A

A good starting point for the analysis of such apparatus is the accountability framework presented in the World Development Report (WDR) 2004 [World Bank 2004], perhaps the most potent expression of the current poverty alleviation agenda. The report, concerned with making service provisions work for the poor, focuses on the delivery of those services that have a most direct link to human development: health, education, water, sanitation and electricity. It is thus highly relevant for this discussion for the following three reasons.

First, service delivery is a key component for the achievement of the MDGs. More specifically, it reflects the shift from the budget-cuts agenda of structural adjustment programmes to a more positive agenda, the incorporation into mainstream of a multidimensional definition of poverty, the stated pro-poor targeting of current development interventions. Effective service provision is a key element for the realisations of the stated development goals. Second, the delivery of services is defined as a public responsibility, whose onus is firmly set at the state level. Governments’ responsibility is justified on two different grounds [World Bank 2004: p 3]. On the one side, public services’ externalities2 imply that the private sector alone is not able to reach levels of provision that are adequate to society’s needs. Governments need to create a favourable institutional framework for such provisions, by financing, regulating and/or providing such services. On the other side, states’ responsibility for public provisions is framed by considering basic health and basic education as fundamental human rights. The “challenge”, thus, becomes one of identifying the appropriate mix between governments, the private sector, communities and outside partners, to meet such obligation. Third, following from this, policy recommendations for effective service delivery reflect the “institutional change” agenda promoted by the post-Washington consensus. The WDR 2004 is, in fact, a potent example of the comprehensive and integrated nature of such agenda: specific policy recommendations contained in the development agenda (namely, its institutional assumptions, stated objectives, recommended actions and their sites of interventions), in sum, its study broadens the discussion not only to its contradictory outcomes in context, but also to the whole development apparatus.

In the following sections, first, the main policy recommendations of the mainstream development agenda are introduced by looking at the World Bank’s Accountability Framework for service provision, contained in the WDR 2004; second, the main analytical categories employed by this framework are challenged by suggesting that its policies and practices redefine analytical boundaries of development-related processes. The argument made is that the current neoliberal agenda, despite being defined at the state level, tends to blur the boundaries between what is internal and external to the state; it redefines decision-making centres within the state; it establishes a complex architecture of governance,around which different forms of social organisation interact. An understanding of the current pro-poor development agenda as an integrated and “coherent” (if assumptions are accepted, that is) apparatus is offered and an analytical perspective for capturing power relations across it is tentatively sketched that focuses on institutional mobility.

The Framework

The problem that the WDR 2004 sets to resolve is that service delivery does not reach the poor. Several examples are cited in support of such consideration; they reflect the report’s concern with access, quantity and quality of services. Four problems are specifically opened up through such examples: lack of propoor resource allocation (e g, a third of developing countries spending is on health and education but is enjoyed primarily by non-poor people); efficient delivery mechanisms (e g, even increasing such allocations, the money does not reach the front line provider); incentives, monitoring and regulation of front line providers (e g, even increasing the share allocated to front line providers “teachers

Economic and Political Weekly July 28, 2007

Diagram: The Accountability Framework

Policymakers Poor people Providers

Source: WDR 2004, p 6.

must be present and effective, as much as nurses and doctors must provide care that patients need”); finally, voice, participation and incentives of clients (e g, even if all the above conditions are met, there might be a lack of demand, due to poor quality of service, distance, or “cultural” sensitivities, for instance in relation to gender). The accountability framework offers an integrated solution to solve these problems.

The WDR 2004 frames the notion of accountability in terms of responsiveness of providers, as well as the ability of recipients of services to monitor and discipline them. Increased voice and participation of beneficiaries, as well as policymakers’ ability to reward and penalise effective/ineffective service providers, are suggested mechanisms for the strengthening of accountability along three key dimensions. The framework is depicted in the diagram.

