Agencies and policies

Contents

1. Introduction

2. Paving the way: Some facts on corruption and development
2.1. Defining corruption
2.2. Measuring corruption
2.3. Corruption and development
2.4. The role of donors in fighting corruption

3. Towards an AC regime in bilateral DC
3.1. A few words on regimes
3.2. What could the regime look like?
3.3. Principles, rules, norms, procedures
3.3.1. Principles
3.3.2. Rules and Norms
3.3.2.1. Complexity and timing
3.3.2.2. Mainstreaming AC efforts in donor agencies
3.3.2.3. Supply-side issues, domestic advocacy
3.3.2.4. Knowledge management, evaluation
3.3.2.5. Country specificity of AC actions
3.3.2.6. Coalition building
3.3.2.7. Entry Points, sectoral approaches
3.3.2.8. Strengthening Civil Society
3.3.2.9. Support decentralization and local participation
3.3.2.10. Political issues
3.3.3. Decision making procedures

4. The normative level: Three AC policy papers examined
4.1. USAID
4.1.1. Complexity and timing
4.1.2. Mainstreaming AC efforts in donor agencies
4.1.3. Supply-side issues, domestic advocacy
4.1.4. Knowledge Management, Evaluation
4.1.5. Country specificity of AC actions
4.1.6. Coalition building
4.1.7. Entry points, sectoral approaches
4.1.8. Strengthening Civil Society
4.1.9. Support decentralization and local participation
4.1.10. Political issues
4.1.11. Summary for USAID
4.2. DFID
4.2.1. Complexity and timing
4.2.2. Mainstreaming AC efforts in donor agencies
4.2.3. Supply-side issues, domestic advocacy
4.2.4. Knowledge Management, Evaluation
4.2.5. Country specificity of AC actions
4.2.6. Coalition building
4.2.7. Entry points, sectoral approaches
4.2.8. Strengthening Civil Society
4.2.9. Support decentralization and local participation
4.2.10. Political issues
4.2.11. Summary for DFID
4.3. Sida
4.3.1. Complexity and timing
4.3.2. Mainstreaming AC efforts in donor agencies
4.3.3. Supply-side issues, domestic advocacy
4.3.4. Knowledge Management, Evaluation
4.3.5. Country specificity of AC actions
4.3.6. Coalition building
4.3.7. Entry points, sectoral approaches
4.3.8. Strengthening Civil Society
4.3.9. Support decentralization and local participation
4.3.10. Political issues
4.3.11. Summary for Sida
4.4. Summary of the results on the normative level

5. The implementation level: aid allocation patterns compared
5.1. USA
5.2. UK
5.3. Sweden
5.4. Summary of the results on the implementation level

6. Conclusion

7. References

8. Abbreviations

List of tables

Table 1: Important OECD documents on corruption

Table 2: USAID, DFID and Sida compared

Table 3: Degree of similarities to OECD recommendations

Table 4: The USA’s bilateral ODA in

Table 5: The UK’s bilateral ODA in

Table 6: Sweden’s bilateral ODA in

Table 7: Results summarized

1. Introduction

In this book, I want to examine how bilateral donors perform in fighting corruption. Taking an actor-centred, policy-oriented approach I focus on three selected bilateral agencies that are heavy-weights in the aid scene: the United States Agency for International Development (USAID), the British Department for International Development (DFID) and the Swedish International Development Cooperation Agency (Sida). ^{[1] [2]}

Each of the three agencies that I focus on faces corruption individually. At the same time corruption remains a common problem, given its negative consequences on development. In this book, I take a rationalist perspective. I argue that donors perform well in fighting corruption when they cooperate. In other words: when donors try to fight corruption individually in their target countries, they risk a suboptimal outcome on the global scale - or as the Organisation for Economic Co-operation and Development (OECD) phrases it: “The risks associated with a piecemeal response, in which various donor organisations act in a deliberate but uncoordinated way, are set to increase.”^[3] Donors thus face a problem of collective action, as fragmentation of anticorruption (AC) work is often hindering success.

Regime theory provides tools to overcome problems of collective action. When states establish a regime in a certain issue area, they do so to better pursue their own individual and rational interests. In the current case, it is in the bilateral donors´ interest that their official development assistance (ODA) is used as effectively as possible. To achieve this, they need to fight corruption in their partner countries and at home in a coordinated way.

