Since the 1980s the IFIs have been lending money to African countries on the basis that they accept ‘conditions’ – changes in their economic and social policies that the IFIs think will help them pay back the loan. Critics protested that the process undermined democracy and sovereignty by imposing one-size-fits-all solutions in aid dependent African countries. In response to these criticisms, the IFIs announced in 1999 that these Structural Adjustment Programmes (SAPs) would be replaced by Poverty Reduction Strategy Papers (PRSPs). Under the new framework the IFIs trumpet their enthusiasm for ‘country ownership’. PRSPs, they argue, build ownership because rather than the IFIs imposing policies, countries are supposed to write their own three year national development plans, which detail macroeconomic policies, government spending targets and also social development programmes. The IFIs then judge whether these PRSPs are an acceptable basis for writing off debts or making new loans. They hope governments will be happier to implement plans they have written themselves. Furthermore, to meet the charge that imposing conditions is undemocratic, the IFIs now insists that other stakeholders, such as NGOs, churches, unions and business, rather than just government, are involved in writing the plans. This is called ‘participation’.
PRSPs thus in theory reduce the policy conditions that the IFIs attach to loans and debt relief, but define new process conditions. Having been instrumental in lobbying for these innovations, development think-tanks, Northern NGOs and bilateral donors initially welcomed PRSPs as a means to reverse power relationships between donors and governments, between citizens and states and between North and South. By February 2005, PRSPs were being implemented in 45 countries.
There is now a burgeoning literature assessing the quality of the participatory processes surrounding PRSPs and the impact of the new dispensation on policy conditions. Within this literature four broad characterisations of the changes brought by PRSPs compete. First, anti-globalisation critics charge that participation is a sham and that the shift to PRSPs can be understood as a public relations exercise that has had little impact on the fundamental ‘dependency’ relationship in which the IFIs continue to dictate policy to African governments (Bendaña, 2001). A second view, held by some IFI staff, donors and researchers, is that the transfer of responsibility for initiating policies is more substantive than critics allow. They suggest that limited change in the policies contained in the new national development plans reveals not that the IFIs maintain control over outcomes, but that recipient governments have learnt, through the sometimes difficult experience of structural adjustment, that international economic realities offer few choices but to embrace market-oriented reforms (Harrison, 2001). A third position, held by most NGOs and donors, as well as the World Bank's own Operations Evaluation Department (OED, 2004), argues that participation holds the (as yet unrealised) potential to transform relations between donors and recipients. Maxwell (2005) suggests that the principles of the PRSP sit at the heart of a new meta-narrative for best practice in aid relationships, displacing the Washington Consensus. The debate over how much progress has been made is thus largely informed by case-studies of PRSP processes that focus on process qua process, asking whether participatory planning techniques are being implemented according to their true principles (Piron and Evans, 2004; AFRODAD, 2003; Richmond and Ladd, 2001; Booth, 2001; McGee, Levene and Hughes, 2002). The final position contradicts the view that participation results in gradual empowerment by pointing to the limitations of African ownership resulting not from the way participation is being implemented, but from the underlying principles of participation itself. Brown (2004) and Craig and Porter (2003) argue that participation and ownership can be understood as ‘technologies of control’ since they mystify power relations, depoliticise negotiations, and thus secure IFI control of outcomes. Brown worries that
the discretionary element inherent in this model of participation seems likely to create a markedly quiescent form of representation. On the one hand, those who continue to be excluded under discretionary arrangements have no legal right to demand representation. On the other, those who are included will be put under pressure to accept whatever they are offered, on the grounds that, as supplicants, they could have received much less (Brown, 2004:244).
Second, this article draws attention to potential means by which participation disciplines individual participants. This involves affecting changes in individual and group consciousness and identity. This suggests that, in seeking to understand the PRSP process, researchers need to focus not only on the ability of participants to shape PRSP policy outcomes, but also the ability of the process to re-shape those participants and through them, the social and political constituencies they represent. Third, this article suggests a range of actors and interests involved in designing and wielding the social technology of control. Brown, and Craig and Porter present the PRSP as an IFI policy, to which other actors adapt. On one level this is self-evident. However, I make the argument that the PRSP is also in some senses a joint project of the IFIs, bilateral donors and Northern NGOs, all of whom collaborated to design the process, all of whom expect their interests to be served by it, and all of whom understand that the process cannot ‘perform’ without the active engagement of all the others. They support and engage with PRSPs on a collaborative basis.
In seeking to develop this argument, this article is structured as follows: section one discusses how the IFIs have come to see participation as a potential solution to a crisis of adjustment and outlines how participation came to prominence in the PRSP. It then presents critiques from within international relations and development studies which challenge the liberatory rhetoric of participation and civil society, and describe how they might instead operate as ‘social technologies of control’. Section two analyses a sample of aid industry documents, in order to assess the possibility that PRSPs represent a concrete example of such a social technology of control. The argument is made that actually existing African political identities, institutions and representatives are excluded by the ‘small print’ of the liberal constructions and assumptions that underpin PRSPs. I therefore conclude that, despite the rhetoric of ownership, PRSPs provide a context within which an uneasy coalition of the IFIs, donors and Northern NGOs are seeking to remake African states and societies in their own self-image, thus securing a stable consensus for liberal systems of economic and political management, and legitimating the increasingly intrusive supervision of African political communities by Northern actors. I make no claim that this whole package is a self-conscious project for all of the actors in this coalition, nor that it succeeds on its own terms in transforming African political communities. However, I do assume that such concentrated and intimate interventions will have both intended and unintended impacts on African political economies. Understanding how these play out will imply a quite different research agenda from the majority of existing case-studies on PRSP processes.
Structural adjustment, democracy & the discovery of participation
The IFIs responded to the debt crisis of the 1980s by adopting a leadership role in a creditors' cartel, using their financial leverage to intensify the supervision of African economies. Initially, the IFIs sought simply to ensure debt repayments, but rapidly became more ambitious, attaching increasingly detailed conditions to new loans and debt relief, despite the failure to reverse economic fortunes (UNCTAD, 2001). Throughout the adjustment era the IFIs struggled to reconcile political management of this economic failure with their major shareholders' stated commitment to liberal political systems. Squaring this circle has required a barrage of technical solutions that seemingly conform to their ‘non-political’ mandate (Veltmeyer et al. 1996) whereas, in practice, a strong anti-democratic impetus runs through all of them, up to and including the latest ‘participatory’ turn.