This diagram is meant to represent the service delivery chain, unbundled into actors and relationships.4 In a nutshell, its main tenets can be summarised by distinguishing its normative contents in relation to actors and relationships. Poor people are the target group of interventions and they are identified by/through indicators, household surveys, and other “participatory” devices. Policymakers are defined in the report as “politicians” and, more in general, as “those who have the responsibility of formulating collective objectives”. Providers are those organisations in charge of the actual delivery of services, be they central or local governments, NGOs, community-based organisations, etc. Two routes of accountability are identified in the report.

As citizens, poor people should, in principle, be able to affect the formulation of collective objectives, though in practice, they might be excluded from decisionmaking processes and by a weak electoral system, by non-responsiveness of policymakers, by a lack of “voice”, or configurations of power that prevent the poor from accessing decision-making institutions. The key prescription embedded in this relation is to increase the accountability of policymakers to its poor citizens, for example, through strengthening the electoral process, democratic decentralisation, by increasing the relevance and role of civil society organisations, and consequently the voice of the poor, etc.

Even if policymakers are responsive, or out of an Olsonian encompassing interest5 they care or want to provide services to their poor citizens, they might lack the capacity to deliver: they cannot offer incentives, nor be able to monitor, or the regulatory framework for the provision of services may be weak. Following and expanding recommendations of the WDR 1997, policy-recommendations cover a vast array of reforms, inspired by new public management principles [Minogue et al 1998] and covering anti-corruption mechanisms, civil sector reforms, etc. Key prescriptions include the separation between policymakers and providers and the conceptualisation of their relation as a compact, or perhaps as a contract, which would facilitate the capacity of governments to manage incentives, while letting the market unfold its potential. These two relations constitute the long route of accountability.

There is also a short route, however. This route is fundamental, given the difficult implementation of reforms promoted through the long route. The key prescription here relates to increasing the power of poor people, clients in this case, over providers. In fact, not only the voice of poor people as citizens should be strengthened in relation to policymakers, but also in their capacity as students in schools, patients in hospitals, or consumers of water, poor people’s voice and participation should be strengthened in relation to those providing public services. Clients can play two roles in strengthening service delivery. On the one side, they can help tailor the service to their needs (ibid lack of demand); on the other, being at the point of delivery, they can function as effective monitors of the quality of such provisions (ibid quality). Strengthening their roles as clients, in other words, positively influences the supply of services by making providers responsive to specific demands and subject to market discipline. The fundamental aspect of this relation is thus the increase in clients’ participation and “choice” (read market competition).

The report provides a good overview of the agenda because it embodies and reinforces the game-theoretical, incentive-based nature of neo-institutionalism and would deserve a more thorough discussion. I will only focus on the first leg of the long route of accountability, i e, the relation between the “poor” and “policymakers”, in order to contextualise such neat and self-contained framework into the chaotic reality where it actually takes place.

Who Are the Actors?

The report calls “poor people” alternatively clients and citizens. They are citizens vis-a-vis the long route of accountability, i e, in their relation to the state; they are clients in their relation to providers. They are more specifically “poor citizens” and “poor clients” and thus the target population of development interventions. But how are the poor defined? As mentioned earlier, poor people are identified primarily through surveys, censuses and participatory appraisals, etc. This seems problematic from three perspectives.

First, in relation to the methodology employed, participatory poverty appraisals (PPAs) provide the basis for so-called “bottom-up” approaches to development, but are subject to heated controversies. In nuce, the main criticism that can be levelled against PPAs is their reliance on populist assumptions about “local knowledge”, which reify “local culture” and affirm the agendas of elites and powerful groups in supposedly homogeneous and consensual communities. Rather than revealing and contesting power relations, participation seems, in fact, shaped by them [Cooke and Khotari 2001; Mosse 2005].6

