6 Developing global standards
International institutions seek to govern in the name of and through universals. By claiming to be working on an issue of universal concern – such as global poverty, human rights or sound economics – international organizations (IOs), non-governmental organizations (NGOs) and donors seek to gain authority. At the same time, they often represent the kinds of expertise that they use to tackle these problems as universal. World Bank and International Monetary Fund (IMF) staff drew on both of these universalizing strategies during the structural adjustment era: arguing that they were seeking to achieve the universal good of economic development, while simultaneously drawing on what they saw as the timeless universals of economic theory to identify the policies needed to achieve those ends. As I discussed in Chapter 4, the inflexibility of these universals was one of the central reasons for the erosion of the international financial institutions’ (IFIs) expert authority in the mid-1990s. As a growing number of critics described these rigid economic prescriptions as failures, IFI staff, management and directors sought to respond – not by rejecting universals altogether, but by problematizing and ultimately redesigning them.
If anything, the rhetoric of IFI and donor leaders became more explicitly universalist in the late 1990s and early 2000s, as they sought to build support for their vision of a new global economic order in the wake of the contested failures that they faced. The World Bank President, James Wolfensohn, initiated this new universalism with his famous speech about the “cancer of corruption” at the 1996 Annual Meetings, while the IMF Managing Director, Michel Camdessus, promised in 1999 that the institution would contribute to “civilizing globalization,” the UK Minister for Development, Clare Short, argued that “We have a moral duty to reach out to the poor and needy,” and Camdessus’s successor, Horst Köhler, called in 2002 for a new “global ethics.”1 This is a language that was unashamedly universalist.
Yet the kinds of universalizing strategies being used today by IFIs and donors are very different from those of the structural adjustment era. For one thing, their scope has expanded: when Wolfensohn or Camdessus talk about the universal value of their institutions’ efforts, they are not just talking about economic development, but also recognizing the social and political underpinnings of economic success. The new global norms of development include standards of political as well as economic transparency, judicial as well as central bank independence, and maternal as well as economic health.2 Not only the scope but also the form of these universals has changed. The law-like economic rules of the 1980s and early 1990s have been supplemented and replaced with more flexible standards, often taking the form of best practices and benchmarks.3 These new standards seek to define the norm in far more complex and contested arenas than in the past. Despite efforts to black-box them as thoroughly as the economic rules that they replace, more effort is therefore required to justify and maintain them.4 As the IFIs in particular move beyond the narrower mandate of the structural adjustment era, they have felt the need to justify the universality of these standards in moral as well as expert terms – which helps to explain the strikingly normative tone of some leaders’ rhetoric.5
This chapter examines the evolution and logic of this new universalist strategy of standardization by focusing on two policies pursued by the IMF and World Bank: the development of the “good governance” agenda from the 1980s onwards in response to the perceived failures of traditional aid, and the more recent adoption of the standards and codes initiative after the financial crises of the late 1990s. Both initiatives involve a range of other institutions in the process of governance: practices designed to foster good governance have been adopted by many donors, while the standards and codes initiative relies on organizations like the Basel Committee on Banking Supervision, the International Accounting Standards Board, and the Organization for Economic Co-operation and Development (OECD) to develop the standards. While I will focus my attention here on these policies’ initial development by the IMF and World Bank, these standardizing practices have thus gone on to become part of a wider community of practice in development finance.
Although the standards and codes initiative has received less attention, the rise of good governance has been widely examined in the academic literature, particularly by those working on the World Bank.6 What is gained from revisiting these policies yet again? The simple fact of expanding the discussion from considering just good governance to looking at standards and codes, and of looking beyond the World Bank to the IMF, reveals these changes to be more widespread and profound than the conflicts within and around any single institution or issue area. The focus on the how of governance also reveals the complexity of the changes involved, particularly when one looks at the most recent developments in the good governance agenda – the emphasis on the “demand side” of governance – which has been understudied to date.7 Finally, I have drawn on verbatim minutes recently made available from the IMF and World Bank archives, which provide additional insight into the debates around these policies.
A meso-level focus enables us to see these new policies as part of a broader strategy of developing global standards, and to understand them as a way of governing in the name of and through universality. In order to grasp the character and significance of these emerging policies, we need to pay attention to how good governance and standards and codes do the work of governance. There are important ways in which standardization acts to foster a new kind of normal, a new kind of norm, as Michel Foucault would suggest.8 Yet, such analyses tend to over-generalize contingent, locally driven processes as part of a singular logic of governmentality.9 Within sociology, particularly among those working on science and technology studies (STS), there is a growing literature on processes of standardization that I will draw on throughout this chapter, which draws attention to their often local, contingent and socially constructed character.10
The good governance agenda and the standards and codes initiative both define new global standards of economic development, and pressure governments to adopt them. In pursuing these policies, IMF and World Bank staff have drawn on small “i” ideas to redefine universals, changing how they go about the business of governance. They seek to enrol a wider range of actors in the processes of governance – not only member governments, but also market and civil society actors. As in the ownership strategy discussed in the previous chapter, institutional staff rely on new, more performative and reflexive techniques, including new ways of measuring and publicizing compliance with standards, and new technologies to encourage popular pressure, thus redistributing a measure of governance authority. They draw on different forms of power in doing so, including more indirect and proactive forms that seek to produce cultural changes aimed at longer-term transformations. As power has become more productive, the forms of exclusion that it generates have become more complex, a matter of degree rather than of absolutes.
Together, I will suggest, these shifting factors of governance once again point towards the emergence of a provisional kind of governance. This form of governance uses indirect but proactive techniques to encourage market and civil society to pressure governments to comply with new standards. Its advocates use techniques that are increasingly symbolic and highly performative to achieve these ends, relying on the fiction of credibility to do so. And although all of these various gambits are designed to minimize the possibility of failure, they nonetheless remain aware of its dangers and seek to hedge against them.