Two themes can be traced throughout the period. First, World Bank thinkers have been disappointed by the failure of ‘reform coalitions’ – potential future beneficiaries of adjustment that might provide political support for the policies – to emerge. Given their non-political mandate they saw little they could do to bring such coalitions into being. Second, they have been concerned to suppress popular aspirations for rapid development gains.
By the end of the 1990s, Joan Nelson noted that the repeated failure to achieve IFI growth projections generated,
a corrosive mixture of public cynicism, defeatism and anger – precisely the opposite of the gradual emergence of economic confidence and political trust … crucial to economic stabilisation and … the consolidation of democracy (1989:20).
The emergence of participation
The emergence of participation as an essential element of the PRSP in 1999, and thus as an organising principle of Bank interventions in Africa, can be understood as an attempts to overcome these crises. To understand how participation came to be imagined as the latest one-shot fix, and worked into policy, we need to look briefly at the idea of participation and at the process by which it came to prominence in Bank thinking.
Participation has its roots in Latin American ‘conscientisation’ literature, but its key advocate in the current context is Robert Chambers (1983) who has energetically promoted ‘Participatory Rural Appraisal’ (PRA) as the solution to the failures of ‘top-down’ development projects. Through PRA, participants, including the trained development professional ‘facilitating’ the process, experience a transformation in their own consciousness of the conditions and causes of underdevelopment. Throughout the 1980s, NGOs experimented extensively with PRA in their own programmes and became convinced that it offered an improved method of development implementation, increasing local populations' investment in projects brought by outsiders.
Four overlapping processes saw a gradual increase in the amount and depth of participatory policy and practice within the Bank. First, from the late 1980s, facing criticism from NGOs and within the US political system on issues including the environment, indigenous people's rights and dams, the Bank introduced participation as a safety check in these particular project areas. Second, from the early 1990s, as a range of major bilateral donors introduced participatory planning mechanisms as standard practice in their own projects, they encouraged the Bank to do the same, sponsoring a ‘Bank Wide Learning Group on Participatory Development’ and publishing various policy statements and ‘best practice’ resource books (World Bank 1994, 1996). Third, from the mid-1990s, the Bank also started moving participation up from the level of local project planning to economic policy and sectoral work, and eventually to country-wide political and economic planning. Fourth, the Wolfensohn Presidency brought new openings to expand and formalise these trends. In the late 1990s, questions of civil society participation became tied into demands for debt forgiveness through the Highly Indebted Poor Countries (HIPC) initiative, leading eventually to the PRSP process itself. Only at this stage did the IMF sign up to a policy agenda that emerged very much within the Bank.
Throughout these four phases, collaboration with external academics, bilateral donors and NGOs was central to the Bank's work. Establishing how the PRSP came into being, and the central role of NGOs in its development is an important element of the argument that PRSPs should not simply be understood as a World Bank project, but as a joint-project of a wide range of actors. As Christiansen and Hovland argue
The PRSP story is characterised by a multitude of links between the various players, both policymakers and researchers. The discussion of the main institutional actors … illustrates clearly the high level of contact between individuals in different institutions. It was also significant that individuals moved between institutions and thus carried ideas with them. As one interviewee put it, ‘none of the players is more than two handshakes away from any of the others’. The formal and informal networks contributed to the speed with which the PRSP ideas were spread and accepted in international development policy (2003:viii).
Reflecting on Brown's critique that PRSPs risk undermining democracy and allows for deeper colonisation of political space by Northern actors, Anthony Bebbington argues,
Even if these are – or might one day be – the effects of such scaled-up, ostensible commitments to participatory policy and to such apparent openness to more politicised discussions of poverty and exclusion, I doubt such effects were the intent of those who committed their lives, careers and often families to fostering such changes (2004:279).
I contend that this is the case. Solving the crisis of adjustment was transparently uppermost in the Bank's mind throughout the period that NGOs were lobbying for participation. For example, a 1995 Bank research document on participation in Country Economic and Sector Work lists the potential benefits that chime closely with the Bank's concerns about low implementation rates of structural adjustment programmes:
- •
A sense of partnership with borrowers;
- •
an increase in the Bank's credibility within countries and among stakeholders;
- •
a wider and deeper use of local knowledge, improving the substance of the work itself;
- •
a much increased sense of ownership of the product, which in turn facilitates the adoption of recommendations;
- •
speedier action on report results;
- •
expedited Bank review of the work because of acceptance by the borrower (Aronson, 1995:8).
First, the Bank's great hope was clearly not that participation would change the Bank, but rather that it would generate a new willingness on the part of borrowing country governments and civil societies to ‘learn from’ the Bank and implement that knowledge. For instance, the Participation Sourcebook notes that participation offers opportunities for ‘social learning’ wherby, ‘people within a local system learn the value and rationale of new social behaviors specified by an expert’ (1996:4).
Second, participation offers the possibility of direct IFI and donor contact with domestic political constituencies and debates. The World Bank's Participation Sourcebook discusses the notion of stakeholders applying to all the different actors involved in the aid and development business – the IFIs themselves, foreign aid donors, borrowing government, and civil society. The image is one of all stakeholders engaged in a dialogue of equals to find a rational solution. This blurs the distinction between stakeholders rooted within the local society and those not, while the idea of a non-hierarchical, dialogical process allows powerful, non-local actors to be more forceful in expressing its own views.
There were, however, limits on what the Bank thought could be done. Despite excitement about the possibilities of participation, the Bank remained aware of legal, theoretical and practical problems with it. Perhaps the most significant was the concern that new processes would not deliver if they were established in the absence of demand for them from either borrowing country governments or civil societies. This generated concerns about how to encourage borrowing country governments, if not to initiate, then at least to accept participation, and about what such processes might look like in the absence of the kind of civil society organisations that might be willing and able to participate. Bank researchers agonised about how far the IFIs could engage in solving these problems without offending against the sovereignty clause in its Articles of Association. In the mid-1990s, the last line that the Bank appeared unwilling to cross was taking responsibility itself for organising interests within borrowing countries, and directly mediating between them and the state.