Thus the question which this book addresses is: Does current cooperation between bilateral donors constitute a working international regime for fighting corruption in bilateral development cooperation?^[4]

In order to answer this question, I use the following variables: the cooperation in the field of anti-corruption in bilateral development cooperation will be the independent variable, whereas the existence of a regime in this field will be the dependent variable. Other possible independent variables such as the power distribution in the international system shall be regarded as fixed. This assumption shall be permitted as the idea of a hegemonic state in a regime entirely devoted to more or less altruistic development cooperation seems to be somewhat absurd.

To operationalize the independent variable, I take two steps. In the first step, I take a look at the normative level in examining key policy papers of the three donors. My assumption is the following: states cooperate when they all incorporate the same contemporary policy thinking in their papers. To define contemporary policy thinking, I use the work of OECD/DAC.^[5] This group's documents, which draw upon the earlier work of the World Bank and Transparency International, reflect the smallest common denominator of ideas of the most important donor countries.

I realize that by focussing on the OECD/DAC countries, I leave out a fraction of donors that are not members of this group. According to Browne^[6] about one tenth of total ODA – perhaps between 5 billion USD and 10 billion USD per year – is accounted for by donors of the south who are not members of the OECD/DAC. The largest donors from this group are China and India.

Yet leaving these donors out of the picture seems to be justified when theorizing about a possible AC regime in development cooperation. Both India and China are not signatories of the Paris Declaration on Aid Effectiveness. For the moment they do not seem to be interested in finding any regime on aid effectiveness or fighting corruption. It remains to be seen if the UN's new bi-annual Development Cooperation Forum (DCF)^[7] can bridge the gap between OECD/DAC donors and the donors from the south.

To sharpen the picture of the current AC state of the art in bilateral DC, I have added recent insights from scholars to better reflect the edge of academic reasoning in the field. My assumption: if donors incorporate all the core findings from the OECD papers as well as fresh academic insight simultaneously in their policy papers, they cooperate. For every donor, I map the concordance between the agency's strategy and the OECD recommendations on an ordinal scale (high, middle, low).

In a second step, I look at how the policies are actually applied by taking a look at the aid allocation patterns of the three donors at issue. I examine whether similarities between donors can be found in this case. Through coordinated actions at the country level donors could avoid supplying aid to the most corrupt states. In doing so, they would show that their AC policies are more than just lip service. Only in this case can a working regime develop.

To operationalize the dependent variable, I need to define what regimes actually are. I use Krasner's classical definition of regimes as “sets of implicit or explicit principles, norms, rules, and decision-making procedures around which actors’ expectations converge in a given area of international relations”.^[8]

This book is structured in the following way: the section following this introduction provides information on the environment that donors operate in. Here, I deal with problems of defining and quantifying corruption as well as with the influence of corruption and development and the role of donors in fighting corruption. In the section thereafter, I examine whether roles and norms for a possible regime exist by looking at key policy documents of the OECD, USAID, DFID and Sida. After that I turn my attention to the three donors’ aid allocation patterns to see if normative and implementation aspects of fighting corruption coincide. In the last section, I present the key findings of this book and suggest ideas for future research.

2. Paving the way: Some facts on corruption and development

This section seeks to provide some essential information that is needed to understand the complex environment that donors operate in. It starts by explaining that donors still have difficulties to pinpoint what corruption actually is. Despite many attempts there is no universally agreed definition so far. Consequently, some of the most important attempts are briefly presented. The section then continues by explaining a second key problem that donors face: how can corruption actually be quantified? It is shown that measuring corruption is equally challenging as defining it. Following this, the effects of corruptions on development are explored which donors need to take into account when operating in their target countries. The section concludes by presenting some ideas on the particular role of donors in fighting corruption.

2.1. Defining corruption

Corruption is not a new problem. In fact, it seems to be a problem that is as old as the concept of government itself. Klitgaard provides some historical evidence: Even 2300 years ago, he writes, the Brahman Prime Minister of Chandragupta was mentioning “at least fourty” ways of stealing money from the government. And in ancient China, there was even an anti-corruption provision in the public sector: officials were granted “Yang-lien”, a special allowance to “nourish incorruptness.” Nevertheless, Klitgaard writes, “such nourishment often failed to achieve that purpose”.^[9] Ancient Greeks understood corruption as “a deviation from or perversion of sound government systems.”^[10]