In the last half-century we have developed a better understanding of what helps governments function effectively and achieve economic progress. In the development community, we have a phrase for it. We call it good governance. It is essentially the combination of transparent and accountable institutions, strong skills and competence, and a fundamental willingness to do the right thing.
Over the past two decades, it has come to seem natural that IFIs and donors would make good governance and limits on corruption part of their development programs. Governance conditions now apply to over 85 per cent of IMF programs.12 Governance factors also account for over two-thirds of the country policy and institutional assessment (CPIA), which the World Bank uses to determine how much concessional assistance poor countries are entitled to through the International Development Association (IDA).13 Governance is also front and centre in many donor assistance programs, from the United Kingdom’s Department for International Development (DFID) to Canadian international development programmes and the United States’ Millennium Challenge Corporation (MCC).14 When governance was first introduced into the World Bank and IMF, it was politically sensitive and the subject of considerable debate.15 Since then, however, good governance and anti-corruption efforts have come to be seen as an essential part of development finance.
The evolution of a governance agenda
Given that the term “good governance” is ambiguous, it should come as no surprise that it has come to mean different things at the World Bank and the Fund. While there are overlaps in the practices of fostering good governance at the two institutions, it is worth looking at each separately, since the path that each has taken is different.
The World Bank
The World Bank prides itself on being the first institution to recognize and act on the idea that governance is central to economic development. Although its staff and management have certainly retained that conviction over the past twenty-odd years, they have defined and acted on that idea differently. It is possible to define two broad phases in the evolution of the governance agenda at the World Bank: the first phase, from 1989 to 98, was an extension of neoliberalism and defined governance primarily in public choice terms as an effort to avoid rent-seeking by creating a leaner, more effective government. The second phase, dating roughly from 1999, saw a broadening of the governance agenda to include the Bank’s new emphasis on poverty reduction and a shift in the theoretical framing from public choice theory to new institutionalism; this phase also saw the development of more experimental policies aimed at fostering the “demand side” of governance.
The term “good governance” first emerged in response to the perceived failures of development in sub-Saharan Africa. In an influential 1989 report on the subject, the authors argued that the principal source of this failure was not external – in declining terms of trade, for example – but internal, based on a failure of investment linked to bad public management.16 This “crisis of governance,” they argued, must be addressed before economic progress can be expected.17 The report is framed in neoliberal terms and can be read in part as a neoclassical rebuttal of dependency theorists’ claims that the causes of underdevelopment are within the capitalist system.18 While the report places some responsibility for failure on the Bank for its inability to recognize the importance of institutions, it also implies that the ultimate blame for underdevelopment rests with poor countries’ governments. Although the report was never formally discussed by the Executive Board, nor served as the explicit basis for policy discussions, it framed later discussions on the legality of the Bank’s move into the governance area.19 In fact, during discussions of the Bank’s legal mandate in this area, the Libyan Executive Director, Salem Omeish, went so far as to charge the Bank staff with trying to “sneak” the results of that study into policy discussions.20
This report launched a broader process of debate and problematization at the Bank, as the issue of good governance became a central preoccupation, with Bank staff producing several further reports on the subject and Executive Board members debating the emerging policy.21 Not surprisingly, those involved in these initial debates about governance drew heavily on economic theory, using a public choice conception of state–market relations that treated all of the players as self-interested and individualistic agents.22 Perhaps the most pervasive argument made throughout staff documents at this time is that rent-seeking is the central problem of governance. Rent-seeking is a concept that assumes that the state’s ability to make decisions about resource allocation – e.g. the building of a dam in a particular location – can have perverse consequences as it encourages the unproductive use of resources by those seeking to affect government decisions.23 The most touted solution to rent-seeking in the public choice literature is to reduce the scope of state decision-making by shifting greater responsibility to the markets.
By the mid-1990s, the governance agenda was having a concrete impact on Bank operations: in 1994, a report on the Bank’s experience of governance programs noted that the volume of governance-related lending was significant and increasing, with as many as 68 per cent of lending operations containing some kind of governance dimension.24 Yet, even as the idea of governance began to take hold within the institution, it was clearly a fraught issue. The Bank’s General Counsel, Ibrahim Shihata, was asked to provide a legal opinion on whether the institution’s mandate allowed it to address governance issues. Shihata’s opinion – which he only provided after significant pressure from management – narrowly reduced the scope of the Bank’s involvement to factors that had a direct impact on economic development.25 When the Bank’s executive directors discussed the legal opinion in an Executive Board seminar, their conversation was intense, with one director noting that the subject was “clearly delicate and perhaps emotive,” while another went so far as to charge the institution with attempting to “move into the distinct political arena and dictate [a country’s] political agenda and ideology.”26
It was not until James Wolfensohn took the helm of the World Bank in 1995 that the issue of governance – and the related problem of corruption – took centre stage. In a famous speech at the 1996 annual meetings, Wolfensohn called for an end to the “cancer of corruption.”27 It was during his tenure that the 2000–1 World Development Report (WDR), Attacking Poverty, and the 2002 WDR, Building Institutions for Markets were released.28 Together, these two reports altered the good governance strategy in several respects.29 First, they justified good governance based on its capacity to reduce poverty. Second, they reframed the theoretical justification for good governance in terms of new institutionalist economics.30 This shift is significant because although an institutionalist approach remains consistent with much neoclassical economic theory, it emphasizes the problems of market failure – instances in which the state must step in because markets are unable to allocate resources effectively.