Nonetheless, while participation claims to vest speaking and decision-making rights in the hands of local people, the Bank was clearly keen to ensure its own voice and decision-making power were strengthened in the process. With the range of new powers the Bank accrued to itself through participation, it seems hard to identify the benefits of supporting the strategy for those seeking to oppose adjustment. However, throughout the period, mainstream NGO lobbyists were pushing the Bank to go further, crossing even the lines it lacked the courage to.
The NGO campaign: Participation as Trojan Horse or strategic retreat?
How then did NGO lobbyists understand their campaign for participation to be increasingly scaled-up and for the Bank to directly support civil society? Just before James Wolfensohn took over the Presidency in 1994, the World Bank's fiftieth anniversary meeting in Madrid faced street protests. Campaigners that considered the Bank had done nothing significant to address their concerns marked the anniversary by denouncing the engagement strategy of more moderate groups, boycotting meetings and declaring, ‘Fifty Years is Enough!’
For the ‘insider’ lobbyists that believed they were making progress pushing participation, this provoked a challenge. Should they also walk out? Paul J. Nelson, one such lobbyist, writing his strategic reflections on Bank-NGO relations in 1995, came to a different conclusion. Nelson reasoned, first, that the arrival of Wolfensohn marked an opportunity to ensure that insider influencing continued, rather than the Bank and NGOs turning their backs on each other, and the participation experiment. He predicted (correctly) that Wolfensohn's strategy, rather than supporting the kind of backlash many frustrated Bank staffers would have liked to see against NGOs, would be to divide ‘reasonable’ from ‘extremist’ voices, and to use the reasonable NGOs to drive a reform agenda within the Bank.
Nelson argued that opposition to adjustment was becoming politically outmoded, suggesting that NGOs, ‘Acknowledge the need for macro-economic adjustment – balanced current accounts and good economic management – and help stimulate development and discussions of concrete, locally-based plans in the national political context’ (1995:190).
The key issues Nelson identified for lobbying were, firstly, debt relief and, second, participation, which he argued should be ‘a top strategic priority for NGOs … no other single measure would penetrate the Bank's standard operation and its information system as thoroughly as the expanded use of participatory methods’ (1995:191). Nelson's understanding of the relationship between adjustment, debt and participation, and his belief that the latter two could be strategically disentangled from adjustment illustrate clearly that the ‘debt relief plus participation’ agenda, which was to become a celebrated success amongst NGO campaigns, can also be understood as a strategic retreat from previous, more critical positions.
I have sought to make the argument that NGOs have been closely involved in the development of the PRSP despite the fact that a social control agenda was already clear within the Bank's approach. I also suggested that a sense of defeat over structural adjustment and political timidity (at least partially resulting from funding dependency on bilateral donors) provide one potential explanation of NGO positioning. However, an alternative reading suggested by Nelson would be that NGOs so believe in the power of participation methodologies to ‘reverse the perceptions’ of the powerful, and thus to secure the priorities of the poor, that they saw it as a Trojan Horse. Their assumption was that once the Bank accepted the need for and logic of participation it would have unleashed forces beyond its control. Such perspectives, with an almost religious belief in participation (Henkel and Stirrat, 2001), inform a good deal of contemporary NGO lobbying for participation. Before moving on to make the argument that the Bank and NGOs share a substantive agenda for reform, it is worth pausing to consider how participation might operate differently – as a disciplining mechanism.
Participation as a social technology of control
Arguments developed by Cooke (2001) and Hopgood (2000) in the fields of development studies and international relations provide some clues as to how participatory processes might operate as a social technology of control. Cooke challenges the dominant view that Participatory Rural Appraisal empowers participants by drawing attention to potential tyrannies of the group, capable of re-moulding the self-perception of participants. He argues that work-shopping and role-playing involve particular techniques of coercive persuasion, such that, ‘group processes can intentionally be shaped to set-up specific psycho – and group dynamics to achieve particular outcomes’ (2001:116). This facilitates the ‘unfreezing’, ‘changing’ and ‘re-freezing’ of personal identity. Un-freezing occurs through a dramatic illustration of the potentially bad consequences of maintaining the status quo. Visioning exercises can then be used to promote a change of consciousness by illustrating a better future. Finally, a re-freezing of identities and objectives occurs through techniques designed to reinforce and secure group ownership of any consensus achieved through the workshop.
Many of Cooke's arguments find a close parallel in Hopgood's challenge to ‘postmetaphysical’ theories of international relations. He builds on Andrew Linklater's use of Habermas' notion of a ‘communications community’ whereby the outcomes of discussions are ethically defensible, because each individual has an equal right to participate in a rational dialogue to design the rules by which participants will coexist. The rules of the communications community look a little like those of the PRA meeting, including conditions that no person or moral position can be excluded in advance, and that participants challenge their own ontological bases / truth claims, and anticipate that their positions will change through dialogue. Norms emerging can only be validated with the consent of everyone whose interests stand to be affected. These constructions claim to create ‘value-free’ or ‘post-metaphysical’ accounts, which secure elements of human freedom, generating virtuous outcomes without asserting a specific form of the good life. In other words, they claim to impose process conditions, without defining policy outcomes.
However, Hopgood argues that these models rely on an idealised model that, when applied to real people, requires them to accept a number of rules of the game. The ‘thin’ liberal conception of a neutral political process is thus not sustainable either theoretically or practically in the absence of a ‘thick’ normative content. The outcomes are in fact shaped by the fact that, in order to participate, real-world individuals have to be ‘unencumbered’, theatrically renouncing their existing particular identity, until they are stripped bare as abstractions. They can then be ‘reencumbered’, a process of education, supported by incentives that generate acceptance of a series of liberal norms. Thus what appear as neutral processes in fact require as a pre-condition, the inculcation of liberal values amongst participants. The very act of establishing the discourse community therefore, ‘presumes that persons are at least minimally liberal (other-regarding, egalitarian) in the first place. But what of the real people, people possessed of a vast array of prejudices, antagonisms, preferences, wants and desires, often embodied in highly legitimate social practices?’ (Hopgood, 2000:10). Section two assesses the potential for apparently neutral processes of dialogue to act as the kind of social technologies of control described by Cooke and Hopgood in the concrete case study of the PRSP process, investigating the capacity of the process to un-make and remake personal and group identities.
