lanka and India."

       Working Paper 236. University of Maryland, Center for Institutional Reform and

       the Informal Sector, College Park, Md.



                                              66

------. 1999b. "What Determines the Effectiveness of Community-Based Water

        Projects? Evidence from Central Java, Indonesia, on Demand Responsiveness,

        Service Rules, and Social Capital." SCI Paper 14. World Bank, Environmentally

        and Socially Sustainable Development Network, Washington, D.C.


Isham, Jonathan, Deepa Narayan, and Lant Pritchett. "Does Participation Improve Project

        Performance: Establishing Causality with Subjective Data." World Bank

        Economic Review 9(2): 175­200.


Itaya, J., D. de Meza, and G. D. Myles. 1997. "In Praise of Inequality: Public Good

        Provision and Income Distribution." Economic Letters 57(3): 289­96.


Jackson, Cecile. 1997. "Sustainable Development at the Sharp End: Fieldworker Agency

        in a Participatory Project." Development in Practice 7(1): 237­47.


Jalan, Jyotsna, and Martin Ravallion. 2003. "Estimating the Benefit Incidence of an

        Antipoverty Program by Propensity-Score Matching." Journal of Business and

        Economic Statistics 21(1): 19­30.


Jimenez, Emmanuel, and Vicente Paqueo. 1996. "Do Local Contributions Affect the

        Efficiency of Public Primary Schools?" Economics of Education Review 15(4):

        377­86.


Jimenez, Emmanuel, and Yasuyuki Sawada. 1999. "Do Community-Managed Schools

        Work? An Evaluation of El Salvador's EDUCO Program." World Bank

        Economic Review 13(3); 415-441Katz, Travis, and Jennifer Sara. 1997. "Making

        Rural Water Supply Sustainable: Recommendations from a Global Study." Water




                                               67

       and Sanitation Program, Washington, D.C.


Khwaja, Asim Ijaz. 2001. "Can Good Projects Succeed in Bad Communities? Collective

       Action in the Himalayas." Harvard University, Department of Economics,

       Cambridge, Mass. Processed.


King, Elizabeth, and Berk Özler. 2000. "What's Decentralization Got to Do with

       Learning? Endogenous School Quality and Student Performance in Nicaragua."

       World Bank, Development Research Group, Washington, D.C. Processed.


Kleemeier, Elizabeth. 2000. "The Impact of Participation on Sustainability: An Analysis

       of the Malawi Rural Piped Scheme Program." World Development 28(5): 929­44.


Krishna, Anirudh. 2002. "Global Truths and Local Realities: Traditional Institutions in a

       Modern World." Duke University. Processed.


Krishna, Anirudh, Norman Uphoff, and Milton J. Esman, eds. 1997. Reasons for Hope:

       Instructive Experiences in Rural Development. West Hartford, Conn.: Kumarian

       Press.


Kurien, V. 1997, "The AMUL Dairy Cooperatives: Putting the Means of Development

       into the Hands of Small Producers in India," Chapter 7 in Reasons for Hope:

       Instructive Experiences in Rural Development. Anirudh Krishna, Norman Uphoff,

       and Milton J. Esman, eds. West Hartford, Conn.: Kumarian Press.


La Ferrara, Eliana. 2002. "Inequality and Group Participation: Theory and Evidence from

       Rural Tanzania." Journal of Public Economics 85(2): 235­73.


Matthew, George, and Nirmala Buch, eds. 2000. Status of Panchayati Raj in the States



                                              68

       and Union Territories of India 2000. New Delhi: Concept Publishing.


Michener, Victoria. 1998. "The Participatory Approach: Contradiction and Co-option in

       Burkina Faso." World Development 26(12): 2105­18.


Mompati, Tlamelo, and Gerard Prinsen. 2000. "Ethnicity and Participatory Development

       Methods in Botswana: Some Participants Are to Be Seen and Not Heard."

       Development in Practice 10(5): 625­37.


