An Efficient Anticorruption Sanctions Regime? The Case of the World Bank

I. Introduction

While international development banks seek to support governments in their development efforts, corruption in recipient societies can undermine this support. For this reason, these banks attach to their financial transfers a range of integrity criteria and mechanisms aimed at securing value for money and preventing theft.1 These include, most commonly, professional procurement rules, external audits, and the enforcement of rules that deem illegal a set of actions found damaging to development initiatives, including fraud, corruption, and collusion. Many recipient governments are not able or trusted to adequately enforce the requested integrity mechanisms, or they lack motivation to secure funds coming from an international organization; therefore, the development banks have started to operate their own controls and act upon detected violations of the agreed-upon rules and laws. They operate hotlines for internal and external whistle-blowers, instruct project managers on what suspicious acts they should report, and demand careful audits of the projects and programs they finance. If they did not, they could easily be seen as condoning fiduciary risks and even as facilitating crime and corruption, given the large amounts of money they transfer to countries with weak law enforcement institutions.

A financial institution established to promote development, however, is not vested with the powers usually associated with efficient law enforcement, such as the authority to investigate, obtain evidence, and summon actors to court. Facing cases of corruption, the World Bank’s impact on criminal justice will rely largely on collaboration with law enforcement systems at the national level. Their support cannot be taken for granted—particularly not in cases where high-ranking politicians or civil servants are suspected of involvement in wrongdoing. Accordingly, a development bank’s power to enforce laws and sanction offenses independent of national criminal justice systems and sanctions regimes is limited to its capacity as a supreme financial institution—it can exclude players from the financial services it offers. Despite this limited enforcement mechanism, development banks have attempted to develop their own anticorruption sanctions regimes. This step can be understood as an attempt to compensate for law enforcement weaknesses in recipient countries as part of the banks’ strategy toward the overall goal of development.2

This Article addresses the function of these sanctions regimes by looking at the case of the World Bank. Among international financial institutions, the World Bank has been at the forefront in developing a sanctions regime. Its regime is the earliest, most comprehensive, most open, and most recognized. In the period from 2007 to 2014, the World Bank imposed sanctions on around 250 firms and individuals by temporarily suspending them from operating on World Bank-financed development projects and programs.3 When governments fail to meet certain integrity criteria or to act appropriately upon suspected fraud, collusion, or corruption, the World Bank and the other development banks can suspend or cut off financial support to governments; such actions are often referred to as remedies. Yet these remedies are not used frequently and seem not to be implemented with an intention to identify the individual leaders responsible for governance failure or clear-cut crime.4 Instead, the sanctions regime is targeted primarily at the suppliers of World Bank-financed contracts and operations.

The Legal Vice Presidency of the World Bank, which oversees the sanctions regime, has modified the regime several times following reviews of its functionality.5 As part of these review processes, the institution has invited external experts to share their views about the sanctions regime. A next step in its review, according to the Bank, is to assess “the overall efficiency and effectiveness of the system—i.e., whether the system as a whole is meeting its objectives of excluding corrupt actors and deterring fraud and corruption in World Bank Group operations, at an appropriate cost to the World Bank Group.”6

Motivated by such calls and processes, this Article discusses the performance of the sanctions regime in terms of how well the Bank protects funds, promotes fair markets, reduces corruption, and secures confidence in the World Bank. While these aims exceed the narrowly-stipulated goal of the sanctions regime to protect the funds for development support, they are each relevant for this debate. In this Article they are viewed as cumulative criteria, meaning that it would not be satisfactory if the sanctions regime were dysfunctional in any of them. Therefore, we set out to address both efficiency and effectiveness. Efficiency in this study refers to the achievement of objectives in a cost-efficient way, given a choice of alternative strategies—that is, an economic understanding of the term. Effectiveness, however, is understood as “doing the right thing” given the overall goals of the institution. With this distinction, we go beyond an assessment of the sanctions regime per se. Instead of addressing a sub-system’s efficiency separately from the overall development goal of its organization, we consider the regime in the broader development perspective associated with the World Bank. The objective of rebuilding domestic law enforcement in client economies7 is therefore addressed, in addition to the mentioned aims of protecting the funds, promoting fair markets, reducing corruption, and securing confidence in the institution, because it is essential for development.

Given this perspective, we argue that the sanctions regime faces problems in delivering efficiently on these specified aims, even if the World Bank overall plays a central role in promoting anticorruption in the world. The organization seems to have developed this regime as if the fiduciary risks and problems associated with governance in developing countries could be detached from the very institutions and societies where the problems are embedded, and brought into a sphere where the Bank itself can exercise control. Within this sphere, the focus on ensuring due process in the implementation of sanctions seems to have outweighed the importance of considering whether the regime is having its intended impacts on the ground.

The Article consists of three main sections. In Section II the sanctions regime in the context of the World Bank’s mandate is considered. Section III describes the World Bank’s efforts to ensure due process, modeled on the procedures typically found within national justice systems. Finally, Section IV addresses the system’s effectiveness and efficiency more specifically. Conclusions and recommendations follow. 

II. The World Bank's Position and the Purpose of Sanctions

At first glance, the World Bank’s sanctions regime comes across as irrational. The organization offers loans and aid to the governments of developing countries and makes these governments responsible for pro-development spending. However, when the money disappears because of corruption and fraud, the World Bank does not hold the government responsible, but instead places sanctions on the suppliers operating in contract with the recipient government. A closer look at the particular position and authority of the World Bank reveals the rationality of this regime. This Section aims at outlining some of the basic features of the World Bank’s regime. First, in II.A, we take a closer look at the mandate of the World Bank. Thereafter, in II.A we discuss how the World Bank collaborates with its member countries, as its clients, particularly with a view on anti-corruption measures. Lastly, for explanation of the sanctions regime’s current structures and rules, we address in II.C this sub-system’s historical background and development.