The good governance agenda also began to focus more explicitly on the importance of public participation and demand. The role of participation had been controversial in the early 1990s: Shihata’s 1991 legal opinion had identified participation as a borderline case that might be viewed as too political, yet a number of directors were vocal in their advocacy for its importance.31 By 2000, in the context of the development of the ownership strategy, the idea that governance reform should be driven by the “demand” of public and private actors (and not just the “supply” from lenders) had become a defining feature of efforts to foster good governance.32 When Wolfowitz took over as Bank President in 2005, he continued this emphasis on the demand side of good governance. As one World Bank staff member put it in 2011,
A lot of Wolfowitz’s enthusiasm for governance and anti-corruption has given a real boost to an interest in citizen participation, because – we can have a discussion about neo-conservatism and Strauss and some very interesting philosophical ideas – but a central neoconservative idea is let’s support human rights, citizen rights and grassroots democracy. . . The [World Bank] President loves this stuff.33
This focus on local participation was integrated into Wolfowitz’s governance and corruption strategy (GAC), which remains the principal framing document for governance activities at the Bank.34
International Monetary Fund
The story of the evolution of the IMF’s governance policy is somewhat shorter since the Fund only got on board with the good governance agenda in 1997 and has remained more consistent, broadly in line with the Bank’s early public choice-driven approach.
In 1996, following a directive from the IMF’s Interim Committee,35 the Executive Board sat down to discuss a staff paper on “The Role of the Fund in Governance Issues.”36 Like the Bank’s early governance strategy, Fund staff drew extensively on public choice theory to frame the problem of governance, arguing that the central issues were those of rent-seeking and ad hoc decision-making, which could be best resolved through continued economic liberalization to reduce opportunities for government mismanagement.37
Despite Fund staff and management’s attempts to define governance issues as consistent with the institution’s traditional role, the attention to governance clearly expanded the organization’s mandate. A 2001 report on the Executive Board’s discussion of governance issues provides a useful overview of the new practices undertaken by the IMF under the rubric of “good governance”:
[S]trengthening revenue administration; enhancing financial accountability of state enterprises; improving bankruptcy laws and procedures; consolidating extrabudgetary funds into the budget; enhancing transparency in tax and tariff systems; reinforcing central bank independence; extending prudential bank supervision; and improving economic and financial statistics.38
In concrete terms, such governance issues were not only raised in the context of the Fund’s usual Article IV consultations with member states, but were also integrated into its lending conditions.39 Good governance practices therefore had teeth, since program approval could be “suspended or delayed on account of poor governance.”40
In a 1998 speech on the IMF’s role in good governance, Managing Director Michel Camdessus argued that there existed a “universal consensus” on the importance of good governance.41 Yet, in truth, this expansion of the Fund’s mandate was more contested on the Board than Camdessus suggests. Jack Boorman, then head of the Policy Development and Review (PDR) department, which was responsible for developing the governance policy, has since noted, “the resistance to the governance agenda was amazing in the mid-1990s.”42 Camdessus himself admitted as much in the same speech cited above, suggesting that initially “Some of our shareholders feared that in taking on such issues the [Bretton Woods] institutions would become politicized and lose their effectiveness.”43
Those pushing to broaden the Fund’s role to include governance issues ultimately won the day. In fact, as the 2001 review of the IMF’s experience noted, engagement on governance had expanded well beyond the staff and Board’s initial expectations when they had first approved the Guidance Note on Governance in 1997. The principal reasons for this expansion were the Asian financial crisis in 1997–8, which led to the standards and codes initiative discussed below, and the decision to link governance conditions to debt relief as part of the Highly Indebted Poor Countries (HIPC) initiative.44 The emphasis on governance issues also continued to increase in the Fund’s bread-and-butter policies – in its Article IV surveillance consultations and lending programs.
Analysing good governance
A new kind of universal
The good governance agenda was an effective response to critics’ charges of the failure of past efforts at financing development: it allowed the IFIs to shift significant responsibility for those failures onto low-income governments while at the same time developing new forms of expertise to respond to the “problem” of governance. In the process, institutional actors did not so much reject their earlier economic universals, as modify and supplement them through a different kind of universal: one that was broader in scope, more symbolic in character, and that combined technical and moral appeals to its authority.
These new universal standards of good governance are defined partly in technical terms. Through their use of public choice theory, Bank and Fund staff have framed the challenges of governance in universal terms, viewing them as the logical outcomes of human self-interest and the difficulties of collective action. The solution for developing states is no different from what (it is assumed) applied to industrialized states many years ago – the development of rules and institutions capable of keeping self-interested tendencies in check. The institutionalist economics literature, which has played a more important role in framing good governance policy in recent years, is somewhat more nuanced, focusing on institutional rather than individual dynamics, and paying more attention to historical and geographical variation. Yet, here as well, the problems of transaction costs and market failure are represented as universal. The 2002 WDR on institutions begins with a discussion of eleventh-century Maghribi trades, suggesting that the challenges that they faced in expanding trade, and the solutions that they found to overcome problems of information and cheating, are parallel to those faced by people everywhere today.45
Given the prominence of such technical economic claims in the good governance agenda, how different is this standardizing strategy from that of the structural adjustment era, which also relied on technical economic universals? If we take a closer look at the kinds of universals that are being deployed through good governance policies, we find several significant differences.
The first of these differences is the broader scope of the new universals, which not only define “sound economics” but also explicitly identify the kinds of political and social institutions needed to achieve such economic goals.46 This is a much messier and more contestable undertaking – one that runs the risk of overstepping the IFIs’ mandate. IMF and World Bank staff have worked hard to police the boundary between “politics” and “economics.” By insisting that only those institutional reforms needed to achieve economic stability or development are appropriate, they have managed the policy in such a way as to “protect [the institution’s] reputation for technical excellence, professionalism and objectivity.”47 Yet, in practice, the move into governance issues has further blurred these distinctions. As Michel Camdessus noted in an interview,
We had to promote a shift in the mentality of staff members. Many saw the mission of the Fund as technical, and were not comfortable mixing a more technical emphasis on problems such as inflation with other policies focusing on human living standards. The challenge was to bring both together in a new approach.