PRSPs in practice: The view from the aid industry
As we have seen, one of the key claims made for the PRSP process is that it will generate ownership of policy conditions. Given that ‘the concept of ‘ownership’ is a relative one’ (Eurodad, 2001:3), this section asks whether changes in practice under the PRSP are likely to bring us any nearer to autonomous, democratic development. In seeking to answer this question, I will look firstly at a range of assessments of the experience of the first round of PRSP negotiations. These studies were all undertaken, or commissioned by, agencies that have not only argued for greater participation, but are actively involved in the negotiation and implementation of the PRSP process, including NGOs such as Christian Aid (Painter, 2002; Richmond and Ladd, 2001); Eurodad (2001); Oxfam (2002, 2004); Save the Children Fund (Marcus and Wilkinson, 2002); and the multi-agency project on ‘The Reality of Aid’ (German et al. 2002), as well as bilateral donors, particularly the UK's Department for International Development (DFID) (McGee et al. 2002; Godfrey and Sheehey, 2000). As such, they tend to have a technocratic bent, identifying ‘problems and solutions’ within the context of broad acceptance of the PRSP framework. I then look at the recommendations for policy change that these ‘constructive critics’ offer, and finally at the advice they provide African partners. This latter analysis suggests a series of more fundamental limitations on autonomy and democracy implicit in the PRSP and based on prejudices shared by the IFIs and constructive critics.
Assessments of the PRSP process
NGOs and donors, who have led the field in developing participatory approaches to development, find numerous faults with the PRSP process. Although practice across different national contexts varies, McGee et al. describe most processes not as participation, but as ‘information-sharing’ or ‘consultation’ exercises. Assessments typically find that the experience of many of the PRSP processes has involved:
- •
narrowly defined agendas, established by government in advance;
- •
detailed proposals for policy change ‘pre-cooked’ and presented as items for discussion within defined constraints;
- •
limited processes of iteration. Rather than initiating debate, civil society criticisms and alternatives have been ‘processed’ by government in an extractive manner;
- •
domination of participatory processes by urban, professional groups, humanitarian NGOs and their umbrella bodies. Many of the major mass-based constituencies within civil society, rural based and peasants' groups, trade unions and religious groups, have not engaged with the process;
- •
limited time allocated to participation, resulting in invitations to meetings arriving at short notice and very brief gaps between the release of consultation documents and the deadline for comments;
- •
presentation and discussion of PRSPs in highly technical language and a failure to translate documents into local languages.
Within the assessments, blame for the failure of the PRSP process to influence macroeconomic policies is shared between civil society, national governments and the IFIs. Many assessments display a marked disappointment at the failure of civil society in developing countries to develop well-articulated positions on macro-economic debates. Many also suggest that national governments, as the lead agency in designing and managing participation, have frustrated the process, considering participation an external condition, which they have to fulfil to satisfy donors.
Most commentators also raise concerns about constraints on ownership enforced by the IFIs. Although frequently discussed as if they are simply national planning processes, discussions under the PRSP are not primarily about planning. IFI ‘approval’ of the final strategy determines access to debt relief and new credit. In many cases this is the real reason why ‘planning’ is taking place at all. IFI approval has become more and more influential as the IFIs increasingly act as ‘gatekeepers’ to numerous other potential sources of capital. As German et al. note,
If the IMF and World Bank reject a government's PRS, the government would lose access to trade credits, aid and finance and probably default on its debt obligations. Ultimately, its domestic economy could collapse (2002:10).
This pre-empting of IFI preferences is made possible by the IFIs' willingness to expound on the values and policies by which they make their assessments. For example, IMF head Horst Kohler, told a 2002 ‘Town Hall Meeting’,
the macro-economic framework is discussed with the country and it is not a prescription or dictate from Washington. … We do not have a one-size-fits-all approach.
if you have inflation rates of 10, 20 or even more percent, that's devastating for the people … and it's the same with the budget … the stability of a society really also depends on the fact that this society knows it has to live within its means (World Bank, 2002b).
The statement hints at the expanded challenge facing the IMF under the PRSP process: how to teach a society, not just a government, how to live within its means.
Given the range of tools available to the IFIs and the enormous risks for any poor country that sought to test the limits of the IFIs' tolerance, it is not surprising that standard neo-liberal policies continue to dominate. As a result, as Tim Kessler argues, if one PRSP looks a lot like a standard IMF package - with a larger social protection budget – that doesn't necessarily prove that alternative policies were ignored. But if most of them - or all of them – have that familiar laundry list of macro and institutional reforms, then there's a real problem. If the Bank and Fund will only ‘endorse’ PRSPs that fit into its preconceived box of standard policies, then it'll be one-size-fits-all adjustment lending with a politically correct name (Kessler, 2000, quoted in Marshall et al. 2001:8).
The range of criticisms described above illustrate that many who argued for participation are far from happy with how it is operating. However, in order to understand where the boundaries on their vision lie, we also need to look at the recommendations they make for change.
Policy proposals
The same broad recommendations are present in almost every assessment studied, including the World Bank's own ‘retrospective study’ (World Bank, 2002a). First, there is a call for ‘capacity building’ to ensure that Southern civil society groups have the information and skills to participate effectively. Second, every one of the assessments reviewed argues that there should be more participation in the development of PRSPs. Where the contradiction between ‘ownership’ and increased external involvement is recognised, this seems to be a price the constructive critics are willing to pay to ensure that maximum pressure, including the ultimate sanction of withholding funds, is placed on governments unwilling to play the participation game. Even those commentators who reject IFI involvement fall short of advocating allowing developing country governments to develop their own plans without international supervision. German et al. for example, describe PRSPs as,
the antithesis of domestically rooted and owned national poverty strategies … Foreign Aid has undermined decades of collectively negotiated governance processes in Africa, destroying the values that held societies together.
authentic ownership of these strategies depends not only on the quality of national efforts to consult and reach social consensus … but also requires donors to open the political space in which such efforts will evolve (emphasis added) (2002:8).