There has been a whole battery of definitions of the term corruption in the AC literature of the recent years. Although I do not want to enter too deep into this – still unfinished – discussion, I nevertheless would like to outline a small number of influential attempts to define the term. My main aim here is to illustrate that the notion of corruption still remains somewhat unclear, even after at least two decades of intensive research. One of the fundamental definitions of corruption has been undertaken by Klitgaard in the late 1980s. He used a principle-agent-model to define corruption “in terms of divergence between the principal's or the public's interests and those of the agent or civil servant: corruption occurs when an agent betrays the principal's interests in pursuit of their own.” Klitgaard has summarized his findings in a “stylized equation” of corruption: Corruption equals Monopoly plus Discretion minus Accountability. This definition has been very influential and formed the working-base base for a larger number of scholars and institutions. Among others, the World Bank based its anti-corruption work in the late 1990s on this approach.^[11] Yet scholars have noted that there seems to be a flaw in Klitgaard's definition: this is due to the fact that his definition requires that the principle is not corrupt himself: “This is often not a reasonable assumption in the most corrupt countries.”^[12]

Transparency International (TI) defines corruption as “the misuse of entrusted power for private gain”. This very broad definition thus encompasses corruption in both the private and the public sector as well as in civil society. OECD/DAC has used the following definition: “Corruption is usually defined as the abuse of public office for private gain, however, it is most helpfully viewed as an outcome, the consequence of any of a number of the failures in accountability relationships which characterise a national governance system.”^[13]

Aside from the attempts to define the term that have been presented here, a large set of others exist. From the three donors that I focus on, only two – USAID and Sida – have defined corruption in their policy papers (see sections 4.1 and 4.3. of this paper). DFID does not use a particular definition, neither does the United Nations Convention against Corruption (UNCAC). Yet, a large amount of scholarly and practically oriented literature proves that one can fruitfully work in the field without defining the term. This is also what I shall try to do.

2.2. Measuring corruption

Since there is rarely hard evidence of corruption taking place in the public, donors face a measurement problem: how can reliable data on corruption be obtained? The solution has been to collect information about people’s personal perceptions of the extent of the corruption problem: “How much corruption do people think is going on?”^[14] This data is usually gathered from people that live in the country under review, from businesses that operate there and from country analysts that are in a position to assess and compare the situation in different countries. It has been argued though, that indicators which are mainly based on such subjective criteria such as TI's Corruption Perceptions Index (CPI)^[15] and Bribe Payers Index (BPI)^[16] are only of a limited value. On the other hand, “since corruption usually leaves no paper trail, responses about corruption based on individuals' actual experiences are sometimes the best available, and the only, information we have”^[17]. Among others, Soreide has argued that the construction of indexes is key to further improve the understanding of corruption.^[18] It seems to be academic consensus though, that corruption indicators that are based on subjective criteria are of high value, despite some methodological problems. Kaufmann and his colleagues put it in simple words: “corruption can, and is being, measured”^[19] Or as Soreide puts it: “The value of corruption indexes should not be doubted”^[20] – despite the known problems. There are ongoing attempts, though, to find other than perceptions based indicators. Recently, Dreher et al. have proposed a new index that not only derives the rank but also the distance between countries in the corruption index. They believe that “this has the potential to give a more accurate (or less biased) estimate of how the impact of corruption varies across countries and hence the dependent variable of interest.”^[21]

2.3. Corruption and development

There are multiple – and sometimes rather confusing – connections between corruption and development that donors need to consider. Or, as Fjeldstad and Andvig put it: “The ‘causes’ and ‘effects’ of corruption are closely interrelated and can hardly be separated.”^[22] For example, the level of corruption seems to be influenced by the level of development in the country in question. It has been argued that misuse of public office is more likely to be exposed in more economically developed countries as “economic development increases the spread of education, literacy and depersonalised (‘arm’s-length’) relationships”.^[23] Studies have shown a strong correlation between the average income (Gross Domestic Product [GDP] per capita) and a country’s ranking on corruption indexes. But as Lambsdorff explains, no causality between GDP and corruption can be derived from this insight.^[24] So, so for donors the puzzling question remains: “Is a country poor because of corruption, or is a country corrupt because of poverty?”^[25]

Interestingly, there has been a debate about possible beneficial aspects of corruption. In some parts of the literature, one can find the argument that the economic benefits of corruption outweigh its costs in some countries - and that the buying and selling of political favours could have certain political and economic advantages: “One point often made is that bribery ‘greases the wheels’ by cutting red tape, and thus is improving efficiency.”^[26] This efficiency argument is based on the assumption that bribery might actually help to reduce the transaction costs for companies who find themselves in a jungle of public regulation provisions.

According to this line of thinking, donors would not necessarily need to fight corruption because of its possible benefits. But Kaufmann and Wei have deconstructed the efficiency argument as a myth. They found that firms that pay more bribes are nevertheless likely to spend more management time with bureaucrats negotiating regulations – and face higher cost of capital: “In fact, a consistent pattern is that bribery and measures of official harassment are positively correlated across firms.”^[27] That means that donors have no excuse not to fight corruption, since the efficiency argument does not hold.