Eventually, Camdessus suggests, staff grew more accepting of the need to move into this new terrain.48 The 1996 IMF staff paper on governance, as well as the later staff guidelines, simultaneously argue for defining the Fund’s role in terms of the macroeconomic consequences of governance, while also noting the difficulty in clearly separating political from economic dimensions.49 World Bank Executive Directors also noted the difficulties in separating politics from economics: as the UK Director, David Peretz, put it, “There is always going to be some ambiguity, some borderline cases.”50
The second major difference from structural adjustment-era universals is the form that good governance standards take. Because good governance is more complex and pushes the boundaries of the IFIs’ (formally) apolitical mandates, these standards tend to be more visibly constructed, even arbitrary, and therefore contestable. Of course, the “rules” of sound economic practice are also social constructs, built up over centuries as the institutions of the market economy were developed and propagated and homo economicus was formed.51 Yet these economic norms have been black-boxed: naturalized as expert knowledge and integrated into everyday life, so that the principles of low inflation and independent central banks now appear as law-like propositions that are natural facts rather than symbolic constructions.
The principles underpinning good governance, in contrast, are less commonsensical and more obviously Western in their origins. IFI actors advocating their adoption face more vigorous opposition and are more aware of the need to actively define, construct and police the boundaries of these new governance standards. These are standards of good practice, rather than fundamental rules: they are therefore more clearly contingent in their construction, making visible their character as what STS scholars, Timmermans and Berg, describe as “local universals,” a term that emphasizes the way that “universality always rests on real-time work, and emerges from localized processes of negotiations and pre-existing institutional, infrastructural, and material relations.”52 As these specific, localized standards are generalized and universalized, however, they lose contact with the particular Western liberal contexts in which they were produced and become increasingly symbolic, even arbitrary.
The strategies that IFI actors have used to justify these new, more symbolic and visibly constructed standards have taken two different forms: on the one hand, as I discussed above, IFI staff have sought to justify them in technical terms through the use of public choice theory and new institutionalist economics. On the other, they have begun to justify the good governance agenda in explicitly moral terms. This moral turn may appear self-defeating in a context in which actors are supposed to be objective and apolitical. In fact, by framing good governance standards in universal moral terms, the IFIs are working to move these standards into a category that is beyond dispute – not because they are technical, but because they are universal goods.53
Both Fund and Bank leaders have made strong moral claims for the importance of the good governance agenda. The quotation that begins this section makes that clear: when Wolfowitz claimed in 2006 that good governance requires “a fundamental willingness to do the right thing,” he was making an explicitly normative argument regarding the universal “goodness” of the good governance agenda. Although many of the universal moral principles being alluded to are largely drawn from industrialized, Western economies, the claim that the “problems of governance are universal” suggests that this is not a policy directed only at developing economies, but rather at a global challenge.54 Executive Board discussions make it clear that it was important to some members that governance not be seen as “a peculiarly African problem,” but rather that “the principles and practices are universal or ought to be so.”55
This combination of moral and technical universalism has several interesting effects: although the technical logic remains the predominant one, it is supplemented by a thicker set of universals, enhancing the basis of the institutions’ claims to legitimacy. This thicker set of universalist claims provides a more robust foundation for expanding the institutions’ mandate to include increasingly contested and politically charged areas in their programs, thus expanding the basis of their authority.
New actors and sites of authority
The shift towards broader, messier and more self-consciously constructed universals entailed in the good governance agenda has led to the enrolment of new state, market and civil society actors, and the redrawing of the boundaries that separate them. Although these new standards do intervene more closely into the workings of borrowing states, they do not seek to render local actors “docile,” to again borrow a term from Timmermans and Berg.56 Instead, IFI staff have sought to encourage the engagement of more proactive, reflexive local actors in the process of governance: citizens and market participants who will demand better governance, and government actors who will respond. This strategy distributes the authority for governing more widely to a range of new actors, while simultaneously reconstituting them in particular terms.
One of the most striking aspects of the changes taking place in IMF and World Bank policies and pronouncements over the past decade has been their renewed interest in the state and government actors – after several decades of denigrating their role. Yet that attention has retained a certain scepticism about state actors, and a belief that their role should remain secondary to that of the market. The 1997 WDR, The State in a Changing World, carefully differentiates the renewed emphasis on the state from earlier state-led development efforts in the 1950s and 1960s. In the 1990s, the report argued, developing-country government actors could only become effective if they first pared down the role of the state by shifting some of its “burdens” to the private sector and to local communities, and made “the state’s central institutions work better.”57
These policies seek to redistribute some governance authority by enrolling state functionaries in the development process – but only insofar as those government actors learn to behave in certain ways. The World Bank in particular has expended an enormous amount of energy over the past decades in fostering Western-style public services in borrowing countries.58 The goal is not only to change public-sector management practices but also to transform bureaucratic actors by fostering a different kind of culture. As I will discuss in Chapter 8, in the context of results-based management, many of these efforts are inspired by new public management theory, which is based on public choice assumptions and seeks to make bureaucracies work more like markets. Drawing on these ideas, World Bank and donor staff seek to transform developing-country bureaucrats into more efficient, accountable and market-like actors.
One of the key ways of cultivating this new kind of bureaucratic actor is by creating checks on their actions, which reduces the potential for corruption among public officials – a classic concern in public choice theory.59 These checks can take the form of rules, procedures and internal audits, but also include other actors: specifically, active citizens and other non-state actors. The World Bank has long been interested in using market actors and forces to check government action – that is part of the logic underpinning decades of privatization and efforts to build up a strong private sector. The logic behind these new initiatives is, however, somewhat different: there is a genuine attempt in the demand-side initiative to encourage both market actors and citizens to press for better governance. Civil society actors are also therefore granted a degree of authority and responsibility in the practice of fostering better governance. Yet these are citizens of a peculiar kind: “citizen-consumers” who are both consumers of basic services, such as water or health care, and citizens of a state responsible for the provision of those services.60 Their identity as public actors is linked to their role as private consumers. The architects of the good governance agenda thus seek not only to enrol a range of new actors in the practices of governance, but also to reconstitute them as more responsible and active participants.
New techniques: governing through universals
Bank and IMF staff use a range of new techniques in their operationalization of good governance practices. Some of these are similar to techniques used for structural adjustment programs, while others are far more innovative, experimenting with ways of generating popular support for governance reforms.