Morris, S., J. Hoddinott, J. M. Medina, and G. Begeron. 1999. "Will Rural Development

       Enhance Food Security: A Case Study of Western Honduras." International Food

       Policy Research Institute, Washington, D.C. Processed.


Mosse, David. 1997a. "Colonial and Contemporary Ideologies of Community

       Management: The Case of Tank Irrigation Development in South India." Modern

       Asian Studies 33(2): 303­38.


------. 1997b. "The Symbolic Making of a Common Property Resource: History,

       Ecology, and Locality in a Tank-Irrigated Landscape in South India."

       Development and Change 28(3): 467­504.


------. 2001. "`People's Knowledge,' Participation and Patronage: Operations and

       Representations in Rural Development." In Bill Cooke and Uma Kothari, eds.,

       Participation: The New Tyranny. London: Zed Books.


Narayan, Deepa. 1995. "The Contribution of People's Participation: Evidence from 121

       Rural Water Supply Projects." ESD Occasional Paper Series 1. World Bank,

       Washington, D.C.



                                               69

------. 1998. "Participatory Rural Development." In E. Lutz, H. Binswanger, P. Hazell,

       and A. McCalla, eds., Agriculture and the Environment. Washington D.C.: World

       Bank.


------, ed. 2002. Empowerment and Poverty Reduction: A Sourcebook. Washington,

       D.C.: World Bank.


Narayan, Deepa, and Katrinka Ebbe. 1997. "Design of Social Funds: Participation,

       Demand Orientation and Local Organizational Capacity." World Bank Discussion

       Paper 375. Washington, D.C.


Narayan, Deepa, and Patti Petesch, eds. 2002. Voices of the Poor: From Many Lands.

       New York: Oxford University Press.


Newman, John, Menno Pradhan, Laura Rawlings, Geert Ridder, Ramiro Coa, and Jose

       Luis Evia. [2002] "An Impact Evaluation of Education, Health and Water Supply

       Investments by the Bolivian Social Investment Fund." World Bank Economic

       Review 16(2): 241­74.


North, Douglass. 1990. Institutions, Institutional Change and Economic Performance.

       Cambridge, United Kingdom: Cambridge University Press.


Oliver, P., and G. Marwell. 1988. "The Paradox of Group Size in Collective Action: A

       Theory of the Critical Mass II." Amercian Sociological Review 94: 502­34.


Oliver, P., G. Marwell, and R. Teixeira. 1985. "A Theory of the Critical Mass I.

       Interdependence, Group Heterogeneity, and the Production of Collective Action."

       American Journal Of Sociology 91(3): 522­56.



                                                70

Olson, Mancur. 1973. The Logic of Collective Action: Public Goods and the Theory of

       Groups. Cambridge, Mass.: Harvard University Press.


Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for

       Collective Action. New York: Cambridge University Press.


Ostrom, E., W. Lam, and M. Lee. 1994. "The Performance of Self-Governing Irrigation

       Systems in Nepal." Human Systems Management 13(3): 197­207.


Paxson, Christina, and Norbert Schady. 2002. "The Allocation and Impact of Social

       Funds: Spending on School Infrastructure in Peru." The World Bank Economic

       Review 16(2): 297­319.


Phillips, Sue and Richard Edwards. 2000. "Development, Impact Assessment and the

       Praise Culture," Critique of Anthropology, 20 (1): 47-66


Portes, Alejandro. 1978. "Social Capital: Its Origins, and Applications in Modern

       Society." Annual Review of Sociology 98(6) 1320­50.


Pradhan, Menno, and Laura B. Rawlings. 2002. "The Impact and Targeting of Social

       Infrastructure Investments: Lessons from the Nicaraguan Social Fund." World

       Bank Economic Review 16(2): 275­95.


Pritchett, Lant. 2002. "It Pays to be Ignorant: A Simple Political Economy Model of

       Rigorous Program Evaluation." Harvard University, Kennedy School of

       Government, Cambridge, Mass. Processed.