But which is the exact influence of corruption on the growth of GDP in a particular target country? This question is not easy to answer as various scholars did not manage to find robust correlation between the two. For example, Brunetti et al. observed only insignificant effects.^[28] But how can this missing link between corruption and growth be explained? Lambsdorff suggested that the mixed evidence might be explained by the fact that corruption primarily has impacts on the accumulation of capital, but does not clearly affect the productivity of capital.^[29]

It is important for donors to note that – without any question – corruption has a strong influence on the level of poverty in a country. TI has noted that corruption does not only reduce the net income of poor people “but also wrecks programmes related to their basic needs, from sanitation to education to healthcare. It results in the misallocation of resources to the detriment of poverty reduction programmes.”^[30]

In the recent years, there has been a strong consensus in the academic world that a strong link between good governance and development exists. Thus, donors have sought to promote “good governance” in their target countries: “Today effective governance is recognized as a sine qua non for development and poor governance, in turn, as a contributing factor to poverty, conflict, and state failure, with repercussions for global security.“^[31]

But what actually is “good governance”? Kaufmann defines governance as “the traditions and institutions by which authority in a country is exercised for the common good.” In this definition, he includes three dimensions of governance: a political dimension, being “the process by which those in authority are selected, monitored, and replaced”, a economic dimension, being “the government’s capacity to effectively manage its resources and implement sound policies”, and an institutional respect dimension, being “the respect of citizens and the state for the country’s institutions”.^[32] He argues that “[i]mproving governance and controlling corruption matter enormously for development, and countries can substantially improve, even in the short term, if the appropriate strategy and political resolve are present.”^[33]

Donors promoting good governance often base their policies on research by Burnside and Dollar on aid effectiveness.^[34] They showed that aid has been effective in countries “with good policies, good governance, and less corruption”^[35]. At the same time, aid proved to be much less effective in countries with weak policies and high corruption. Consequently, they argue that “aid would be more effective if greater effort were made to direct it to good policy performers.”^[36] Following these insights, the developmental agenda has increasingly been dominated by the quest not simply to pursue initiatives “that focus on particular activities or sectors, but proposals that also deliver anti-corruption measures through wider reforms to the state.”^[37]

In theory, one would assume that democratic institutions in target countries would make it easier for their citizens, media and watchdog-organisations to denounce cases of corruption. In practice, researchers have taken two opposite positions: “Some analysts argue that corruption is on the increase as a direct result of democratisation, since democratic political systems provide incentives and opportunities for corrupt practices. Others stress the potential of established democracies to devise institutions through which corruption can be tackled and contained.”^[38]

The scholarly consensus that links the quality of public administration to economic development is thus contested by some researchers. There has been a longer academic discussion about findings by Kaufmann and Kray who suggest that while better governance clearly tends to promote economic growth, growth per se does not tend to promote better governance.^[39] Recently, Kurtz and Schrank argued that “that there is far more reason to believe that growth and development spur improvements in governance than vice versa.”^[40] They question fundamental cross-national measures of good governance and thus the empirical foundations of the causal linkage between growth and governance. They believe “the dominant measures of governance” to be “problematic, suffering from perceptual biases, adverse selection in sampling, and conceptual conflation with economic policy choices.”^[41] This critique has been answered by Kaufmann et al. who defend their work in the field. They argue that the claims by Kurtz and Schrank are unsubstantiated and that their empirical work on the effects of governance on growth “is far removed from the best-practice frontier in crosscountry growth empirics.”^[42] Nevertheless, Kaufmann et al. argue that there remain limitations of existing cross-country measures of governance. They explain a “need to complement cross-country indicators with more detailed and nuanced within-country data in order to inform efforts to improve governance at the country level.”^[43]

2.4. The role of donors in fighting corruption

Today it seems agreed that only by fighting corruption can donors help to build a better world: “Corruption undermines governance, economic growth, and, ultimately, the stability of countries and regions.”^[44] But this view has not always been mainstream: “For decades, international institutions had little if anything to say about corruption.“^[45] Players in the international scene hardly acknowledged the existence of the word “corruption” in their daily business. They rather nebulously referred to it as the “c-word”. The World Bank (WB) is often referred to as a striking example for this practice. Theoretically, the Bank team had understood the mechanisms of corruption for a long time, as a WB paper from the early 90s shows: “Pervasive corruption is particularly damaging to development. Corruption occurs in all countries and in many different forms. It tends to thrive when resources are scarce, and governments, rather than markets, allocate them; when civil servants are underpaid; when rules are unreasonable or unclear; when controls are pervasive and regulations excessive; and when disclosure and punishment are unlikely.”^[46] But in practice not much happened for a longer period of time: “Ten years ago, the aid donors, the World Bank, and the IMF seldom discussed corruption or dictatorship.”^[47] But things seem to have changed, as a speech by then World Bank President Wolfowitz on 11 April 2006 in Jakarta, Indonesia, illustrates: “Today, one of the biggest threats to development in many countries […] is corruption,” he said. “It not only undermines the ability of governments to function properly, it also stifles the growth of the private sector.”^[48]