Since the mid-1970s, the World Bank has relied on the CPIA to help establish the level of funds that the poorest countries receive from the IDA, the World Bank’s concessional financing facility. The CPIA translates countries’ performances in a wide range of areas into indices, which it ultimately reduces to a single numerical score: a classic technique for making disparate contexts commensurable with reference to a set of universally applicable criteria.61 The CPIA initially focused on macroeconomic criteria, but starting in 1997 the Bank introduced governance criteria into the mix, which now account for 68 per cent of the final score.62 At the IMF, more traditional techniques include the inclusion of governance concerns in Article IV consultations and the addition of governance conditions in stand-by agreements and concessional loans.
The good governance agenda has also led to some innovative practices. At the IMF, the chief example of this is the standards and codes initiative, which I will discuss in the second half of this chapter. At the World Bank, this kind of innovation is most apparent in the institution’s move to foster the “demand for good governance” (DFGG) at the country level, through the development of new technologies of community.63
Bank staff have identified several practices as key to supporting DFGG: transparency and the dissemination of information, consultation and participation, and ongoing monitoring and evaluation. These techniques should be familiar to anyone who has examined liberal conceptions of the public sphere as a site where individuals engage in publicity (typically through a free press) and debate. Bank staff thus seek to mobilize popular support for good governance in a way that combines liberal normative assumptions about publicity and participation with public choice ideas about the need to develop checks on government excess. In the process, deliberation is transformed into a kind of thin participation that is somewhere between consumer feedback and citizen consultation.
If citizen-consumers are to act as checks on government, they need information about what government actors are doing. The Bank’s goal is not simply to engage public actors in the initial formulation of policies, but rather to create mechanisms through which they can monitor, evaluate and report on government policies on an ongoing basis.64 For this to happen, it is necessary to gather information about the effects of those policies, to assess them against predefined indicators, and to communicate them to the public. This involves the creation of new kinds of inscriptions. These can include Doing Business indicators, a Bank initiative that scores countries on how easy it is to set up a business, complaints mechanisms, media investigations, and “citizen report cards,” in which civil society organizations grade public services.65 Transparency, participation and monitoring thus come together, as information on service performance is transmitted to encourage public actors to “voice” their views, producing data that is in turn used to improve service delivery.
While such thin public practices could potentially spill over into thicker, more genuinely political activities, this possibility is constrained by the tendency to frame civil society actors as consumers of services first and foremost. Narrowly economic forms of consultation such as obtaining customer feedback thus come to redefine and constrain activities that might have produced more political kinds of engagement.
New forms of power and authority
Any efforts to reshape the policies and institutions of developing-country governments clearly involve power relations, whatever claims the IFIs may make to the contrary. The IMF and World Bank have always relied on a range of forms of power, from the more coercive power of traditional conditional lending to the informal power of technical advice and assistance. The incorporation of governance criteria into the CPIA at the World Bank and of governance conditions into loans at the IMF are examples of how the good governance agenda has extended the IFIs’ more traditional coercive forms of power. Yet when World Bank executive directors discussed the institution’s initial move into good governance, they were explicit about the importance of using “persuasion” rather than “constraint” as the key tool for achieving objectives.66 Efforts to develop new standards of good governance have made the IFIs more reliant on productive, indirect and proactive forms of power.
As Barnett and Finnemore suggest, IOs like the IMF and World Bank have always exerted a kind of productive power that uses their capacity to define and categorize objects of governance in order to give them real meaning and presence.67 The idea of good governance is a classic example of a term whose invention has had performative effects by making possible a range of practices and interventions that would not have been possible before. “Good governance” can be seen as an extension of earlier such categories, like “sound economics,” which have been used for much longer. Yet whereas calls for sound economics once defined state and market actions in largely negative terms, as a matter of deregulating and liberalizing, the category of good governance seeks to define far more explicitly – and positively – the role of government, civil society and market actors.
Thus the IFIs aim to create important changes in the organization of low-income governments and their relationship with the market and citizens by creating the conditions in which others (chiefly citizen-consumers) will demand those changes. This is a kind of indirect power that relies on information and transparency to achieve its ends. Not only is the form of this strategy somewhat unusual, but its goal is also novel, at least in the developing world: the attention to market and popular demand together with the emphasis on transparency and accountability makes it clear that the objective is to create what Mitchell Dean, drawing on Peter Miller, has described as a kind of reflexive government:
The imperative of reflexive government is to render governmental institutions and mechanisms . . . efficient, accountable, transparent and democratic by the employment of technologies of performance such as the various forms of auditing and the financial instruments of accounting, by the devolution of budgets, and by the establishment of calculating individuals and calculable spaces.68
Michel Foucault’s discussions of the different forms that power take are useful here: the universalist strategy that underpinned structural adjustment operated both like a juridical rule, dictating orthodox economic practice, and like a form of discipline, differentiating normal economics from abnormal forms.69 The standardizing practice that underpins the turn to benchmarks, standards and best practices shares with the structural adjustment era a disciplinary logic, as it sorts economies into normal and abnormal. But it also increasingly relies on a more governmental form of power, focusing on managing circulations around the norm (rather than drawing lines between what is normal and what is not), and seeking to foster a more active, self-disciplining kind of subjectivity among the bureaucratic, market and civil society actors that it enrols.
The form that power takes here is not one of domination, since the goal is not to produce docile subjects, but to enrol a wide range of actors in the process of demanding and providing better governance. Yet these less direct mechanisms still work to generate forms of exclusion and inequality. Standards may appear inclusive but, as Timmermans and Epstein put it, “every standard inevitably implies an evaluation at the expense of some other, often obfuscated, devaluation.”70 In the case of good governance standards – and even more clearly with the standards and codes initiative, as I will discuss below – those kinds of institutions and practices of governance that have been deemed “good” are almost entirely drawn from Western, liberal, free-market societies. The patterns of inclusion and exclusion enabled by standards are therefore subtle: countries are graded and ranked, rather than simply excluded. Moreover, because their “grade” translates, through the CPIA and performance-based conditions, into different levels of funding, this ranking has very real consequences for the poorest in the world.71
Standards and codes
Although the good governance agenda is the most prominent of recent efforts to develop global economic standards, it is not the only such policy. By taking a look at a second, related, policy change – the development of standards and codes – we can gain a more concrete sense of how the IFIs have deployed the strategy of standardization in practice. Whereas the World Bank was the main driver of the good governance policy, here it is the IMF that has been in the driving seat.