Second, even accepting the potential benefits stemming from such techniques, PPAs embody a residual notion of poverty, a notion that defines poverty as a series of lacks – of education, health provisions, income, etc, Kandiyoti (2000) makes an extremely potent critique in this respect, arguing that surveys do not capture the quite complex arrangements behind a household’s allocations of roles and resources in relation to production activities and domestic chores. In the context of post-Soviet central Asia she questions the boundaries of what are defined as households in surveys and administrative registers, i e, the very basic unit of targeted intervention, as much as the ambiguities of concepts such as employment, access to land, income and expenditure, if set in specific contexts and assessed through the prism of effective social

Economic and Political Weekly July 28, 2007 relations. Such critique – on its own – undermines the whole poverty reduction apparatus, because it questions the very basis upon which the poverty alleviation agenda is constructed. The relational nature of development processes, whereby different pre-existing social formations (be they households, tribes or “ethnic groups”) interact with, shape and are being shaped by development-related forms of knowledge and power, undermines the constitutive element of poverty interventions – the “poor”– and remains an unsolved issue of mainstream development.

A third point can be made in relation to “the poor”. Not only in this framework are the poor both poor citizens and poor clients, but also, depending on the mandate and orientation of different development agencies the same group of people can be seen and defined as “poor children” “poor women” “poor migrants” or even “poor souls”, depending on the institutional mandate, or specific concern, of each INGO, NGO or faith-based group concerned with them (e g, UNICEF, the Global Fund for Women, Concern International, Action by Churches Together, respectively). Following the techno-scientific paradigm of poverty-assessment indicators,7 these differences might not matter so much. Effectively however, each definition implicitly brings about a set of concerns, specific best practices, and a whole series of academic case studies based on such concerns and practices. Each definition informs different agencies’ specific understandings of beneficiaries, and thus of appropriate interventions; and is not neutral.

The definition of policymakers can also be problematised. This is important because it indicates who is “accountable” to whom, as different definitions of policymakers imply different institutional locations and power relations. The report defines policymakers as “politicians” or “those who have the responsibility of formulating collective interests”. Once again, it is possible to question who is effectively covered by this definition from various angles.

In relation to domestic social relations, a report by UNDP (2002) focuses on the concept of elite capture. The report argues that in many developing countries, despite formal democratic procedures being in place, small elites are able to influence if not outright occupy key decision-making institutions, and thus are able to influence the definition of the collective interest.

They can be propertied classes, the military, religious elites, etc. Through policymakers (a subset of national elites), different elite factions are thus able to shape, at different levels, the content and outer boundaries of policies and of their implementation. Clearly, each country has a particular historical trajectory and needs to be studied in detail,8 though it is possible to generalise processes of negotiation involved in the definition of “the public interest” through the notion of political settlement.

Khan (1997) assesses new institutional economics’ approaches to explaining state failure and defines “political settlement” as the balance of power of classes or groups subject to a particular institution. He explains both the possibilities of transitions to new institutions, as much as their performance, on the basis of such balance of power. The notion of political settlement radically challenges the technocratic and depoliticised assumptions behind development interventions, and is a second issue that seems to undermine the whole apparatus. However, Khan’s analysis is concerned with the balance of power within the state and it is not only “domestic” social relations that affect the definition of the public interest.

In the case of service provision, for example, donors seem an overriding force. Harrison (2001) assesses the postconditionality regime of the current neoliberal agenda arguing that it blurs the borders between internal/external interests.

He looks at reforms implemented in Tanzania and Uganda and attempts to demonstrate that the relation between governments and donors is one of mutual dependence and of mutual assimilation, making it more difficult to distinguish what is state-based or nationally-owned and what is donor-induced. Cammack (2004), similarly, calls development a “total process” and argues that the World Bank and its poverty alleviation agenda has consistently sought to embed disciplines of capitalist competition not only in its official discourse but also into mechanisms that bring about institutional and social transformation in developing countries. The point made is that the relation between donors, governments and providers is increasingly one of co-penetration or osmosis.