A. The World Bank’s Mandate for Having a Sanctions Regime

The World Bank sanctions system is anchored in the International Bank for Reconstruction and Development (IBRD) Articles of Agreement, last amended in 2012. According to Article I(i), the institution’s purpose is “to assist in the reconstruction and development of territories of members by facilitating the investment of capital for productive purposes, including . . . the encouragement of the development of productive facilities and resources in less developed countries.”8 The phrase “facilitating the investment of capital for productive purposes” refers to the World Bank’s prime function of managing funds transferred from the more developed countries and offering loans and aid to governments for the purpose of reducing poverty and promoting development. The organization, steered by a board consisting of member-government representatives (as stipulated by its Articles of Agreement), controls development funds on behalf of donor governments. These donor governments, who have to defend to their voters and national auditors the amounts of state revenues transferred for development support, of course expect accountable spending.9

According to Article III, Section 5(b), “The Bank shall make arrangements to ensure that the proceeds of any loan are used only for the purposes for which the loan was granted, with due attention to considerations of economy and efficiency and without regard to political or other non-economic influences or considerations.”10 The responsibilities spelled out in these articles are often referred to as the World Bank’s “fiduciary duty,” or its “obligation to reduce fiduciary risks.”11 At the same time, however, the institution’s authorization is anchored in the idea of redistribution; it is expected to meet a demand in wealthy countries for sharing with those who have less. It is obliged to offer favorable monetary services to member countries in need, even if there is generally a higher risk of corruption in developing countries. This duty implies the Bank cannot easily cut support to poor countries just because their governments cannot be trusted. It is therefore well-established that a core rationality for the World Bank is to transfer funds to poor areas, including societies with a high risk of fraud and theft.

B. Balancing Collaboration and Anticorruption Imperatives

The World Bank collaborates closely with its client countries regardless of integrity concerns. The governments can be encouraged to introduce integrity systems and can be influenced to enforce their laws, but they hold a monopoly on law enforcement. Good dialogue is essential to the Bank’s ability to promote recipient governments’ capacities and pro-development priorities, especially in countries where integrity mechanisms are not strongly institutionalized. The World Bank does raise concerns about various corruption risks with recipient governments.12 If the Bank voices such insinuations too loudly or explicitly, however, it may jeopardize its collaboration with a recipient government, and this in turn may have repercussions for various development programs introduced to improve health, education, and infrastructure, for example.13 Instead, the World Bank encourages the introduction of integrity-promoting laws and institutions, seeks to enlarge the space for civil society watchdogs, and, to some extent, puts pressure on governments by including pro-accountability steps as part of loan agreements.14 Nonetheless, in circumstances where initiatives to promote integrity and law enforcement come into conflict with the interests of a corrupt elite or the dishonest intentions of an incumbent regime, there are limits to what the World Bank can do. The institution is then placed in an awkward situation, as it is expected to transfer funds to the recipient government, but also to protect the funds.

This is why the sanctions regime can be seen as a pragmatic response to the risks of fraud and corruption. It is built on the authority that the World Bank does have—the Bank can deny spending to players in the private sector if they are involved in fraud, as they are not clients. By limiting the scope for corruption to state-private sector interactions, the World Bank strives to reduce the risk of corruption. The Bank operates a number of other governance and anticorruption initiatives directed at governments, including investments in domestic justice sector reform programs as well as law enforcement components of projects it finances in other sectors.15 Upon discovering indicators of funds mismanaged by governments, there is usually a prompt reaction by the Bank, followed by a clear request for facts, investigation, and enhanced supervision, sometimes resulting in halted development support. In addition, the Bank has initiated a number of broad cross-country anticorruption initiatives, facilitated numerous expert working groups, and developed methodologies for analyzing corruption-related problems and toolkits for developing efficient strategies. It also supports several international programs such as the Stolen Asset Recovery Initiative (StAR), the Extractive Industries Transparency Initiative (EITI), the Construction Sector Transparency Initiative (CoST), and a network for collaboration with anticorruption agencies worldwide. When it comes to holding someone responsible for corruption, however, the organization focuses on suppliers of government contracts. Therefore, while governments are the ones that receive and control funds from the World Bank and are told to act responsibly, it is the firms on contract to the recipient government that can most easily be held responsible when fraud and corruption occur.

C. Background for the Current Sanctions Regime

The current sanctions regime must be understood as a result of ongoing developments in the World Bank. An early form of sanctions was introduced in 1996, not long after James Wolfensohn was appointed president of the World Bank and around the time he gave his famous “cancer of corruption” speech.16 By 1996, several multilateral organizations, including the United Nations (U.N.) and the Organisation for Economic Co-operation and Development (OECD), had started policy work on corruption, including preparatory steps toward the U.N. and OECD conventions on corruption. However, Wolfensohn’s speech came to represent a paradigm shift, a turning point after which corruption could be addressed more easily in diplomatic circles and research.17

It was the World Bank president himself who initiated the new focus on corruption, since before that, corruption had been very much a taboo in the organization. According to Sebastian Mallaby, speaking about corruption was thought to “violate the Bank’s apolitical charter,” and the Bank would be seen as interfering with politics if concerns about political fraud and theft were raised.18 Despite staff members’ attempts to defend anticorruption as a pro-development strategy19 —that is, independent from politics regarding the allocation of scarce resources—the organization still struggles with this concern. While its “apolitical charter” has not prevented the World Bank from attempting to influence real politics—on budget priorities, infrastructure, liberalization, health, education, and so on—the concern seems to have prevented it from efficiently raising issues of grand-scale “grabbing,” operationalizing the recommendations of the organization’s many political economy analyses, and placing governance conditions on its lending.