Thus, while many of the constructive critics recognise that PRSPs are being used as a legitimation mechanism for neo-liberal economic conditions, most are loath to let the PRSP go. They remain committed to building ‘bottom-up’ conditionality (externally policed conditionality based on policies established through participatory processes), see the PRSP mechanism as a flawed but powerful means of securing it and show concern for the reputation of the process and the continued engagement of civil society groups with it. How then should we understand this apparent paradox?
Practical engagement
Constructive critics might argue that they support PRSPs because more radical strategies, aimed at securing autonomous, democratic development, (for example by reducing the IFIs' structural power over developing countries through total, unconditional debt cancellation), are politically unfeasible, jeopardising widespread support for participation, aid and debt relief. However, we can look beyond assessments of the PRSP and recommendations for policy change. In order to understand how actors operationalise their approach to the PRSP, we can also look at ‘How To’ guides to PRSPs such as Oxfam's Guide to the PRSP Process (2002), targeted at their Southern partners, and at the World Bank's PRSP Sourcebook (Tikare et al. 2001). By making visible the contradictions, evasions and assumptions in this literature, I aim to show that the dogged attachment to the PRSP reflects more than a simple victory of hope over experience and that, in fact, both the Bank and many of the ‘constructive critics’ find autonomous, democratic development politically unattractive. In other words, despite tensions over policy and practice, the IFIs, bilateral donors and Northern NGOs are all wedded to a ‘liberal project’ (Young and Williams, 1994) which desires, and acts to secure, the intensive international supervision of African political communities. The PRSP represents an exceptionally useful tool for the promotion of this liberal project, since a ‘division of labour’ between the IFIs, donors and NGOs helps to prop up the fragile myth of the equal sovereignty of states, whilst securing the globalisation of developing country politics.
In order to make this argument, I aim to show that in the PRSP, Northern actors share a definable set of ideal political outcomes and act to delegitimate alternatives to liberal political and economic arrangements through a wholesale assault on existing political identities, projects and political institutions in Africa. These are gradually being replaced by international norms and standards, by accountability to external assessment, and by internationalised elite policy networks. ‘Bottom-up conditionality’ thus paradoxically legitimates increasing intervention and implies insulating political processes from the majority of the population.
The assault on existing political identities
As previously discussed, Hopgood notes that, before entering the ‘discourse community’, participatory theories require individuals to be unencumbered of their existing social and political identities and re-encumbered with appropriate modern, liberal identities. Illustrating that such a transformation could be attempted in relation to African civil society in the concrete circumstance of the PRSP is no easy task since the aid industry constantly stresses its respect for people's own identities and its recognition that people and communities are situated within specific contexts, that local knowledge is central to participation, that civil society is diverse, and that this diversity represents its strength. However, the task of critical analysis is to investigate the underlying limitations that such apparently progressive terms write into debate, asking what range of identities and political projects are legitimated and promoted, and what alternatives de-legitimated and squeezed out. I want to argue that respect for identity, context, knowledge and diversity are severely constrained by the fact that in the PRSP process, primacy is accorded to ‘poverty discourse’, and the generation of a form of ‘planning knowledge’ useful to the development of a national anti-poverty plan.
Nelson argued that in the pre-PRSP era, the diagnostic phase, ‘pre-appraisal’, based on future profitability calculations was, ‘deeply rooted in the World Bank's management and operating style’ (1995:117). Bank staff took a strong interest in the identification of the ‘problematique’, establishing – in the process – constraints on ‘appropriate’ knowledge to apply to the task in hand, and the solutions that could be proposed, and thus enabling them to disengage from the later stages of the process. Under the PRSP, ‘pre-appraisal’ has been replaced by the ‘poverty diagnostic’ phase. Researchers undertaking Participatory Poverty Assessments (PPAs) during this phase are encouraged by both the IFIs and NGOs to be particularly attuned to marginality, and specifically, to gender, physical ability, ethnicity, generation and most importantly, poverty. The ‘local knowledge’ of the objects of research is thus legitimated with reference to a proven experience of marginality. This process enables a particular construction of the problematique that PRSPs set out to solve, and suggests a new one-size-fits-all solution – an anti-poverty programme, developed according to international norms of ‘best-practice’.
By focusing attention on what people lack (power, wealth, status) and on their ascribed identities (old, child, poor, female, disabled etc.), poverty discourse thus describes people and countries as objects, victims of their circumstance. Francis writes that,
Even those without material means can still be consumers of development, at the price of being reduced to atoms of poverty with no more social identity or history than being ‘poor’ (2001:85).
The ascribed identities of poverty discourse can be contrasted to identities based on the consciousness of particular groups, formed in the interactions of memory, history, culture, religion, place, group and perhaps most significantly, of interests and political ideology. Investigating these subjective identities and desires implies viewing people as subjects, able to think and act independently, to shape events, mandate legitimate representatives and mobilise collective political action. However, since the knowledge represented here may not be expressed in terms of poverty, it can legitimately be excluded by a process that declares itself interested above all in ‘poverty reduction’. We can see this exclusion at work in McGee et al.'s concern that civil society participation will be dominated by ‘interest groups more interested in pressing their own case’ (2002:9). Though classical liberal theory understands civil society's principal role as the defence of private ‘interests’, McGee et al.'s position illustrates the narrow range of projects this contemporary (more moralistic?) liberal project can support. In the process, it delegitimates, amongst others, trade unions and producer groups that exist primarily to promote their members interests.
Similar constraints are also at work in the assumption, shared by the IFIs, donors and NGOs, that what PRSPs both can and should do is to generate a national consensus. This focus on consensus asserts, or insists upon, a harmonious civil and political society, bereft of mutually exclusive ideologies and interests. Abrahamsen notes that under governance discourse, ‘civil society emerges as undifferentiated and harmonious, and there are no classes, races, no genders, ethnic groups or oppressors’ (Abrahamsen, 2000:56). Actors who disrupt this equilibrium immediately place themselves in a position to be condemned for their inappropriate or ‘nonparticipatory’ attitudes.