Pritchett, Lant, and Michael Woolcock. 2002. "Solutions When the Solution in the

       Problem: Arraying the Disarray in Development." Working Paper 10. Center for



                                                71

      Global Development, Washington, D.C.


Putnam, Robert D. (with Robert Leonardi and Raffaella Y. Nanetti). 1993. Making

      Democracy Work: Civic Traditions in Modern Italy, Princeton, Princeton

      University Press


Rao, Vijayendra. 2001. "Celebrations as Social Investments: Festival Expenditures, Unit

      Price Variation and Social Status." Journal of Development Studies 38(1):71­97.


------. 2003. "Symbolic Public Goods and the Coordination of Collective Action: A

      Comparison of Community Based Development in Indonesia and India." World

      Bank, Development Research Group, Washington, D.C. Processed.


Rao, Vijayendra, and Ana Maria Ibanez. 2003. "The Social Impact of Social Funds in

      Jamaica: A Mixed-Methods Analysis of Participation, Targeting and Collective

      Action in Community Driven Development." World Bank Policy Research

      Working Paper 2970, World Bank, Development Research Group, Washington,

      D.C.


Rao, Vijayendra, and Michael Walton. 2004. "Culture and Public Action: An

      Introduction." In V. Rao and M. Walton, eds., Culture and Public Action: A

      Cross-Disciplinary Dialogue on Development Policy. Palo Alto, Calif.: Stanford

      University Press.


Rao, Vijayendra, and Michael Woolcock. 2003. "Integrating Qualitative and Quantitative

      Approaches in Program Evaluation." In François Bourguignon and Luiz Pereira

      Da Silva, eds., Evaluating the Poverty and Distributional Impact of Economic

      Policies: A General Introduction. Washington, D.C.: World Bank and Oxford

                                              72

        University Press.


Ravallion, Martin. 2000. "Monitoring Targeting Performance When Decentralized

        Allocations to the Poor are Unobserved." World Bank Economic Review 14(2):

        331­45.


Ravallion, M., and Q. Wodon. 1999. "Poor Areas or Only Poor People." Journal of

        Regional Science 39(4): 689­711.


Ribot, Jesse C. 1995. "From Exclusion to Participation: Turning Senegal's Forestry

        Policy Around?" World Development 23(9): 1587­99.


------. 1999. "Decentralization, Participation and Accountability in Sahelian Forestry:

        Legal Instruments of Political-Adminisitrative Control." Africa 69(1): 23­65.


Rose, Kalima. 1993. Where Women are Leaders: The SEWA Movement in India. London:

        Zed Books.


Schady, Norbert R. 2000. "Political Economy of Expenditures by the Peruvian Social

        Fund", The American Political Science Review, 94 (2) : 289-304


Scott, James. 1998. Seeing Like a State: How Certain Schemes to Improve the Human

        Condition Have Failed. New Haven, Conn.: Yale University Press.


Seabright, Paul. 1996. "Accountability and Decentralization in Government: An

        Incomplete Contracts Model." European Economic Review 40(1): 61­89.


Sen, Amartya K. 1985. Commodities and Capabilities. Amsterdam: Elsevier.


------. 1999. Development as Freedom. New York: Knopf Press.


Serrano-Berthet, Rodrigo. 1996. Who Knows What's Best for the Poor: Demand-Driven

                                               73

       Policies and Rural Poverty in Northeast Brazil. MIT/State of Ceará Project II.

       Master's Thesis. Massachusetts Institute of Technology, Department of Urban

       Studies and Planning, Cambridge, Mass.


Somanathan, E., R. Prabhakar, and Bhupendra Singh Mehta. 2002. "Collective Action for

       Forest Conservation: Does Heterogeneity Matter?" Discussion Paper in

       Economics 02-01. Indian Statistical Institute, Planning Unit, New Delhi.

       Processed.


Sullivan, John. 1992. Local Government and Community in Java: An Urban Case-Study.