Several of President Wolfensohn’s reform initiatives seem to have been started on his (fairly good) instinct, but without sufficient anchoring in the organization’s bureaucracy of development experts, Mallaby explains.20 Some high-level decisions led to hasty solutions in the organization—which might have been the case with the new fight against corruption. While Wolfensohn’s commitment to anticorruption led the Bank’s researchers to develop new cross-country data sources, background information for corruption indices, procurement systems, and integrity mechanisms more generally, it seems that little thought was given to how to secure country ownership of this agenda. Instead, the Bank developed best-practice policy procedures as a menu of options for governments, and it started to rank countries on how their various governance institutions performed. Moreover, the Bank resolved to lead by example—there should not be corruption and fraud among World Bank staff, or in World Bank-financed operations.21

Toward this end, the organization’s Integrity Vice Presidency (INT) was established in 2001. It was supposed to investigate allegations of fraud and corruption in World Bank-financed projects and present its findings to the already established Sanctions Committee.22 Its attention was directed at World Bank staff overseeing development programs, as well as those possibly involved in client economies. Hence, from the outset, to the extent that the World Bank was inclined to place sanctions on offenders, it focused on those who violated its own regulations and procurement guidelines. The sanction process did not relate to the governments that were expected to control the funds, nor was it anchored in efforts to strengthen their criminal justice systems. This means that the World Bank responded to cases of corruption in its projects as issues to be sanctioned and solved internally, in and by its own systems, rather than as problems to be addressed in the larger context of corruption in recipient governments.

In order to assess its sanctions system in terms of efficiency, fairness, and transparency, the World Bank invited a group of experts led by Dick Thornburgh to make recommendations. 23 The Thornburgh report, as it is known, significantly influenced the design of the sanctions regime, especially with respect to the fairness of procedures and the position of the sanctions-determining body vis-à-vis investigations.24 It also helped establish the protection of funds as the main purpose of the regime.

III. The Current Sanctions Regime

Following the Thornburgh report and internal reviews, the World Bank further developed its definitions, strategies, and procedures for imposing sanctions upon detected offenses.25 This Section addresses some characteristics of the resulting regime. Hence, we focus specifically on the role of sanctions in a development perspective, leaving aside the whole range of other Bank-initiated anticorruption efforts. In III.A we look closer at the meaning of “offenses” and “sanctions” in the World Bank’s regime while in III.B we pay particular attention to the due process mechanisms of the regime.

A. Forms of Offenses and Sanctions

The offenses subject to sanctions by the World Bank sanctions regime include corrupt practices, fraudulent practices, collusive practices (usually referring to cartels in markets), coercive practices (involving harm or threats of harm), and obstructive practices (hindering investigations), as stipulated in the World Bank Sanctions Procedures of April 15, 2012.26 The definitions of the offenses are constant across all World Bank lending and operations. Within the Bank they are considered as violations/offenses that should be reacted to regardless of how the acts are legally defined and sanctioned by lending governments. Individuals or firms that have committed one of the listed offenses while operating on government contracts financed by the World Bank are consequently viewed as offenders; in the World Bank terminology often referred to as respondents.

The sanctions in use are primarily conditional or unconditional temporary debarment (suppliers are excluded from operating on World Bank-financed contracts worldwide), “naming and shaming” (those debarred are listed on a public website), a form of negotiated settlement with the INT, and a letter of reprimand.27 The offense most frequently sanctioned is fraudulent practices. This was the offense in 86 percent of cases of debarment in the period 2007–2013, according to the Office of Suspension and Debarment. Corruption occurred in 14 percent of cases, and collusion in 9 percent.28 The different percentages primarily reflect differences in the ease of investigating cases and do not necessarily reflect the distribution of ongoing crimes. Compared to detecting fraud, uncovering corruption and collusion usually requires more difficult and time-consuming investigations, and therefore the figures on the latter forms of crime are lower.

In 2010 the impact of sanctions was intensified when the biggest development banks reached an agreement on cross-debarment, meaning that a firm or individual found guilty by one of the organizations will almost automatically be debarred by the others.29 The agreement includes the African Development Bank, the Asian Development Bank, the European Bank for Reconstruction and Development, the Inter-American Development Bank Group, and the World Bank Group.30 This means that the “sanction,” typically debarment that varies in length from six months to several years, now implies exclusion from tenders on all contracts financed by any of these banks and, therefore, exclusion from a large world market.31

B. Due Process in Sanction Decisions

The sanctions regime implies that the World Bank exercises legal powers that are normally associated with national sanctioning systems. The Bank itself has always maintained that the system is fundamentally administrative.32 However, looking at the types of violations that are regulated, the far-reaching sanctions and the relatively independent system for investigating breaches, and considering the kind of national remedies it effectively replaces, one could ask whether the World Bank’s sanctioning regime does not also reflect criminal justice mechanisms.33

In any case, it seems clear the Bank places substantial weight on due process procedures—intended to ensure fairness, regularity, and transparency—in making certain decisions about the application of sanctions. The result is a procedure that in many respects resembles criminal law procedures at the national level. A separate unit, the INT, is responsible for carrying out investigations and can act independently in response to complaints of sanctionable practices.34 These complaints are typically received from World Bank representatives with oversight responsibility for funded programs, from government representatives, or from other stakeholders or witnesses. In carrying out these procedures the INT operates within strict institutional structures designed to ensure due process. Another unit, the Evaluation Officers, assesses the allegations of offenses (as defined by the system) and determines whether the evidence presented is sufficient to warrant moving ahead with a process akin to prosecution, aimed at sanctioning the alleged conduct.35