Despite a heavy rhetorical focus on local knowledge and diversity, participation under the PRSP is mainly interested in people's ability to perform according to established liberal norms of political behaviour and to insert their ‘knowledge’ into the framework of planning tools. These tools are designed to collate and represent experiences of poverty towards a pre-determined end, the production of a plan for national development backed by IFI lending.
In this context, the ‘failure’ of African civil society to engage with the PRSP process becomes more understandable. PRSPs are designed by and for groups capable of expressing their project in terms of planning knowledge and poverty discourse – but few such groups exist, and few are likely to emerge from domestic social processes. PRSPs thus inevitably fail to engage with the assumed identities through which active human agents define their own ends; in other words, with the forces driving actually existing civil society in Africa, particularly groups which mobilise around conflicts of interest and identity and pursue their projects through strategies of resistance and non-co-operation. The problem of a lack of ‘capacity’ in civil society can, therefore, be understood as a lack of capacity to rehearse political arguments in the form of ‘planning knowledge’. And ‘capacity building’ can be understood as a process of teaching elite cadres within civil society to express themselves in an appropriate and professional manner – to perform as a version of Western ‘good citizens’.
Attacking existing political institutions
Before the PRSP era, African states were recognised as the only authority mandated to negotiate with external actors on behalf of their populations. However, a deep suspicion amongst donors and the IFIs towards African states has long informed their desire to engage wider stakeholders. Furthermore, the bias towards consensus involved in participation tends to delegitimate struggles for a majoritarian mandate between conflicting interests, ideologies and representatives. This orientation has serious implications for the legitimacy accorded those political forces which have ‘captured’ formal state power through either democratic and more physical ‘struggles’ – in other words all existing governments. The state thus appears as the antithesis of bottom-up, local, sensitive and consensual processes, and the demand for participation emerges as a challenge to the state's monopoly on representing the citizenry, and from a claim that an alternative popular sovereignty resides in civil society. As Brown notes,
the legitimacy of the PRSP process … derives as much from the alleged limitations of representative democracy as from the manifest qualities of the participatory approach (2004:243).
I am deeply troubled by the distance that the Bank has gone in democratic countries towards engagement with groups other than the Government. … when there is an attempt to reach within societies to develop Country Assistance Strategies, there is a real possibility, it seems to me, of significantly weakening democratically elected governments (quoted in Eurodad, 2001:5).
The primacy accorded representative institutions over civil society has featured prominently in NGO documents as a cause of ‘poor governance’. Christian Aid, for example, commented disapprovingly that in Malawi, ‘The Government acted on the belief that once elected it had a mandate to govern and design policy without being questioned by unelected bodies’ (Painter, 2002:5). Typically the constructive critics argue that this democratic deficit can be remedied by participation, presenting the possibility of moving up the ‘ladder of participation’ from ‘information sharing and consultation’ to a point where stakeholders ‘initiate and control’. The way in which these ideas, drawn from the literature on PRA, have been (carelessly?) dropped into the discussion of IFI-led national planning processes is telling. By imagining African governments into the same position as external facilitators, they are constructed as politically illegitimate actors.
More recently, however, both the Bank itself, and donors such as the UK's DFID, which has been a staunch advocate of PRSPs, have started to worry about the relationship between PRSPs and democracy. Furthermore, some of the most effective campaigns against PRSP conditions have adopted the Parliamentary route. For example, the resistance to adjustment conditions documented by Larmer (2005) in Zambia involves trade unions creating pressure on elected representatives, and in turn on the Executive, to seek a re-negotiation (so far unsuccessful) of the terms of the PRSP. Some of the more radical NGOs have seen a way in to criticism of the policies, arguing for the engagement of Parliaments and elected political representatives in the PRSP process. This has included a global petition of Parliamentarians arguing that they should have the right to debate and vote on PRSPs before they are endorsed by the IFIs. However, even where NGOs have been at their most reflective about participation (Rowden and Ocoya Irama, 2004), the role of Parliamentarians is very much alongside participation of the poor – the question seems to be how to draw Parliament into involvement in the PRSP – to ensure political ownership of the process. This is somewhat different from asserting the primacy of elected institutions and formal democratic political processes over hazy, non-representative participatory processes.
Increasing surveillance of the state
The challenge to state authority under PRSPs is not merely rhetorical. Through the PRSP the IFIs aim to secure adjustment conditions against any government backsliding. This explains the shift towards releasing credit in ‘tranches’, triggered by reviews. However, surveillance is political as well, and the PRSP provides a means of empowering ‘reform coalitions’ to defend adjustment against the re-assertion of popular demands (classically through wage strikes from government employees), and resulting ‘political’ budgetary allocations. It is thus not surprising that ‘budget monitoring’ is promoted by the World Bank as the most significant role of civil society in the PRSP:
Participation in decisions regarding budget allocations, spending patterns, and public service delivery is thus a key entry-point for civil society engagement … enhanced participation in budgeting means a more powerful voice for the poor, improved public transparency, strengthened external checks on government and ultimately, greater efficiency and efficacy (emphasis added) (Tikare et al. 2001a:119-120).
This positioning could be understood in three ways. Constructive critics may trust that they will, in the long run, be able to secure their spending priorities throughout the policy cycle. Alternatively, they may not actually prioritise macro-economics, being satisfied with any welfarist crumbs falling from the PRSP table. Or it may be that they share the IFIs' concern to establish the primacy of the PRSP over domestic political processes and welcome the leverage that the IFIs provide in securing a long-term shift in power away from widely condemned political cultures and institutions and towards civil society. In other words, the constructive critics may not trust the IFIs, but they appear to trust unsupervised African political processes a lot less.
Constructing a professional civil society
In participation theory, as with the PRSP, ‘process’ is often presented as more important than ‘outcome’. Advocates of participation suggest that they are responding to a demand, driven by the ‘emergence’ of civil society, to open up new political spaces, and that it is from these that innovative solutions to underdevelopment will emerge. However, this assertion stands in awkward contrast to widespread recognition of the absence, in many countries, of the type of civil society that might like to participate in PRSPs.