       Singapore: Oxford University Press.


Summers, Lawrence. 2001. "Speech at World Bank Country Director's Retreat--2001,"

       Washington DC (date was not available)


Tendler, Judith. 2000. "Why are Social Funds So Popular?" In Shahid Yusuf, Weiping

       Wu, and Simon Evenett, eds., Local Dynamics in an Era of Globalization: 21st

       Century Catalysts for Development. [Washington-DC World Bank and Oxford

       University Press.


Tendler, Judith, and Rodrigo Serrano. 1999-Berthet. "The Rise of Social Funds: What

       Are They a Model of?" Massachusetts Institute of Technology, Department of

       Urban Studies and Planning, Cambridge, Mass.


Thomass-Slayter, Barbara P. 1994. "Structural Change, Power Politics, and Community

       Organizations in Africa--Challenging the Patterns, Puzzles, and Paradoxes."

       World Development 22(10): 1479­90.




                                              74

Trawick, Paul. 2001. "Successfully Governing the Commons: Principles of Social

       Organization in an Andean Irrigation System." Human Ecology 29(1): 1­25.


Uphoff, Norman. 1986. "Improving International Irrigation Management with Farmer

       Participation: Getting the Process Right." Boulder, Colo.: Westview Press.


Uphoff, Norman, and C. M. Wijayratna. 2000. "Demonstrated Benefits from Social

       Capital: The Productivity of Farmer Based Organizations in Gal Oya, Sri Lanka."

       World Development 28(11): 1875­90.


Uphoff, Norman, Milton J. Esman, and Anirudh Krishna. 1998. Reasons for Success:

       Learning from Instructive Experiences in Rural Development. Kumarian Press:

       West Harford, Conn.


Vasan, Sudha. 2002. "Ethnography of the Forest Guard: Contrasting Discourses,

       Conflicting Roles and Policy Implementation." Economic and Political Weekly

       [volume 39] October 5th

       http://www.epw.org.in/showArticles.php?root=2002&leaf=10&filename=5016&f

       iletype=html


Vedeld, Trond. 2000. "Village Politics: Heterogeneity, Leadership and Collective

       Action." Journal of Development Studies 36(5): 105­34.


Wade, R. 1987. "The Management of Common Property Resources: Finding a

       Cooperative Solution." World Bank Research Observer 2(3): 219­34


------. 1988. Village Republics: Economic Conditions for Collective Action in South

       India. Cambridge, United Kingdom: Cambridge University Press.



                                               75

White, Howard. 1999. "Politicising Development? The Role of Participation in the

       Activities of Aid Agencies." In Ken Gupta, ed., Foreign Aid: New Perspectives.

       Boston: Kluwer Academic Press.


White, Sarah. 1996. "Depoliticizing Development: The Uses and Abuses of

       Participation." Development in Practice 6(1): 6­15.


Woolcock, Michael. 1998. "Social Capital and Economic Development: Toward a

       Theoretical Synthesis and Policy Framework." Theory and Society 27(2): 151­

       208.


Woolcock, Michael, and Deepa Narayan. 2000. "Social Capital: Implications for

       Development Theory, Research and Policy." World Bank Research Observer

       15(2): 225­49.


World Bank. 2001. World Development Report 2000/2001: Attacking Poverty. New

       York: Oxford University Press.


World Bank. 2003. World Development Report 2004 Making Services Work for the Poor,

       Washington DC, World Bank and Oxford University Press


World Bank, Operations Evaluations Department. 2002. Social Funds: Assessing

       Effectiveness. Washington, D.C.


World Bank, Social Development Department. 2000. "New Paths to Social Development:

       Community and Global Networks in Action." Environmentally and Socially

       Sustainable Development Network, Washington, D.C.


Zyl, Johan Van, Loretta Sonn, and Alberto Costa. 2000. "Decentralized Rural



                                              76

Development, Enhanced Community Participation, and Local Government

Performance: Evidence from North-East Brazil." Processed.




                                    77