There are surely many good reasons for such a formal system, even besides ensuring legal certainty for the accused, so commonly associated with due process. It is, among other things, clear that a wrong decision hurts not only the debarred party, but also the Bank and its borrowers in the form of reduced competition. At the same time, the sanctions system is not consistent in upholding these high standards of due process. Not all cases require equally high due process standards, and some cases might benefit from a more administrative approach. In many cases, suppliers are effectively debarred from all World Bank-financed contracts from the moment investigation begins.36 In addition, governments around the world may lose confidence in suppliers under World Bank investigation, and may, on that basis, choose to debar them from their public procurement tenders. Moreover, in many cases those suspected can reach a negotiated settlement with the INT. This arrangement is similar to a formal plea bargain or a conditional withdrawal of charge. It consists of an agreement that the INT makes with “the accused,” typically a supplier. Through that agreement, the accused can influence the charge, the sanction, and what facts of the case to keep confidential, typically in exchange for the INT’s access to information, implementation of a convincing compliance system, and monitoring of the supplier’s business activities. Usually, the offender must prove commitment to the agreement before the INT will release it from temporary debarment. Like negotiated settlements in criminal law, this option of reaching an agreement with the INT arguably speeds up case processing while also incentivizing compliance.37 As in other contexts, however, such arrangements easily compromise principles of predictability, protection from self-incrimination, and transparency.

If no settlement is reached and the supplier is found guilty by the Evaluation Officer, the supplier can appeal the decision to a different body, the Sanctions Board. This unit, consisting of both World Bank staff and external experts, functions as an appeals court, again in line with due process considerations.38 The Sanctions Board applies a low standard of evidence—if the offender “more likely than not” is guilty, a sanction will be imposed. This implies a weak presumption of innocence, particularly when compared to national standards of criminal evidence. In this aspect the regime is comparable to procurement law debarment regimes operated by state administrations, where it in many cases is a sufficient basis for rendering a supplier ineligible to participate in a tender that he or she has the status of suspect in a crime.39

In the most recent makeover of the sanctions regime, one of the objectives was to increase the regime’s degree of transparency.40 At present, all principles, rules, definitions, and procedures are set forth in the World Bank Sanctions Procedures, which are publicly available. Sanctions imposed are made public to some extent. While the debarred suppliers are officially blacklisted, few details about the negotiated settlements (or “conditional non-debarment”) are made public. The supplier can negotiate the content of the press release, and the World Bank may offer confidentiality in exchange for evidence or commitments to a compliance system. The provisions of these “deals” therefore cannot be assessed by the general public.41

IV. The Efficiency and Effectiveness of the Sanctions Regime

When it comes to the overall goal of promoting development, the World Bank has played an indispensable role in putting good governance and anticorruption on the policy agenda for governments around the world. However, what this Article addresses is not the World Bank’s convincing role in anticorruption in general, but rather the sanctions system’s role in relation to the World Bank overarching aims such as promoting development.

According to the Sanctions Procedures (Article I), the purpose of this regime is to protect Bank funds and to serve as “a deterrent upon those who might otherwise engage in the misuse of the proceeds of Bank financing.”42 The aims of deterrence and protecting the funds are important in and of themselves. Nonetheless, one cannot ignore the regime’s role in contributing to the wider goals and activities of the World Bank, especially its goal of promoting development. Despite its more specific objectives, the expectation is that the regime is coherent with and promotes the World Bank’s overall policies. After all, the regime was set up by the World Bank itself. Therefore, the sanctions regime should be expected to promote, or at the very least not work against, integrity in the relevant markets. Second, it should be expected to actively reduce the risk of corruption. Third, the regime should be expected to secure confidence in World Bank operations. Finally, it should be expected to encourage and support enforcement of anticorruption measures, or at least not undermine them, especially in terms of enforcing domestic law enforcement. This Section addresses these different expectations—or aims—in terms of their contributions toward the overall goal of promoting development and their implications for law enforcement systems in client countries. It is difficult to assess empirically the extent to which the aims are met in practice, especially since there is not only an acute lack of data available for such assessment, but also because there are often a number of initiatives with similar objectives introduced in a given country. The debate in this article considers only what can be rationally expected. To what extent can we assume that these aims are met—or at least not undermined?

From this starting point, we will discuss further how the World Bank’s sanction system relates to the protection of funds in Section IV.A; the promotion of development is also dependent on the protection of fair markets, as market fairness affects the prices and quality of products and services available to firms and citizens in Section IV.B; closely related to the protection of fair market is the securing of confidence in the World Bank operations addressed in Section IV.D; lastly, in this Section IV.E we will also discuss the importance of rebuilding domestic law enforcement.

A. Sanction Strategies for Protecting the Funds

Development loans are transferred to governments in need of the World Bank’s monetary services. Governments are expected to spend the funds in accordance with the agreed-upon development goals and existing integrity mechanisms. Once funds are transferred, as mentioned, the recipient government—and not the World Bank—conducts the spending, including procurement of services from domestic and international suppliers. When corruption, collusion, or fraud occurs and distorts the intended benefits of World Bank aid and loans, the wrongdoing often involves government representatives—sometimes high-ranking civil servants and politicians—in addition to players in the private sector, both individuals and firms. Whoever is to blame, once the funds are received, the recipient government is in charge of protecting the funds and spending and is at least indirectly responsible for any inadequate management and control of the funds.