Developing the initiative
The standards and codes initiative ultimately became the centrepiece of the IFIs’ response to the financial crises of the 1990s.72 Over time, the standards also came to be viewed, particularly by the IMF, as their central contribution to the spread of good governance practices. Yet the first standards were not developed with such grand objectives in mind. In fact, the standards and codes initiative evolved gradually, eventually taking on the central role that it plays today. By tracing the initiative’s development, we can also track the evolution of thinking about the role of these standards and recognize the choices that were made in pursuing this path to financial stability and good governance – choices that reflected a desire to govern more provisionally.
The standards and codes initiative was developed in two major stages in response to two major crises – the Mexican and Asian financial crises. While external critics saw these two crises as evidence of a profound failure of the IFIs’ efforts, IMF staff and management interpreted them rather differently: as evidence of a more modest failure in the information that governments made available to market actors, and of a more serious failure in borrowing countries’ institutional quality. It was not markets (or the IMF), but states that were to blame for these failures, they argued, due to their poor institutional capacity and failure to provide the markets with timely and accurate information. In debating the nature of these failures and the appropriate responses, IMF staff and management thus began to problematize the role of information and transparency in domestic governance, both of which played a central role in the global standards that they developed in response.
The first stage in this standardization strategy was the IMF’s development of a highly technical and, at least on the surface, unexciting set of standards for statistical information: the special data dissemination standard (SDDS). Reacting to G7 calls for action in the wake of the 1994 Mexican crisis, the IMF created the SDDS to encourage member countries to begin publishing statistics on their economies in a standardized and timely manner.73 The IMF, the G7 and others saw the poor quality of Mexican data as one of the causes of the crisis, and believed that better data would improve market confidence. Shortly after creating the SDDS, which was designed for countries that were able to borrow from international financial markets, the Fund created a second general data dissemination standard (GDDS), which gave poorer states an incentive to develop their statistical capacities and publish the data that they obtained. Central to both standards is the technique of publicity: the strategy not only seeks to get countries to regularly publish their statistics, but also publicizes countries’ compliance with the standards through an electronic bulletin board.
This same preoccupation with transparency and publicity characterized the second stage in the development of this standardizing strategy. In 1997, the IMF Executive Board sat down to discuss two staff papers, one on “Transparency in Government Operations” and the other on “Fiscal Policy Rules.” The first of these papers discussed the value of transparency – not only in the provision of statistical information but also in the everyday fiscal activities of a state.74 The second document focused on the usefulness of fiscal policy rules – such as balanced budget rules, or a maximum budget deficit threshold such as the 3 per cent of gross domestic product (GDP) limit in the Maastricht treaty.75 In their discussions, IMF staff and directors focused on the problems of the kind of “creative accounting” used by some European Union countries in order to meet Maastricht rules. They expressed concerns about unintentional opacities – when a government does not have the capacity to provide the necessary information – and intentional forms, in which a government attempts to “escape public scrutiny of its behaviour – especially in the run-up to elections – in order to avoid or postpone possible adverse reaction from the electorate and from financial markets.”76
The outcome of these discussions was the next step in the creation of the standards and codes initiative. Although executive board members considered the possibility of integrating fiscal rules into their programs, they decided instead to take a less direct and more flexible approach to transparency. At an October 1997 meeting, the board asked Fund staff to compile current best practices into a manual that would be available to all members. By March of the following year, pressure from the G7 and various government leaders for more explicit guidelines was strong enough that the Fiscal Affairs Department drafted a more comprehensive code of conduct on fiscal transparency. The code emphasized the importance of providing timely and accurate information on the budget to the public, ensuring that the budgeting process was open and that there were independent audits of public accounts.77 Shortly after the code on fiscal transparency was approved in 1998, a second code on monetary and financial policies was developed.78
The IMF Board and staff gradually developed techniques for monitoring compliance with the codes.79 This process initially took the form of experimental case studies, but eventually grew into the more standardized reports on observance of standards and codes (ROSCs). Like the SDDS and the GDDS, the ROSCs are based on the principles of voluntary compliance, publicity and market discipline. Both the adoption of the standards and codes and the publication of the ROSCs are voluntary. Staff and directors believed that peer pressure and a desire for market approbation would lead governments to adopt the standards and codes, and publish information on their compliance. They hoped that markets would provide a further crucial incentive by rewarding compliant states with lower borrowing costs.
The list of standards and codes continued to grow, ultimately including twelve different issue areas covering everything from accounting standards to bank supervision and the prevention of money laundering.80 The World Bank and the Fund together have responsibility for overseeing their implementation. What initially began as a rather modest effort to reform statistical capacities has thus grown into a vast array of standards covering a wide range of different aspects of political economic life.
An analysis of standards and codes
A new kind of universal
The standards and codes initiative, like the good governance agenda, is a universalist strategy that seeks to promote new global standards of good political economic practice. Yet, once again, when we look closely, we find a visibly constructed universal at work, broader and more flexible than the rigid economic rules that defined the structural adjustment era.
As I discussed in the previous chapter, a particular small “i” idea – credibility theory – was highly influential when the IMF Board was debating whether and how to develop new standards of economic practice.81 Advocates of credibility theory and the time inconsistency problem argue that as long as governments are able to revise the policies that they committed to earlier, they will not be seen as credible by market actors, and will therefore find their efforts at constructive market intervention undermined; in such circumstances, the only credible alternative is for the government to relinquish much of its policy discretion and commit to binding rules.82 This is seen as a universally applicable axiom, since it is assumed that all rational market participants will recognize good or bad economic policy, and therefore judge the credibility of a government’s actions accordingly.