A watershed was the World Bank’s decision in 1996 to radically reverse its longstanding policy. Before, the bank held the position that it could not explicitly recognize or seek to address the acute problems of corruption in many of its borrowing countries, because local politics were outside the Bank's official mandate. Now, those problems were given a high priority: “While World Bank lending to promote economic reforms fell by 14 per cent annually between 2000 and 2004, its lending to improve governance rose by 11 per cent annually during that period, and by the latter year 25 per cent of its lending was committed to law and public administration in borrowing countries.”^[49]

The eventual rise of AC work in the strategies of donors is closely related to the aid effectiveness debate (see previous section of this book). Development assistance has seen some tremendous changes in the recent years – with the question of aid effectiveness in the centre. With international conferences and declarations such as the Millennium Declaration (September 2000), the International Conference on Financing Development (Monterey, March 2002), the World Summit on Sustainable Development (Johannesburg, August/September 2002), the Rome Declaration (February 2003) and the Paris Declaration (March 2005), the international community has been working heavily on the construction of a new aid architecture. One of the basic ideas is: “Not only should the recipient countries show commitment and make their national policies coherent in the pursuit of poverty reduction, but so should the donors. In addition, donors should harmonize their activities, an increasing share of donor funds should be general budget support, and donors' policies and activities should be aligned with the recipient country.”^[50]

There is a complex relation between aid and corruption, as – among others – a recent discussion paper by TI illustrates.^[51] Both occur in a particular form of interdependence. Of course, donors are indeed affected by corruption in recipient countries in their work and in service delivery. “Corruption within recipient countries can seriously undermine the achievement of intended results directly, by diverting a percentage of aid away from intended purposes and beneficiaries, or indirectly, by promoting inappropriate uses of aid.”^[52] But at the same time donors are also part of the problem: “Donors (bilateral or multilateral) who made and continue to make large amounts of funds available knowing the extreme weaknesses of a country's financial and other institutions, and the effect of corruption on those institutions, likewise share responsibility.”^[53] But this responsibility seems to be too often neglected in the Northern World.^[54] Take, for example, the Paris Declaration on Aid Effectiveness: although it states that corruption and lack of transparency “erode public support, impede effective resource mobilisation and allocation and divert resources away from activities that are vital for poverty reduction and sustainable economic development”^[55] it does not speak of the role of donors as part of the problem. The paper just deplores that “where corruption exists, it inhibits donors from relying on partner country systems.” Surprisingly, not a single word is wasted on how donors could help to reduce corrupt practices – or at least avoid fostering them.

The role of donors is crucial. Easterly, a former senior economist at the World Bank criticises: “High aid revenues going to the national government benefit political insiders, often corrupt insiders, who will vigorously oppose democracy that would lead to more equal distribution of aid.”^[56] The magnitude of the difficulties that donors face cannot be overestimated: “The role that donors can play in strengthening government ownership, PFM systems and domestic accountability is therefore more complex than many are willing to admit. Different actors and interests play different roles in shaping budget policies and priorities.”^[57]

A couple of recent studies suggest that aid actually decreases democracy and makes government worse: Knack has found that higher levels of aid worsen bureaucratic quality and lead to more violations of the law and to more corruption.^[58] Djankow et al. similarly have found that high aid caused setbacks to democracy in the years 1960-1999.^[59] They actually showed that the effect of aid on democracy is worse than the effect that oil has on democracy.^[60] In a somewhat cynical statement Easterly concludes: “Maybe bad governments attract donors who want to reform it just as sinners attract televangelists. However, if you control for this effect, donors make government worse.”^[61]

In the scholarly debate, there have been various contributions on how donors need to improve policies in order to achieve better development results in their target countries. Birdsall has compiled “seven sins” that donors must seek to avoid in order to improve the quality of their aid:^[62]