Given the World Bank’s position, mandate, history, and collaboration with governments, the sanctions system primarily targets suppliers on government contracts. As explained in Section II, the World Bank needs to collaborate closely with governments to ensure progress of the various development programs it funds. Those involved in overseeing the programs might be worried that a more demanding line on corruption would place this collaboration at risk. Such concerns may explain the current solution, where primarily the firms actively involved in the specified forms of crime, including when bribes have been demanded by government officials, are held responsible. The question is not whether there should be consequences for suppliers involved in corruption, which is rather undisputed, but rather, if these private actors are the most responsible offenders, given the World Bank’s explicit allocation of responsibilities for the management of funds and its own definitions of crime. Are suppliers the ones who should be the main subjects of investigation and prosecution? Should the investigation of their offenses be the responsibility of the World Bank? Or could there be alternative solutions whereby the World Bank investigators function primarily as advisers to domestic police forces while the recipient government is under pressure to enforce the law?43

One argument for targeting primarily private sector suppliers on World Bank financed contracts could be that if government representatives are involved in a crime, they will do what they can to prevent an investigation that puts their own position and properties at risk. But this is not necessarily a problem in all cases worth investigating and prosecuting; very often the perpetrators are not influential enough to stop an investigation. Instead of finding alternative solutions for circumstances involving highly influential perpetrators, the World Bank has chosen not to discriminate, and acts in general as if it cannot rely on recipient governments’ law enforcement systems.44 Instead, the Bank conducts its own law enforcement initiatives in all cases of suspected violations in Bank-financed projects.45 In light of this strategy, the World Bank will not automatically rely on any other government’s or court’s judgment of a supplier, and excludes from procurement only those suppliers that have been found guilty by its own investigators, through its own sanctions process.46 This means that a company found guilty of bribery by a court in South Africa, Sweden, or Chile, for example—but not by the World Bank sanctions system—can still be found eligible to participate in tenders for World Bank contracts. This asks if the risk of mistakenly trusting the judgment of a foreign court is more serious than the risk of allowing suspected suppliers to participate in tenders. In practice, however, World Bank investigators are inclined to react to alleged cases of fraud or corruption, and upon their own assessment of the evidence, they might proceed with a process toward debarment.

In any case, the organization’s insistence on sanctioning only those found guilty by its own system can be understood in light of two considerations. First, the World Bank has country offices in about one hundred countries and closely follows the performance of state institutions.47 In a good number of countries, it cannot rely on the fair judgment of the national criminal justice system, for reasons that may include low capacity, integrity issues, or politically steered decision making.48 If the Bank were to officially trust the courts in certain countries and not in others, it could easily compromise its own attempts to maintain good dialogue with governments whose courts were deemed untrustworthy. Second, while the World Bank’s “legal powers” in corruption cases are based on nothing more than the decision not to buy from a certain supplier (just as ordinary people can bypass a certain shop they do not like), the Bank seems very aware of the potential consequences of imposing such a sanction on suppliers. On this issue, however, its signals are somewhat inconsistent, given the mentioned fact that suppliers are effectively debarred from the moment they are under World Bank investigation. At the same time, the organization maintains the importance of playing it safe in terms of protecting its reputation and finds it difficult to sanction a supplier unless it has used its own institutional judgment and can clearly defend the process behind the sanction as just and fair.

Nonetheless, given the ultimate goal of protecting development funds, it appears even more inconsistent to hold suppliers responsible for offenses and not to focus primarily on the governments that are charged with the responsibility to manage the funds efficiently. It cannot be claimed that funds are better protected by holding only suppliers responsible for decisions that are, in fact, controlled by government representatives.

B. Promoting Fair Markets

The first expectation of the sanctions regime, listed above, is that it promotes—or at the very least does not work against—the integrity in the relevant markets. When it comes to promoting development, the protection of fair markets is essential, as market fairness affects the prices and quality of products and services available to firms and citizens. Excluding corrupt players is seen as a strategy to level the playing field for honest, development-promoting competition. The problem with this solution is that we need to keep suppliers in the market for the sake of competition. The more suppliers are excluded, the more competition is distorted and the more market power accrues to the remaining suppliers. At this point, the price-quality combination easily deteriorates—to the detriment of development. Temporary debarment of corrupt players is nevertheless justified if it has longer-term impact on integrity and fairness, with positive indirect consequences for the market—in this case, the global market for World Bank-financed contracts.

The World Bank relies on debarment of dishonest suppliers as a strategy against corruption, apparently without regard to the level of integrity it expects of the recipient-government agencies involved in procurement. An implicit assumption behind the debarment logic is that the procurement agency is honest, but in countries where corruption is known to be a problem, that assumption may not hold. If a government agency prefers to collaborate with dishonest suppliers, it is unlikely that top-down enforced debarment of suppliers will have positive effects in terms of promoting fair markets. Excluding certain firms may even enable the agency to extract higher bribes from the firms that remain in the market if, given weaker competition, there are more profits to share between a corrupt government agent and the supplier. In the most extreme scenario, a corrupt procurement agent could demand very high bribes from a firm that stands to gain from sole source procurement. Moreover, the more suppliers are excluded, the fewer the number of remaining firms, and the easier it is for them to facilitate cartel collaboration.49 This is exactly in the interest of the organized crime that infiltrates private sector markets, a substantial problem in a number of developing countries. In either case, the power to control contract allocation as well as the size of a bribe rests largely with the procurement agent and not so much with the targeted suppliers. Removing a supplier will not necessarily reduce the procurement agent’s inclination or opportunity to demand bribes, unless it deters corruption by raising the risks.50