The fiscal rules paper that the IMF Board discussed was particularly explicit in its reliance on both public choice and credibility theory to justify the application of universal rules that would bind governments’ hands.83 In the end, the Board decided not to adopt fiscal rules as their primary technique for fostering standardization. Yet the logic of credibility remained central even as they developed more flexible techniques for achieving their ends. Staff believed that fiscal transparency and the publication of the ROSCs would have a similar effect on credibility, by providing markets and the public with additional information on fiscal plans, reducing governments’ leeway to back-track on policies, and thus increasing their credibility.84
Although technical expertise served as the primary grounds for justifying the universality of these new standards, this was a different universal claim from those underpinning the economic rules of the structural adjustment era. Credibility theory relies on a more contingent kind of universal from those in traditional neoclassical theory – or rather, credibility theory makes the symbolic and performative dimensions of economic universals more visible. As I noted in the previous chapter, the very acts by market participants that are supposed to be signals of objectively credible or non-credible policy actually create or undermine policy credibility.85 For example, an inward flow of foreign capital, which is often seen as a sign that a government’s policies are credible, actually generates that credibility by providing the government with resources that enable it to keep its promises, while an outward flow of capital has the opposite effect. Similarly, the role of external experts who are often imported to play key roles in monetary institutions do not simply verify and communicate a government’s latent credibility, but actually help to create it through their presence as symbolic markers of Western economic expertise.
These standards and codes are not only more performative but also far broader in their scope than past economic norms – defining best practices in the fields of auditing and accounting, corporate governance, money laundering and terrorist financing, to name a few. In their effort to justify a move into these new domains, the IMF leadership, like the World Bank, moved beyond a narrowly technical discourse and began to frame the policy in moral terms. Horst Köhler, the managing director of the IMF during the early years of the initiative, suggested “While standards and codes deal with highly technical matters, there is nothing narrow or technical about their purposes.”86 Michel Camdessus, Köhler’s predecessor, also suggested that universal standards could help to “civilize globalization” by creating new “rules of the game” to tame the wilder excesses of the global economy. Given the interdependent character of that global economy, Camdessus argued, “a duty of universal responsibility is incumbent upon all. Every country, large or small, is responsible for the stability and quality of the entire world growth.”87 These new universals are therefore not only more symbolic and performative than their predecessors, but also more complex in their justification, relying on both technical and moral appeals to their universality.
More performative techniques
An interesting aspect of the standards and codes initiative is the way in which its creators sought to put the policy into practice through new and innovative techniques. The key technique at the heart of the standards and codes initiative is the ROSC. This technique was developed through an experimental process in which IMF staff undertook several case studies in a range of different countries, published their findings, and solicited public feedback. The IMF Board initially deemed these “experimental ROSCs” a success, but continued to advocate a “gradual and interactive approach” in developing the Fund’s involvement in standards.88
As their name suggests, ROSCs are reports: they are inscriptions that seek to translate the messiness of a country’s strengths and weaknesses into a single document. They were intended to establish “best practices” in a range of areas from corporate governance to banking regulation, and to benchmark individual countries’ performance in each of these areas. Part of the goal behind these benchmarks is to enable governments to assess their own progress over time. Once the ROSCs are published, however, their role shifts significantly and credibility becomes key, as an early IMF report notes:
By highlighting actual practices and identifying those in need of improvement, transparency reports could reduce the likelihood that market participants are uninformed or misled, and enhance the credibility of those national authorities following sound practices.89
The role of ROSCs as inscriptions is thus not only to translate complex economic and political actions into a single document, but also to signal good (or bad) economic practice to market actors. Like the conditions I discussed in the last chapter, standards and codes are not only important because of what they are (guidelines for best practice) but because of what they signal through the ROSCs. Their value is increasingly symbolic and it is through that symbolic role that they do their most important work: IMF staff and management believed that this signalling would have significant material effects, as those interested in buying government bonds would pay attention to ROSCs and reward good economic performers with lower borrowing costs.90
Although many policies based on credibility theory, such as fiscal policy rules, ignore or downplay the performative character of the policy, advocates of ROSCs were highly reflexive about the performative nature of this kind of signalling device, seeing the reports not simply as descriptions of an objective reality but as tools for creating the kind of self-fulfilling cycle that pushes countries towards more “credible” policies.
New actors and sites of authority
These new techniques rely for their effectiveness on the active involvement of a range of actors: not only the IFI staff who conduct the evaluations, or the member government actors who are to use the benchmarks established, but also civil society and market actors who, it is hoped, will pay attention to the ROSCs and reward good performers.
As in practices to foster good governance, information, publicity and transparency are the key mechanisms through which these new actors are to be informed and enrolled. As I have discussed elsewhere, although a policy of transparency may appear to be minimalist, its objectives are not.91 Take the two data standards discussed earlier: although the SDDS and the GDDS seem highly specialized and technical, they are important for several reasons. They frame the problem of economic governance in terms of the quality of information. They also seek to constitute a particular capacity to obtain and communicate that information, particularly among emerging and developing countries.92 And in providing that information they also hope to influence not just the actions of government, but also those of civil society and market actors, by providing them with the information that they need to keep the government in check. This same logic underpins the other standards and codes. As the introduction to an early draft of the code on fiscal transparency put it:
Increased fiscal transparency should lead to better-informed public debate about the design and results of fiscal policy, make governments more accountable for fiscal policy and management, and thereby strengthen credibility as well as mobilized popular support for sound macroeconomic policies.93
The standards and codes initiative thus seeks, like the good governance agenda, to redefine the relationship among state, market and public actors. Here again we see a new emphasis on the demand side of governance, and an effort to redistribute some governance authority based on the belief that market and public actors can pressure governments to act in certain ways. Although one might argue that these assumptions about the superior rationality of the market and the public are rather naïve, it is important to note that the goal of fiscal transparency is not simply to rely on existing public actors but also to educate them about the budget, thus constituting a more informed and active public citizenry. This belief in the importance of external scrutiny is key to the particular form of power that the standard and codes initiative relies on.