Furthermore, it is possible to extend the number of “external actors and institutions” defining the “internal” collective interest to other domains. On the one side, the state is not the only actor in this redefined agenda: NGOs and various providers are too. As such, poverty alleviation provides a context for the transformation of the developmental state along neoliberal lines and for the deconcentration of its core functions, now redistributed to non-state entities. On the other, international treaties such as TRIPs, and institutions such as the WTO seem to define specific boundaries within which development processes, such as service provisions for health, may or may not operate.

Escaping Accountability

The WDR 2004 frames the relation between the poor and policymakers in terms of accountability: to improve the responsiveness of policymakers to their “poor citizens”, not only their voice needs to be made “louder”, but also their power to hold accountable governments need to be strengthened. “Poor people need to have the legal, economic and political means” [World Bank 2004: 8] for an effective accountability. Which institutional mechanisms are devised to solve this problem?

The report itself focuses on the role of the media, of CSOs and of community organisations. Set in the context of the broader development agenda, this relation involves democratisation and decentralisation interventions, which are the key institutional mechanisms devised to strengthen the relationship between the two “boxes” under consideration. Once again: the state needs to get “closer to the people” to increase responsiveness; people’s voices need to be made louder, to strengthen their roles as clients and citizens; a proper institutional framework should be established to increase incentives for all parties involved. This agenda is very potent because it uses very few concepts which can abstractly be shared by a variety of people, and applies them across the board. Refrains, if repeated ad infinitum, become “common sense” and common sense is the first dimension that seems to escape any form of accountability.

Furthermore, even accepting the desirability of getting the state closer to the people, such reforms do not necessarily result in improved accountability. The specific institutional configuration resulting from decentralisation processes is not uniform across countries, but seems to be dependent on the particular political settlements (in its expanded connotation, see above) shaping the effective outcomes of these processes. In other words, the concrete institutional architecture resulting from “decentralisation policies”, for instance vis-a-vis administrative levels (e g, provinces/states and municipalities) or vis-a-vis powers associated to each (e g, taxes, policing, or appointments), once again, opens up opportunities for different social groups to shape the outer boundaries of such architecture [Slater 2003 for the case of Bolivia]. On the other, even focusing on a predetermined “local” administrative unit, the relation between “class” and the “local state”, as well as the (limited) possibilities that the state might serve interests other than the ones of the dominant classes, is dependent on the “way the overall political control of the state administration interacts with the different positions of social actors within and outside the local state apparatus”, as suggested by Jeffrey and Lerche (2000, p 875) in the case of Uttar Pradesh. From both angles, in other words, it seems that institutionalised accountability mechanisms are shaped, both in their form and content, by the a priori interaction between different social formations; an interaction which seems to escape accountability.

Furthermore mechanisms to increase the voice of the poor are not exclusively based on citizenship rights and democracy. Another institutional mechanism not explicitly mentioned in the report, but which is very relevant for service provision is PRSP.9 PRSPs are based on processes of consultation whose aim is to achieve a propoor and effective allocation of resources. They are based on similar concerns and deploy a similar rationality, but through different institutional mechanisms. PRSPs in other words, create a different type of relation between the same poor citizens and policymakers: they bypass parliamentary consultation, they reinforce the tension between internal and external constituencies (in this case the “political” voice of the poor vs the technocratic legitimation conferred by international financial institutions), they superimpose another layer of institutions which are, in principle, geared towards similar objectives [Craig and Porter 2003], thus multiplying the sites where accountability for services can be demanded.

There is more. If one were to consider health and education as human rights, the same poor citizens, who in this case would actually be “poor humans”, would see their collective interests defined in universal terms and would be encompassed by other types of institutions and organisations (human rights NGOs, national human rights commissions, rights-based donor-funded projects, etc). As discussed in the previous section, each would define the problem in its own way and thus “the intervention” from that perspective. Other regimes could be included, but the point is that the political settlement defining the collective interest, be it related to service provisions, recognition of specific instances, claims to protection, or anything else, seems to be negotiated in multiple institutional sites. Each regime embodies different forms of knowledge, defines different types of actors, provides different forms of legitimacy, and confers different types of authority. They all shape a complex “propoor” institutional agenda.