Debarment is not even an optimal solution when procurement agencies are honest—as most are likely to be, even in countries with perceived high levels of corruption. In terms of its impacts on a given society, the decision to debar entails a trade-off: one accepts fewer bidders with the likely result of weaker competition with inferior price-quality combinations in the short run in order to reduce corruption risks in the long run. Externally imposed (global) debarment of suppliers, however, does not allow a government to weigh the short-term costs against the presumed long-term benefits of reduced corruption in their own markets. Instead, by weakening competition, it imposes a cost on society in terms of higher prices and lower quality. If corruption is not really a problem because procurement agencies are honest, suppliers will not be able to bribe agents for market benefits, and, the costs incurred by weakening competition will not be compensated by higher levels of integrity (because this level is already high). The impact of debarment in a society where honest procurement agents control the allocation of contracts, then, is simply to restrict the market—corruption in procurement cannot be reduced if it is not a problem in the first place. Dishonest suppliers should be kept away from tenders, but this is already in the interest of honest procurement agents, who will not give contracts to suppliers in exchange for bribes. The undesired market consequences of debarment are especially pronounced in oligopoly markets with few suppliers, a not uncommon market situation in developing countries.51

In sum, it is difficult to see how global debarment of dishonest suppliers will improve integrity in public procurement markets, whether procurement agents are honest or dishonest. In order to reduce the risk of unintended market distortions, governments can introduce measures that promote competition, such as competition authority controls and an investment-friendly business climate. When it comes to competition-friendly designs of the sanctions regime specifically, the option of negotiated settlements might be a good model, since it secures the intended effects of debarment while returning suppliers more quickly to the market.52 However, given the lack of transparency about these settlements, it is difficult to tell how they work from a development perspective.

C. Reducing the Risk of Corruption

Over the last decade we have witnessed increasing attention to the consequences of corruption, implementation of international anticorruption conventions, and a growing availability of information about the integrity performance of both firms and governments.53 Firms increasingly are rewarded by criminal justice systems for their internal compliance programs. In this context, the World Bank’s sanctions regime has been one of several important forces that have pushed the integrity agenda in the right direction, and whatever inefficiencies it might have, the powerful signaling effect of this regime cannot be ignored. Companies operating in the relevant markets are pushed, if not forced, to take into account the threat of sanctions and consider what this means for their business operations. Moreover, now that the main development banks operate with automatic cross-debarment, it is reasonable to expect the regime to have a certain deterrent and/or norm-developing effect. Most likely, it reduces the propensity among many suppliers to offer bribes. According to Hugette Labelle, then-Chair of Transparency International, “the World Bank’s sanction process is critical to eradicate fraud, corruption and collusion from the projects it finances.”54 To what extent is this assumption correct? The problem with a supplier-focused sanctions regime as a pro-development anticorruption strategy is similar to the issue discussed above: a more hesitant attitude toward corruption among suppliers will not necessarily stop government representatives from demanding bribes. The obtainable profits for those involved and the bribes for the corrupt decision makers may in fact increase if one or more suppliers are debarred from a given market. Typically, in a developing country context or in an emerging market economy, law enforcement institutions are far from robust, and despite the World Bank’s sanctions regime, the risk of being caught in corruption is still very low. For this reason, and despite expected deterrent signaling effects of the sanctions regime, corrupt government officials may still experience a steady supply of firms willing to take part in a corrupt deal. Indeed, the role of a supplier involved in corruption is often to allow the government representatives in charge to use a World Bank-financed contract as a tool for organizing the transfer of funds from the state and into their personal bank accounts. The fewer suppliers who are willing to offer such opportunities for corruption and fraud, for example as the result of a heightened risk of being caught and sanctioned, the more a (remaining) corrupt supplier can demand in contract negotiations and, eventually, the more profitable the deal.55 Therefore, a deterrent effect on suppliers coming from the World Bank sanctions system could reduce the number of corrupt transactions, while leaving the total amount of bribe benefits and the most serious cases unaffected. A decline in the number of firms willing to offer bribes may simply imply huge benefits for those firms that are still involved in such crime. For these reasons, the problem cannot be solved by holding only the bribe-paying side responsible.56 If the government representatives cannot be reached, the system can hardly be seen as an efficient means of reducing corruption in government-controlled spending. To some extent, therefore, it seems that the World Bank sanctions regime in many cases targets the wrong offenders, or at least does not target all the offenders.

D. Securing Confidence in World Bank Operations

A development bank cannot function according to its stated purpose unless it enjoys a level of trust.57 It needs trust from its donors, who support the bank’s operations in the belief that funds will not disappear into the pockets of the wrong people along the way. Well-confirmed risks of fraud and corruption in borrowing countries effectively set limits on a development bank’s trustworthiness. Taxpayers in donor countries are increasingly aware of such risks. They are also concerned about how donor funding may fuel corruption, and possibly have a deleterious, if counterintuitive, effect on development.58 For the World Bank, it is therefore essential to show that such risks are small if it handles the transfer of funds. In an era when donor countries have expressed significant concerns about the transfer of funds through the Bank, the sanctions regime helps restore credibility to World Bank operations.