More indirect power
What is perhaps most striking about the standards and codes initiative is its almost exclusive use of indirect forms of power. Although there were wide-ranging debates both within and outside the IMF about whether the new standards and codes should be mandatory or voluntary, those who believed in the power of peer pressure ultimately won the day.94,95 Underlying this informal strategy was the assumption that it was in states’ interest to adopt the new standards and codes. Thus, Köhler argued, “While it is still early in the game, there is already evidence that meeting standards can pay off,” as investors reward good behaviour with lower borrowing costs.96 No need for the IMF or World Bank to use the blunter instrument of conditionality to achieve their desired goals.
The forms of power involved in making the standards and codes initiative work are also clearly productive: the goal is not simply to make government actors behave in certain ways, but to give them new tools – statistical capacities, best practices and benchmarks – to enable them to see and calculate their actions and objectives in new ways. In Michel Callon’s terms, the strategy is one that seeks to generate a particular kind of economic society by fostering new calculative capacities.97 This is therefore also a proactive strategy: while transparency is partly an end in itself, more importantly it is a means to encourage market and civil society to pressure governments to achieve longer-term goals of economic policy change.
Who gets to decide what count as “best practices” of fiscal and monetary policy? As the initiative’s critics – academics, NGOs and emerging market leaders – were quick to point out, the process for establishing these news standards was exclusive, and largely based on Western, free-market principles.98 Moreover, as Ilene Grabel notes, although scholars and policymakers generally insist that credible policies (low inflation, low taxes, minimal government involvement in the economy) are timeless and universal, they are instead based on neoliberal values, and tend to support a particular set of economic interests. The process of fixing these particular economic values as the standard also devalues alternative forms of political and economic policy as non-credible and therefore not viable.99 Although the mechanisms through which these valuations of better and worse policy are therefore far subtler than a fiscal rule or strict conditions, their effects are nonetheless to grade and rank country practices according to Western norms.
A more provisional kind of governance
Both the good governance agenda and the standards and codes initiative involve a significant shift in how the IFIs govern through, and in the name of, universals: the universal economic rules of the structural adjustment era have been supplemented by a more reflexively constructed and justified set of global standards. In the process, universals based on technical expertise have been combined with moral and popular ones, new actors have been enrolled and authorized to play a role in governance, more performative techniques have been developed, and more indirect and productive forms of power deployed.
As standardization has gradually colonized new terrains of global life, so has the influence of technical expertise. Yet the evolution of this new strategy of standardization has been far messier and fragile than a narrow focus on technical expertise would indicate: IFI leaders have had to bolster their technical justifications for the standards with normative claims, changing the character of the standards themselves in the process.100 Although we might expect a strategy of standardization to be a classic example of the spread of technical expertise to new areas of social, political and economic life, what we find is a more complex picture, in which governance and the expertise that underpins it have changed in form and become more provisional: more proactive, indirect, symbolic and aware of the possibility of failure.
The IMF and World Bank Executive Boards both considered and rejected the option of enforcing these new standards using more direct techniques, opting instead for an indirect approach that works through the publication of performative inscriptions and the pressure of market actors, civil society and peer states to achieve its ends. The techniques involved are also proactive, playing the long game: both policies seek to use these less direct pressures to change the cultures of borrowing country governments, markets and civil societies, giving them new tools and new incentives to pursue the changes sought by the IFIs.
These standards are not only more flexible than the economic rules that they supplement, but also more visibly constructed and symbolic in form. Because their subject matter is more complex and contested, the assumptions and biases that go into a standard of good governance or transparent fiscal policy are more visible than those that underpin the rules of sound economic policy. They are therefore harder to naturalize and black-box. At the same time, the role played by these standards is increasingly symbolic – aimed at signalling credible government intentions, whether to civil society actors or market investors. Whether in the form of the Doing Business Indicators, government “report cards” or ROSCs, these policies all rely on a kind of inscription designed to translate complex on-the-ground realities into a signal of good or bad state behaviour. In a context in which transparency, information and publicity are key techniques, it is this signal that does the real work of encouraging further government compliance: enrolling external actors in the business of pressuring, punishing and rewarding governments.
These new governance practices were developed in response to particular interpretations of the failures of international development and finance. At the same time, the provisional character of the standardization strategy enables IFI staff and leaders to avoid certain kinds of failure. The indirectness of these governance practices creates considerable distance (and thus deniability) between IFI policies and their effects. Moreover, the policies’ proactive effort to give more authority to governments for ensuring their own compliance with the standards also transfers to borrowing countries much of the responsibility for failure.
In spite of these pre-emptive tactics, IFI staff and management continue to be aware of and preoccupied with the strategy’s limits. I will discuss the challenges faced by the standardization strategy at some length in the Conclusion to this volume. They include the gradual unravelling of the compromise that these standards represent – with some seeking to return to hard and fast rules, while others argue for an even more contextual and flexible interpretation of governance norms. These challenges also include some very real practical difficulties in the operationalization of the inscriptions as, for example, the market actors who were supposed to pay attention to the ROSCs have tended to ignore them entirely, thus eroding their performative effectiveness. These limits have in turn ensured that the strategy’s advocates have had a hard time black-boxing these standards, leaving them open to ongoing contestation.
While the standardization strategy remains highly influential today, neither it, nor the strategy of ownership, have managed to fully resolve the problems of institutional authority posed by the contested crises of the 1990s. These two early provisional strategies have been followed, however, by two newer strategies designed to tackle more explicitly the epistemological and ontological problems of earlier governance practices. By measuring results, IFI actors are developing a new epistemological foundation for assessing development financing. And by managing risk and vulnerability, the subject of the next chapter, they are developing a new ontology of poverty, debt and the unknown.