1. Impatience (with institution building)
2. Envy (collusion and coordination failure)
3. Ignorance (failure to evaluate)
4. Pride (failure to exit)
5. Sloth (pretending participation is sufficient for ownership)
6. Greed (unreliable as well as stingy transfers)
7. Foolishness (underfunding of global and regional public goods)

Addressing these issues is not crucial for the partner countries’ futures but also for the donors themselves – because only in this way they can secure domestic support for their work: “Ongoing concerns about the anti-developmental or anti-democratic nature of politics and about endemic corruption in poor countries, however, undermine the case for increased support from rich countries.“^[63]

3. Towards an AC regime in bilateral DC

The main argument of this book is that donors perform well in fighting corruption when they cooperate. I argue that in this case, they would need to form an international regime^[64] to fight corruption in their partner countries and at home in a coordinated way. This would be in their own rational interest as it would secure that their ODA is used as effectively as possible.

In this section, I examine if such a regime indeed exists. I start off by briefly explaining the idea of international regimes in general. I then continue to apply this theory to the field of this work. I use Krasner's definition of an international regime to show how the recent work of the donor community could be forming the basis of an AC regime in the making. I then continue to describe some basic rules and norms of this developing regime, based on key OECD/DAC documents.

However, I do not address the question of compliance. Similar to Bukowansky, I find the proposed regime still too young to judge its effectiveness.^[65] Sandholtz and Gray also do not speak about enforcement aspects in what they call not a regime but “the existence of international norms”.^[66]

3.1. A few words on regimes

In trying to theorize about international governance, scholars such as Krasner and Keohane have developed the concept of international regimes in the late 1970s and early 1980s. These scholars were in fact looking to provide an alternative to neo-realist approaches that dominated IR at that time. Thus, regime theory is distinct from neo-realism on the one hand and liberalism on the other. It can be described as part of a neo-institutionalist world view. One of the basic ideas behind this concept was to understand regimes as “something more than temporary arrangements that change with every shift in power or interests”^[67], that is distinct from one-shot agreements on the one hand and international organizations on the other. A first important use of the regime concept was the area of international trade, where scholars sought to explain how the economic institutions formed after World War II could be sustained in a time when the hegemonic USA were temporarily losing power. Later, the concept of regimes has been expanded to areas such as environment, communication and security.^[68]

In a somewhat classical definition, Krasner describes regimes as “sets of implicit or explicit principles, norms, rules, and decision-making procedures around which actors’ expectations converge in a given area of international relations”.^[69] Krasner further defines principles as “beliefs of fact, causation, and rectitude”, norms as “standards of behavior defined in terms of rights and obligations”, rules as “specific prescriptions or proscriptions for actions” and decision-making procedures as “prevailing practices for making and implementing collective choice”. Thus, principles and rules provide the normative framework for regimes, while rules and decision-making procedures provide specific injunctions for appropriate behaviour.^[70] Krasner notes that “changes in rules and decision-making procedures are changes within regimes”^[71] whereas “changes in principles and norms are changes of the regime itself”.^[72]

Although the definition of regimes has been contested by some scholars, “the study of international regimes made an important contribution by supplementing the technical aspects of formal IOs with the norms and rules governing state behavior”^[73]. The concept has been broadened and deepened since its first appearance in scholarly literature. The social-constructivist influences on the concept of regimes are particularly important.^[74] Constructivist scholars such as Risse namely attacked one rationalist assumption of the regime theory: the fixed preferences of the actors.^[75] Constructivists rather argued that the preferences of the states can indeed be altered by a regime that they are part of. Scholars of the regime theory have responded by including this possibility in their theory.^[76]

For practical reasons, I assume that in the present case of the developing AC regime, donors’ interests do not necessarily need to be altered by the regime in the short run. I have chosen to treat donors' preferences as fixed, as I understand the AC regime in bilateral DC as something nascent. Thus, the period under review is short enough to assume that the regime has not persisted long enough to alter the preferences of its participants.

3.2. What could the regime look like?

The idea of a developing international anti-corruption regime has been raised in the scholarly literature. While Bukovansky has argued for a regime that is largely based on a moral foundation,^[77] I argue in a purely rationalist tradition. This is because it seems to be in the rational interest of the major donors to form a regime on AC in bilateral DC. Young has described three developmental sequences for international regimes. He labels them “spontaneous orders”, “negotiated orders”, and “imposed orders”.^[78] In my view, the developing AC regime in bilateral DC that I want describe here is a form of the “negotiated orders” that Young describes as “conscious efforts to agree on [...] major provisions, explicit consent on the part of individual participants, and formal expression of the results.”^[79] More precisely, I understand the developing AC regime in bilateral DC as a “partial or piecemeal negotiated order” where “many problems [are] to be worked out on the basis of practice and precedent.”^[80] The bilateral donors have the rational interest that their ODA is used as effectively as possible. This is the main driver to set up the regime; the practical problems are to be solved during the ongoing work.