Disentangling different institutional regimes serves two purposes, both of which can be related to key agents’ (lack of) accountability. First, considering each regime on its own, key aspects of decisionmaking processes seem to escape any form of accountability, yet they have profound consequences on the way in which accountability demands can be put forward. For example, accepting the perspective employed by the WDR 2004, assumptions framing pro-poor policy- recommendations escape any accountability, yet they concretely shape states’ institutional apparatus, constraining or enhancing the capacity of different social groups to influence their outcomes, by opening up or closing down different institutional locations where struggles may or may not take place.

Second, considering the agenda as an integrated whole, different forms of power can be identified. The fragmentation of the current pro-poor institutional architecture in fact, reinforces the power of those actors who are able to operate on different levels and scales, across discourses and institutions, and drawing their power from different social domains. International financial institutions and their constituencies are a clear example. The World Bank, for instance, fashions itself as the “knowledge bank”, asserting a clear claim over the boundaries of the discourse it seeks to intervene upon. It influences the form and content of state reforms, be they associated to the civil sector, to financial transparency and anti-corruption, to electoral reforms and democratisation processes, or poverty alleviation. It defines the role civil society organisations should play within statesociety relations. Most importantly, it is

Economic and Political Weekly July 28, 2007 able to back up its ideological claims with considerable material disbursements. In other words, it establishes clear institutional boundaries, yet, it is able to transgress them, operating across all of these domains. Similarly, donors, and those social groups who are able to shape their policies operate bilaterally, fund international financial institutions, international and national civil society organisations, think tanks that further reinforce certain ideological constructs, and INGOs. The latter, in turn, operate both in the context of western societies, sending positive images of charitable development interventions to their household donors and organising campaigns that shape public opinions, as well as in developing societies, through partnerships with local NGOs or grassroots organisations. They also concur to the determination of the international policy agenda through advocacy and lobbying. Furthermore, once key institutional reforms are instantiated, similar considerations can be levelled to other social groups. Powerful domestic constituencies are able to shape the form and content of particular institutional reforms, and to assert their voice to shape their outcomes. In the context of “local” communities, landlords and other types of elites are able to influence participatory processes, resisting or reappropriating its desired outcomes. So on and so forth.

The capability of certain actors to cut across institutional and geographical spaces, as well as social domains: their power seems to stem from their ability to simultaneously assert their influence upon a variety of sites of political struggle, transgressing the boundaries they themselves seek to define and their institutional mobility seems to escape any form of accountability.

The World Bank’s accountability framework is a conceptual tool that, in its stated aims, should be deployed for the improvement of public services’ quality and quantity as well as for increased targeting of, and access to them by, the poor. It is a universal policy recommendation, replicated to suit the specific needs of various countries through national reports, that embodies the core elements of the current neoliberal development agenda and it is very potent because of its internal coherence and intersection with different aspects of such agenda. This article has sought to problematise this framework from a variety of perspectives.

First, the very basis upon which such agenda is formulated, i e, the definition of “poor people” was challenged. Second, accepting such definition, the notion of political settlement beyond the level of the state was expanded and took into consideration supra- infra- and trans-national actors and regimes. This discussion suggested how the complex and fragmented architecture of governance apparently imposed by powerful development actors is actually negotiated and resisted by a variety of organisations and social groups, who interact around the nodes and intersections of this architecture.