This creates a problem for international actors under pressure to deliver the normative and institutional content to make new processes perform. As Hopgood notes, it is not enough simply to strip away inappropriate identities. In order to engage in discourse, actors need to be ‘re-encumbered’ with new ‘ends’ – projects to pursue through participation. What I am suggesting here is that international aid donors and NGOs play a key role in the process of ‘re-encumbering’ civil society to fill new political spaces. This involves creating and ‘capacity building’ a ‘professional’ African civil society, encouraging new and existing groups to engage in the PRSP process, and providing a tailor-made set of civil society ‘ends’ for them to pursue.
The Bank appears primarily concerned with accessing ‘expert knowledge’ on local development challenges and with mobilising ‘reform coalitions’. For instance, the PRSP Sourcebook encourages governments to think carefully about how to win support for adjustment.
Participation is inherently political, and understanding the dynamics of participation, particularly in macro-economic policy-making requires thinking in terms of interest-groups and incentives … Demystification of what the policy is, the rationale behind it, and its impacts will be critical in giving various societal groups incentives to support the policy and build positive coalitions of stakeholders (Tikare et al. 2001a:37)
Encouraging engagement with the PRSP & international actors
Bilateral donors and NGOs thus encourage ‘African civil society’ to fill the political space provided by the PRSP process by providing advice, training and funding to ensure that more and more civil society groups emerge, and that those willing and able to engage do so. Not unlike the relationship between donors and African governments, the funds, information and advice come with implicit and explicit conditions, introducing new sources of ‘accountability’ for a professionalised civil society.
However, the methods and priorities of aspirant participants in these internationally oriented, modern, liberal and technocratic discourse communities stand in stark contrast to those leading ‘new social movements’ – the mass based, sometimes democratic, frequently identity-based organisations that lead political resistance to SAPs. Historically, mass civil disobedience in response to adjustment (the IMF riot) have in some ways strengthened developing country governments' efforts to resist conditions, providing a concrete illustration of the limits of political feasibility of reforms. McGee (2000) suggests that new policy networks emerging from the PRSP process might provide African states with an alternative method of illustrating to the IFIs the limits of political feasibility. As civil society groups provide the state with ‘evidence’ on the negative impacts of adjustment and the realities of poverty, they strengthen the hand of the state in negotiations with the IFIs. She also recognises however that this potential has not been realised. This is perhaps not surprising given the instincts of many in the aid industry to support alliances between the IFIs and civil society, and against the African state. Oxfam's ‘Guide to the PRSP’ for example provides clear advice to Southern partners that they should engage with the PRSP process ‘constructively’, implicitly accepting the legitimacy of the process, and build new networks to promote their agendas. This includes suggesting that groups orient themselves not towards confrontation or critique, but prepare for negotiation, self-censorship and compromise in their engagements. Oxfam also advises Southern civil society to ally with the IFIs and donors in order to influence their governments. The Guide advises, ‘Use Donors and the WB/IMF: These often have a strong interest in Public Expenditure Management and at the same time exercise a lot of leverage over governments … Increasingly donors are actually putting staff inside Ministries and these in particular can be invaluable sources of information. Developing good relationships with these donors is essential’ (2001: 25). This strategy of Northern NGOs has caused resentment amongst social movement activists (Dembele, 2003, for example).
Yet in some places, the complaints are beginning to be heard. One notable exception to the dominant ‘insider lobbying’ strategy advocated by most Northern NGOs is ‘Re-thinking Participation’ (Rowden and Ocaya Irama, 2004) which was recently published as a discussion paper by Actionaid. While the paper argues PRSPs may have brought some benefits, principally in terms of learning and relationship building for civil society, and in improving relations between civil society and the state in countries where little trust previously existed, Rowden and Ocaya Irama also warn about the dangers of dominance in the process by Northern NGOs, and of the process undermining sovereignty and democracy in recipient states. They suggest that NGOs should argue for increased domestic democratic oversight of the process, and focus their own efforts on alternative public forums, civic mobilisation and engagement of parliamentarians – in brief, an agenda for civil society relying less on links to the powerful and international community to build their influence, and more on securing their own domestic support base.
Legitimating intervention
The drive for engagement described above is particularly important in order to underpin the World Bank's argument that, ‘Economic policy is primarily driven by domestic politics, not by outside agencies. The key to successful reform is a political movement for change, and donors cannot do very much to generate this.’ (Devarajan, Dollar and Holmgren 2001). However, I have argued throughout this discussion that many actors in the aid industry are energetically engaged in an effort to do precisely this – to generate internationally-oriented reform coalitions. As Hopgood argues in relation to the ‘civilising mission’, whilst there has been a ‘decline in the publicity surrounding the inculcation of virtue, the process itself, spreading the virtues to make liberalism work, still goes on, unceasingly, just without the fanfare’ (2000:3).
A division of labour between international agencies plays a central role in enabling this inculcation of value to take place without finally condemning the fragile myth of sovereignty. This division of labour involves the IFIs opening up new political spaces within developing countries by enforcing process conditionality, before these spaces are rapidly ‘filled’ both by external actors and trusted local groups.
Thus although the World Bank has hesitantly expanded its mandate, it is precisely because the IFIs cannot confidently engage with domestic political processes that they have sought increasingly close collaboration with bilateral donors and civil society through the PRSP. There is a degree of confusion within the Bank as to how far its mandate will stretch, but by employing a division of labour, no one agency can clearly be accused of a dramatic breach of sovereignty. Rather an ‘extended state’ makes multiple ‘contacts’ or ‘networks’ across state boundaries. Intervention becomes a series of minor, inter-linked actions, none of which, on its own, appears sufficient to cause grave offence against sovereignty but which, as a whole, increasingly shift the locus of decision making from South to North.
In this situation, it is not feasible to argue that particular external actors determine outcomes. Rather, given the multi-faceted supervision of developing country politics under PRSPs, ‘the international is always present at the national and local levels’ (Abrahamsen, 2000:1). Analysis of this situation requires a focus on the interactions between domestic and external actors. As Abrahamsen argues, ‘little will be gained if the recognition of the history and agency of Africa and its peoples leads merely to the opposite extreme, namely of the abstraction of the continent from its international setting’ (2000:7).