But why was the regime formed in the first place? Again, Young offers an explanation. He argues that leadership is a necessary condition, or “a critical determinant of success or failure in the processes of institutional bargaining that dominate efforts to form international regimes”^[81]. From the three types of leadership that Young presents – structural, entrepreneurial and intellectual – I find the idea of intellectual leadership most appropriate for the field in question. It is the power of ideas that “shape[s] the thinking of the principals in the process of the institutional bargaining.”^[82] In terms of leadership, the work of the OCED is important. The organization that is also active in the broader discussion on aid effectiveness pushed for a regulation in the field of AC.

I would like to start from Krasner's definition to show that an AC regime in bilateral development cooperation is currently forming. Let me start with a theoretical assumption. If there really was a regime in the making, we would need to see the following:

a) a distinct problem in the international relations
b) a “set of implicit or explicit principles, norms, rules, and decision-making procedures”
c) various actors with expectations that converge around these principles, norms, rules, and procedures.

But which of these ingredients are actually present?

Ad a) The international community in general and the donor countries in particular have recognized the influence of corruption on development. The Development Assistance Committee (DAC) is the body of the OECD that is mainly charged with issues related to co-operation with developing countries. An important forum for anti-corruption work is its Network on Governance (GOVNET). The OECD/DAC members have publicly recognized corruption as “a severe developmental problem” and have stated that “greater collaboration and the building of partnerships among donors and partners”^[83] is needed to fight it. In a recent policy paper OECD/DAC distinguishes between two types of actions that donor agencies can take to minimise the risks arising from corruption: firstly, a class of actions “whose effectiveness is not compromised by being done by each organisation separately (although the benefits will be greater the more organisations that undertake them)”, and, secondly, a class of actions that will “only be effective if they are undertaken jointly or in a strongly coordinated way”.^[84] OECD/DAC members thus undoubtedly recognize corruption as a distinct problem in international relations that needs to be treated in a cooperative way. The donors see the risk “that donor resources will not be used for the intended purposes”, additionally they fear “that corruption will undermine the achievement of economic growth and poverty reduction by its corrosive effects on government performance and private investment” and finally they are afraid of reputational risks “including the risk that aid to countries with corrupt leaders will tarnish donors' reputations and undermine the case for aid.”^[85]

Ad b) The documents that OECD/DAC has produced in the previous years form the core of a “set of implicit or explicit principles, norms, rules, and decision-making procedures”. On the next page you will find a list of papers that are noteworthy in terms of rule development.^[86]

[...]

^[1] I would like to acknowledge the help of the following people who has made this book possible: Odd-Helge Fjeldstad and Alf Morten Jerve (CMI), Sheona Duff (DFID), Bathylle Missika (OECD), Ina Eriksson (SIDA), Hansjörg Elshorst (TI), Wilhelm Mirow and Henry Thomson (FU Berlin), Christiane Arndt (Maastricht Graduate School of Governance and Harvard University), Kareen Klein (University of Cambridge and Université de Genève). I further would like to thank Desiree Nilsson (Uppsala University) for her comments at the presentation of my ideas at the Sixth Pan-European International Relations Conference 12-15 September 2007 in Turin. And above all, I would like to thank my parents for always supporting me.

^[2] In section 3.4. of this book I explain in detail why I have focussed on these particular cases.

^[3] OECD (2006), p. 3.

^[4] Bukovansky (2002), p. 5. proposes an AC regime that comprises UN, IMF, WB, and OECD on the international institutional side and TI and the International Chamber of Commerce on the NGO and private sector side. One could argue that the anti-corruption regime in bilateral development cooperation that I examine in this book might as well be a sub-set of such a larger AC regime.

^[5] The Development Assistance Committee (DAC) is a specialised committee of the Organisation for Economic Co-operation and Development (OECD). Its members periodically review the amount and the nature of their contributions to aid programmes and consult each other on other relevant aspects of their development assistance policies. For more information on DAC's mandate visit: http://www.oecd.org/document/62/0,3343,en_2649_33721_1918654_1_1_1_1,00.html (All online resources cited in this thesis have been verified on 13 October 2007). OECD/DAC has several subsidiary bodies, among them the Network on Governance (GOVNET) which is responsible for the AC work. For more information on DAC's structure visit: http://www.oecd.org/dataoecd/50/54/18058884.PDF

^[6] Browne (2007), p. 121.