In relation to the framework’s stated objective, above considerations undermine the idea that it necessarily improves the states’ accountability to those who are supposed to benefit most from it, the poor. On the contrary, the above notes point to

Economic and Political Weekly July 28, 2007

the command of institutional and geographical spaces enjoyed by key development actors, which escape any form of accountability. They also point to powerful constituencies within each country that actively shape the form and content of such institutional set up. The state remains a pivotal institution, but the effective sites where political settlements are negotiated cannot be assumed a priori, and need to be studied and explained in their specific legal, economic and social configurations. At the same time, practices of transgression and resistance by a variety of social formations able to engage and shape outcomes of such processes highlight how subjectivities are also overlapping and blurred, with forms of social organisation being dependent both on historical trajectories and specific forms of mobilisation. They also cannot be assumed a priori but need to be studied and explained.

The dynamicity and relational nature of the development process seems to suggest the need for analyses that are both focused on such processes in specific and contingent institutional sites, but that, at the same time, are able to capture the simultaneity of other institutional sites co-determining contingent interactions. Itis only by assessing forms of power that are able to transcend institutional and geographical boundaries that such power can be captured, and its limitations comprehended.

EPW

Email: pn4@soas.ac.uk

Notes

1 MDGs encompass issues such as eradicating extreme poverty and hunger, achieving universal primary education, reduce child mortality, improve maternal health, etc. MDGs thus tackle “poverty” in its multidimensional conceptualisation, as promoted by UNDP’s Human Development Index, or, more recently, by the World Bank’s incorporation of Robert Chambers approach [WDR 2000-01].

2 When a market outcome affects parties other than the buyers and sellers in the market, sideeffects are created, called externalities. Externalities cause markets to be inefficient, and thus fail to maximise total surplus.

3 In particular, WDR 2004 applies the WDR 2000 theme of empowerment to the WDR 1990 emphasis on health and education; it extends the WDR 1993 on health – which prescribed a cost-effectiveness analysis to improve outcomes for poor people – by examining the institutional factors that may give rise to the correct institutional solutions. Furthermore, it complements the WDR 1994 on infrastructure – which focused on private sector participation – by addressing the politics of infrastructure provision for poor people. It expands one aspect of the WDR 1997 on the role of the state – state’s responsibility for service provision – and it builds on two recent WDRs – on institutions for market and institutions for collective action – by identifying incentives for reforming institutions to make service work for poor people.

4 It is interesting to note how these basic modelsare then replicated in, and tailored to, different country contexts. For example, the 2006 reportIndia: Inclusive Growth and Service Delivery

– Building on India’s Success presents a similar framework, and provides specific examples of intervention in that context – see http://siteresources.worldbank.org/SOUTH ASIAEXT/Resources/DPR_FullReport.pdf

5 Olson (1997) provides a “neo-institutional metaphor” populated by “rational” selfinterested bandits to develop an argument in favour of particular institutional arrangements.In brief, by comparing the (assumed) rational interests of individual – or mafia-type criminals and those of roving and stationary bandits, Olson argues that states’ leaders (i e, stationary bandits) can develop encompassing interests which would “rationally” lead them to the provision of public services. Through institutions such as property rights and democracy, in fact, the “bandit’s” narrow interests become encompassing and he would gain from using resources for the provision of public goods, so long as every dollar spent on such provision more than proportionally increases the income of “his” domain.

6 I am indebted to Giuseppe Caruso, my friend and colleague, for the wording of this paragraph.

7 The expression “techno-scientific” paradigm refers to the reliance on technical and scientific forms of knowledge to address what seem tobe essentially political issues. For more thorough discussions, and different approaches to this issue, see Craig and Porter (2003), Nanda (1999), Banuri (1990), Rahnema and Bawtree (1997).

8 Chatterjee (1998), Bose (1997) and Cooper(2002), or, from very different analytical premises, Huntingdon (1993), provide possible lines of enquiry in this respect.

9 PRSPs are the pinnacle of the new pro-poor development strategy. They are meant to be a pro-poor development plan, which should guide all government expenditure/budgets, informed by and agreed with the international development community, and arrived at in consultation with civil society. PRSPs are based on six core principles. They are meant to be: results-oriented, with targets for poverty reduction that are tangible and monitorable; comprehensive, integrating macroeconomic, structural, sectoral and social elements; countrydriven, representing a consensual view of what actions should be taken; participatory, with all relevant stakeholders participating in formulation and implementation; based on partnerships between government and other actors; long-term, focusing on reforming institutions and building capacity, as well as short-term goals. They replicate, in other words, the same core principles justifying the accountability framework discussed here.