The greatest weakness of the dominant literature on the PRSP process may therefore be its acceptance of this abstraction, a move that renders the role of external actors invisible. To give an extreme example, DFID give high priority to PRSPs in a summary document designed to explain the Department's role to a public audience. But they make no mention of the role of the IFIs, commenting, ‘If they are to help people climb out of poverty, developing countries must develop and carry through their own master plans called poverty reduction strategies’ (emphasis added) (DFID, 2002:4). Such accounts present the PSRP not as a central encounter in the international relations of a developing country, but simply as the sort of thing any poor country might do as a matter of best practice in public administration.
Oxfam, who one assumes would still recognise themselves as a ‘Northern’ NGO, also appear not to see any relevant distinction between domestic and international actors, viewing PRSPs as an opportunity for African civil society and for Oxfam to normalise and institutionalise its own influence over policy processes in developing countries: ‘PRSPs offer Oxfam and other NGOs major opportunities to influence policy and practice at local, national and international levels, both at the formulation and implementation stage’ (Oxfam, 2002:6-7).
The few commentators who note the risk of ‘ventriloquising’ Northern demands into local processes typically blame donors, not civil society. German et al. note,
local policy processes … [are] disempowered by the better-prepared, well-informed globally articulate donors who could not resist stepping into the policy vacuum (German et al. 2002:14).
Many international NGOs (and the IFIs themselves) work hard to establish some form of claim to a ‘local’ mandate. Beyond participation, strategies include hiring more ‘local’ staff to run offices in Africa and the North and the adoption of ‘federal’ structures. It is hard to believe that anyone truly believes that these strategies give organisations any kind of a mandate from African societies. However, the blurring of boundaries achieved appears to help Northern actors explain their interventionist role in African political communities. This is perhaps the only way they can present the PRSP as a legitimate, nationally owned process.
Conclusion
This article started by seeking to understand the origins of the IFI's concern with participation, situating the emergence of PRSPs within the context of a multi-faceted crisis of adjustment. An historical analysis suggests that, in seeking to make structural adjustment perform as advertised, the World Bank has experienced a mission creep. The Bank persistently finds that its every solution to the problem of African under-development is undermined by factors below the level of its previous intervention and, driven by that discovery, has burrowed deeper and deeper into the fabric of African political communities in order to finally reach the source of the problem.
An initial concern with getting the prices right soon shifted to a concern with getting the wider policy environment right. Backsliding on policy advice then led to a determination to discipline the ruling elites by imposing conditionality. Nonetheless, democracy allowed the election of new governments hostile to adjustment, and social unrest continued to undermine implementation. The Bank responded by developing a surveillance architecture capable of disciplining democracy. Still, forms of bureaucratic resistance and obstruction persisted. Graham Harrison's (2004) description of ‘Governance States’ provides a detailed account of the manner in which institutional reform, civil service training and donor technical assistance has allowed an ever more intimate supervision of every facet of the state bureaucracy in selected countries. Harrison describes a set of states that have ‘succeeded in internalising the impetus of governance’, meaning that they have achieved, ‘progress towards the resolution of a set of structured tensions generated by a decade of neo-liberal reform in sub-Saharan Africa in the 1980s’ (2004:4).
However, Harrison's argument focuses on the successful disciplining of the state machinery in Africa. It has little to say about politics at a sub-state level, about the ability of African societies to resist the governance agenda and to make their own history. The Bank, on the other hand, has long been worrying about the ‘next layer down’. It did so impotently in the 1980s as the ‘reform coalition’ literature expressed nagging anxieties about the difficulty of the long-distance social engineering of the state and civil society needed to embed adjustment. Despite experiments with participation, it was thus not until the emergence of the PRSP that the Bank developed a self-confident and comprehensive mechanism to attempt such an ambitious feat. The Bank tends to suggest that as a ‘learning organisation’ it recognised weaknesses in its previous moral and developmental scheme, took some good advice, and found the light. PRSPs have generated great pride and a new sense of mission for some IFI staff, allowing them to see themselves setting the pace in aid policy. Alternatively, to the extent that NGOs have gradually become disillusioned with PRSPs, they also present the process as one through which the Bank has taken another ‘NGO word’ and twisted its meaning to secure their own interests.
We can also speculate that the Bank's ambitions do not stop at the level of more or less institutionalised representative bodies within civil society. One layer further down we may find the populations of African states. The beliefs and attitude of these individuals may eventually become the target of Bank interventions. Williams (1999) has argued that, in order to make its market model work, the Bank attempts to remake African people, constructing a ‘homo economicus’. Craig and Porter make the argument that this agenda is intrinsic to the normative project pursued by PRSPs: ‘both poverty reduction and social inclusion policy and strategy represent a refinement of the liberal political project, specifically a mode of ‘inclusive’ liberalism, in which the disciplined inclusion of the poor and their places is a central task’ (2003:54).
We can provide an account then of the possibilities inherent in the participation agenda for key Northern actors seeking to develop PRSPs as a social technology of control. This article demonstrated that this is their aim. However, whatever the intentions of external actors – to impose particular conditions, to inculcate particular values, to neutralise resistance – this tells us little about how these processes are playing out in the concrete reality of African countries. PRSP processes are unlikely to easily supplant ‘deep’ social processes which actually existing political and social movements will react to and construct. Rather, negotiations are likely to proceed as a series of ‘performative rituals’ (Mercer, 2003), in which all actors construct their voices strategically. The extent to which a shift from coercion to consent gradually becomes possible can only be discussed in the context of a new research agenda – one that provides detailed ethnographic descriptions of the impacts of participatory planning processes on the personal and group identities of African actors, and on the bases for the formation or collapse of new coalitions, both promoting reform and resisting adjustment.
Aware of the tendency towards performance, the Bank continues to suffer from uncertainties about its right and ability to transform African political economies, about the failure of the IFIs' core economic propositions to generate dynamic economies in its client countries, and about the long-running institutional and legal tensions between the Bank's historic mandate and its expanded modern mission. Given the wide range of unresolved tensions, the Bank remains aware that the rhetoric of participation has out-stripped the reality. The fall-back position for the IFIs throughout the PRSP period has thus remained ‘business as usual’. In other words, until participation can prove itself capable of generating and policing consent to core IFI priorities, the IFIs hold on to the most important coercive mechanisms at their disposal. We are nowhere near a world of post-conditionality.