^[7] For more details see: http://www.un.org/ecosoc/newfunct/develop.shtml

^[8] Krasner (1983), p. 2.

^[9] Examples are all taken from Klitgaard (1988), p. 7.

^[10] Amundsen (2006), p. 5.

^[11] Riley (1998), p. 135.

^[12] Andvig/Fjeldstad et al. (2001), p. 106.

^[13] OECD (2006), p. 10.

^[14] Soreide (2003), p. 1.

^[15] For more details on the CPI see: http://www.transparency.org/policy_research/surveys_indices/cpi.

^[16] For more details on the BPI see: http://www.transparency.org/policy_research/surveys_indices/bpi.

^[17] Kaufmann et al. (2006), p. 1.

^[18] Soreide (2003), p. 12.

^[19] Kaufmann et al. (2006), p. 1.

^[20] Soreide (2003), p. 12

^[21] Dreher et al. (2007), p. 460.

^[22] Andvig/Fjeldstad (2001), p. 78.

^[23] Andvig/Fjeldstad (2001), p. 63.

^[24] Lambsdorff (1999), p. 6.

^[25] Andvig/Fjeldstad (2001), p. 74.

^[26] Andvig/Fjeldstad (2001), p. 70.

^[27] Kaufmann/Wei (1999), p. 16.

^[28] Brunetti et al. (1997), p. 24.

^[29] Lambsdorff (1999), p. 7.

^[30] TI (2007), p. 2.

^[31] Spector (2005), p. XIII.

^[32] Kaufmann (2005), p. 82.

^[33] Kaufmann (2005), p. 93.

^[34] Burnside/Dollar (2000).

^[35] Spector (2005), p. XIII.

^[36] Burnside/Dollar (2000), p. 30.

^[37] Doig/Marquette (2005), p. 104.

^[38] Robinson (1998), p. 2.

^[39] Kaufmann/Kray (2002), p. 2.

^[40] Kurtz/Schrank (2007), p. 538.

^[41] Ibid.

^[42] Kaufmann et al. (2007), p. 555.

^[43] Ibid, p. 562.

^[44] Spector (2005), p. 7.

^[45] Bukovansky (2002), p. 6.

^[46] World Bank (1991), p. 6.

^[47] Easterly (2006), p. 133.

^[48] Wolfowitz (2006).

^[49] Arndt (2007), p. 3.

^[50] Selbervik/Nygaard, p. 29

^[51] TI (2007).

^[52] Ibid, p. 4.

^[53] CIDA (2000), p. 9.

^[54] Yet, recently, there seems to be evidence that a growing number of donors is starting to acknowledge this point. See e.g. SDC (2006), p. 18.

^[55] Paris Declaration (2005).

^[56] Easterly (2006), p. 135.

^[57] De Renzio (2006), p. 642.

^[58] Knack (2004).

^[59] Djankow et al (2005).

^[60] The bad effect of oil on the likelihood of democracy and good governance in a particular country is well documented. Among others, Kolstad (2007) calls this the “resource curse”.

^[61] Easterly (2006), p. 136.

^[62] Birdsall (2004), p. 3.

^[63] De Renzio (2006), p. 627.

^[64] I am aware that the word “institution” has now largely replaced the word “regime” in large parts of the scholarly literature. Nevertheless, I use the term in the context of this work because I think it is more appropriate for what I want to deal with in the field of bilateral DC.

^[65] Bukovansky (2002), p. 5.

^[66] Sandholtz/Gray (2003), p. 768.

^[67] Krasner (1983), p. 2.

^[68] Little (2005), p. 374.

^[69] Krasner (1983), p. 2.

^[70] Simmons/Martin (2002), p. 193.

^[71] Krasner (1983), p. 3.

^[72] Ibid, p. 4.

^[73] Simmons/Martin (2002), p. 194.

^[74] Zangl (2003), p. 136.

^[75] See e.g. Risse (2000).

^[76] Zangl (2003), p. 137.

^[77] Bukovansky (2002), p. 5.

^[78] Young (1993), p. 113.

^[79] Ibid, p. 99.

^[80] Ibid.

^[81] Young (1991), p. 281.

^[82] Ibid, p. 307.

^[83] OECD (2003), p.6.

^[84] Ibid, p.16.

^[85] Ibid, p.6.

^[86] Currently, DAC/GOVNET's Anti Corruption Task Team (ACTT) seeks to operationalize these documents in joint corruption assessments, and – possibly – common response principles. As these documents are not publicly available yet, I cannot discuss them in detail.