References

Banuri, T (1990): ‘Development and the Politics of Knowledge: A Critical Interpretation of the Social Role of Modernisation Theories in the Development of the Third World’ in F A andS A Marglin (eds), Dominating Knowledge:Development, Culture and Resistance, Clarendon Press.

Bose, S (1997): ‘Instruments and Idioms of Colonial and National Development’ in Cooper and Packard (eds), International Development and the Social Sciences, University of California Press, Berkeley, pp 45-63.

Cammack, P (2004): ‘What the World Bank Means by Poverty Reduction, and Why It Matters’,New Political Economy, 9(2), pp 189-211.

Chatterjee, P (1998): ‘Development Planning and the Indian State’ in T J Byres, The State, Development Planning and Liberalisation in India, OUP, New Delhi, pp 82-103.

Cooke, B and U Kothari (2001): The Tyranny ofParticipation, Zed Books, London.

Cooper, F (2002): ‘The Recurrent Crises of the Gatekeeper State’ in Cooper F, Africa since 1940: The Past of the Present, CUP, Cambridge, pp156-90.

Craig, D and D Porter (2003): ‘Poverty Reduction Strategy Papers: A New Convergence’, World Development, (31)1, pp 53-69.

Harrison, G (2001): ‘Post-Conditionality Politics and Administrative Reform: Reflections on the Cases of Uganda and Tanzania’, Development and Change, 32, pp 657-79.

Huntingdon, S (1993): ‘Democracy’s Third Wave’ in Diamond and Plattner (eds), The Global Resurgence of Democracy, John Hopkins University Press, Baltimore.

Jeffrey, C, J Lerche (2000): ‘Stating the Difference: State, Discourse and Class Reproduction in Uttar Pradesh, India’, Development and Change, 31, pp 857-78.

Kandiyoti, D (2000): ‘Poverty in Transition: An Ethnographic Critique of Household Surveys in Central Asia’ in S Razavi (ed), Gendered Poverty and Well-Being, Blackwell, Oxford, pp 89-113.

Khan, M (1997): ‘State Failure in Weak States: A Critique of New Institutionalist Explanations’ in Harriss et al (eds), The New Institutional Economics and Third World Development, Routledge, London.

Minogue, M, C Polidano and D Hulme (eds), (1998): Beyond the New Public Management: Changing Ideas and Practices in Governance, Edward Elgar, London.

Mosse, D (2005): Cultivating Development: An Ethnography of Aid Policy and Practice, Pluto Press, London.

Nanda, M (1999): ‘Who Needs Post-Development? Discourses of Difference, Green Revolution and Agrarian Populism in India’, Journal of Developing Societies, 15(1), pp 5-27.

Olson, M (1997): ‘The New Institutional Economics: The Collective Choice Approach to Economic Development’ in Clague (ed), Institutions and Economic Development:Growth and Governance in Less Developed and Post-Socialist Countries, The Johns Hopkins University Press, Baltimore.

Rahnema, M, V Bawtree (1997): The Post-Development Reader, Zed Books, London.

Slater, D (2003): ‘On the Spatial Dynamics of Democratic Politics: Analysing the Bolivian Case’, Development and Change, (34)4, pp 607-32.

UNDP (2002): ‘Deepening Democracy in a Fragmented World’, Human DevelopmentReport, OUP, Oxford.

World Bank (2000): ‘Attacking Poverty’, World Development Report 2000, World Bank, Washington.

– (2004): ‘Making Services Work for Poor People’,World Development Report, World Bank, Washington.

Economic and Political Weekly July 28, 2007