The problem with this strategy is that it is not the World Bank’s own activities that jeopardize the institution’s trustworthiness. Rather, it is the actions of recipient governments and third-party contractors operating domestically. Any measure intended to enhance donor trust in the Bank necessarily involves and affects these third parties. At the same time, it is only by means of its own actions and controls that the Bank can influence donors’ perceptions of its trustworthiness. Through its sanctions regime, the World Bank tries to convince donors that it controls a risk that in fact is not under its control.59 By constructing a comprehensive regime instead of using a simple administrative public procurement debarment list, the Bank underscores the seriousness of its commitment to anticorruption.60 Nevertheless, since the main control over state revenues and contract allocations rests with state administrations, a sanctions regime targeted only at fraudulent suppliers cannot, in itself, secure donor confidence in the World Bank’s operations.

Instead of thinking critically about the strategy underlying its sanctions regime, the World Bank seems stuck on the idea that the regime will fulfill expectations if its internal organization and procedures are sufficiently well designed. Over the past decade, the Bank has made continuous efforts to strengthen the design of the system, soliciting reviews and making improvements;61 the assumption seems to be that pro-development consequences will follow automatically if only the procedures are good enough.62 There have been no attempts in this period, however, to empirically explore the consequences of sanctions or to assess systemically what happens in markets where suppliers have been debarred.63 As far as we are aware, there has not been a single study seeking to determine whether sanctions are associated with greater integrity, fairness, and competition in markets or with a noticeable reduction in corruption and fraud. Those overseeing the regime have been preoccupied with developing proper standards and procedures. It is possible that the organization is now preparing to undertake such impact evaluations (although this is not mentioned in the information brief on the review process). At present, however, it appears that the World Bank is more concerned with improving donor confidence by demonstrating an ability to impose sanctions than with verifying that the sanctions actually reduce corruption and fraud.

Given the objectives of the sanctions regime and the importance of securing confidence in World Bank operations, there is a need to think differently about the system’s motivations and purpose. First, it is motivated by the corruption and fraud taking place in World Bank-financed programs and operations. Second, it has evolved as a solution to ensure donor trust in the Bank’s ability to manage funds responsibly. But, third, the system is motivated also by the incapacity of domestic criminal justice systems to tackle the problem. These motivations have led the World Bank to try to deal with corruption and fraud through its own sanctions regime—as if the problem could be detached from the domestic government system in recipient countries. Notwithstanding recently increased collaboration with domestic authorities, the sanctions regime continues to process and sanction cases independently from domestic law enforcement agencies. In the long run, this strategy sustains the distrust in domestic law enforcement.

E. Reinforcing Domestic Law Enforcement

The sanctions strategy is expected to promote law enforcement. However, this strategy may not only fail to reinforce the enforcement aimed at fraud and corruption, as discussed. Considered separately from other law enforcement initiatives, it may also fail to contribute to stronger domestic criminal justice institutions. This is a critical concern, since these institutions are better placed than the World Bank to deal with domestic fraud and corruption. By signaling that domestic criminal justice institutions cannot be relied on, the World Bank contributes to any domestic reputation that they are untrustworthy. Despite the de facto weaknesses of these institutions, it is essential for them to earn trust in their respective societies in order to start functioning efficiently. When law enforcement institutions are not trusted, members of the public find different, usually more informal ways to solve their disputes and complaints. In some cases, they make use of corruption or extortion to “solve” their problems.64 The difficulty for governments that are trying to strengthen their criminal justice systems is that they need to convince citizens to start using and trusting public law enforcement institutions at a time when these are still weak, albeit progressing.

The World Bank cannot trust organizations that are not reliable. At the same time, by refusing to trust any domestic courts the Bank’s sanctions regime upholds a public belief that domestic institutions lack the capacity to enforce the laws of their countries. Such perceptions effectively push both private parties and public officials toward alternative and informal strategies to solve their issues, often through bribery or corruption. This undermines development and is precisely what the sanctions regime is supposed to work against. In order to align the protection of funds with its overall development goals, the Bank must realize that it depends on domestic law enforcement institutions.

V. Toward More Efficient Solutions

The World Bank will have to start expecting borrower governments to secure an independent and legally accountable judiciary, protect the borrowed funds, and enforce their anticorruption regulations. What steps have been taken and can be taken to progress in this direction? First, instead of concentrating so much on improving the internal procedures of its own sanctions system, the organization should seek to identify and strengthen its various channels for demanding workable integrity systems in the countries where it finances development programs.65 INT already collaborates with law enforcement institutions at the national level when an offense has been committed. It encourages these institutions to investigate and prosecute actors involved in the offense who are outside the reach of the World Bank investigation, including government representatives. This collaboration could be extended—for example, INT representatives could serve as advisers to the prosecution and remain in the country as the case moves through the domestic criminal justice system.66

Second, the World Bank will have to envisage alternative scenarios for how fraud and corruption can be sanctioned. Instead of focusing so closely on suppliers, the Bank could also sanction governments for failing to investigate and prosecute guilty government representatives and suppliers. If this is difficult due to the weaknesses of law enforcement institutions and public finance management systems, why not demand collaboration for the strengthening of such institutions as a precondition for lending? And if that too is difficult, the Bank could dust off Paul Collier’s suggestion and introduce what he calls “ex post conditionality” on governance, which means rewarding strong performance with more financial support or more flexibility in spending.67 When it comes to checks and balances, it could make sense, as a second-best solution, to let an external third party control the government’s management of funds. Such services are offered by lawyers, accountants, or investigators who are independent from both the World Bank and the government in question.68

Regarding a more constructive use of financial power for pro-development purposes, the main development banks could collaborate in order to shut down borrower governments’ current option to “find another development bank” when one bank makes annoying demands for integrity. The development banks already collaborate on cross-debarment, so why not collaborate on conditions for lending as well? If they can cooperate on cross-debarment, they could collaborate on “cross-referrals”—meaning, for example, that loans will not be offered unless a given case is brought through the domestic criminal justice system.69 If it becomes too difficult to hold back financial support, given urgent development needs, the financial institutions could place restrictions on the government’s flexibility and control over spending—the less reliable the government, the more external control placed on each dollar spent. The funds could be kept under the control of the government’s more reliable institutions or transferred to more reliable city governments instead of leaving control with the central (monopoly) level.70 Of course, there can be a big leap from ideas about options to practical solutions. One case in which difficulties surfaced was the building of the Chad-Cameroon pipeline, which was completed in 2003 with multilateral and bilateral credit financing, including $100 million from the World Bank. In this case the World Bank insisted that the revenues from the pipeline be kept out of the Chadian president’s control, except for spending on stipulated development needs.71 Although in this case the strategy failed in the end, the World Bank should not give up on such efforts. Strategies to keep funds away from corrupt dictators are necessary to support the populations that are victimized and paralyzed by an autocratic regime. Despite the many difficulties, it should be possible for the Bank to let confirmed corruption risks have implications for lending, and to be stricter in its demands for integrity mechanisms and law enforcement institutions.

When it comes to the sanctions system, the World Bank may well continue to exclude suppliers not found trustworthy. Ideally, the governments should bear the main responsibility for excluding unreliable suppliers from their public procurement markets. In case they do not, the World Bank should maintain a list of firms not found eligible for contracts, although the mentioned tradeoffs associated with global and local debarment in short and long-term development perspectives need to be understood and taken into account. Besides, while it is important that the sanctions regime continue to ensure better due process standards, it is equally important that the World Bank and the sanctions regime encourage due process in domestic law enforcement institutions.

The World Bank could continue to reach its own judgment in each case by assessing the allegations or suspicions, entering into dialogue with public investigators, reviewing available evidence, and finally deciding whether to list a supplier as eligible or not eligible, based on specified criteria and an assessment of the market consequences. Combined with more transparent settlement procedures and efficient compliance control enforced in collaboration with national third-party players, it could be possible to bring debarred suppliers quickly back to their markets, thus protecting competition. It is difficult to predict how various World Bank stakeholders would respond to a simpler administrative solution with more attention to domestic law enforcement when cases arise, but a change in this direction seems necessary for the system to have a real impact on the ground that furthers the Bank’s main objectives. When it comes to collaboration with the other development banks, it is far more important to agree on how to efficiently demand integrity mechanisms on the government side than to cross-debar suppliers.

The World Bank is already considering alternative designs for its sanctions system and anticorruption strategy. However, for the Bank to alter its sanctions system and modify its strategies vis-à-vis governments, it needs the support of its board members, who represent governments. In particular, more efficient solutions will require the support of the most influential governments (or group of governments). Internationally, we have seen significant progress in governments’ willingness to sign and implement new anticorruption laws, and the World Bank has played a central role in promoting such developments. Nonetheless, the organization’s own suggestions on how to strengthen the international anticorruption agenda have sometimes met resistance when presented to the Bank’s own board. Government representatives, including those from countries scoring high on the Human Development Index, are sometimes hesitant to support more efficient anticorruption initiatives.72 Some of these countries also fail to enforce their foreign bribery laws and thus do not meet their own obligations when it comes to controlling corruption in developing countries. For the World Bank to be able to work more efficiently on anticorruption, however, it needs the full support of governments.

VI. Conclusion

The World Bank has a challenging mandate: the institution must offer financial support to developing country governments, including those with high corruption risks, while also trying to secure efficient use of the funds after they have been transferred and are no longer under the control of the World Bank. The sanctions regime must be understood in light of the organization’s need to secure the confidence of donor governments while offering financial support to countries where a high risk of fraud and corruption exists. One of its main strategies has been to develop an internal mechanism of law enforcement that targets fraud and corruption only on the part of suppliers to government contracts while respecting principles of due process.

Nonetheless, there are reasons to be concerned about the sanctions regime’s efficiency in contributing to the World Bank’s overall development goals. Considering not just the internal efficiency and transparency of the process but also the larger role and impact of the sanctions regime, it appears that the regime runs a high risk of not protecting funds; that it does not promote fair markets, but quite possibly promotes unfair markets; that it does not reduce the overall level of corruption; and, finally, that rather than rebuild and reform domestic law enforcement, it serves to undermine these institutions, which should be the primary venue for addressing fraud and corruption in client countries. The World Bank Legal Vice Presidency, which oversees the sanctions system, has shown interest in understanding the root causes of corruption in order to better assess the Bank’s anticorruption initiatives.73 A thorough exploration is likely to reveal that the present design of the sanctions regime is not based upon an understanding of root causes of corruption nor the domestic market consequences of sanctions, and that the calls for debates about the regime’s functionality are warranted.

In order to show strong accountability to donor governments and appear to be acting forcefully against corruption, the organization tries to solve the corruption problems in developing countries at the central headquarters level of the Bank itself—which means not solving them at all. In bypassing domestic law enforcement in client countries, the sanctions strategy declares the Bank’s low confidence in criminal justice institutions at the national level. This is a problem for developing country governments, which need law enforcement institutions that are trusted by the public in those countries to serve their intended law enforcement and conflict resolution functions. The World Bank should realize that it depends on these institutions in order to solve problems of corruption and fraud, in line with its development agenda.

• 1For an overview of initiatives, see Marie Chêne, U4 Expert Answer: Multilateral Development Banks' Integrity Management Systems (Dec. 10, 2010), available at http://www.transparency.org/files/content/corruptionqas/264_Multilateral_development_banks_integrity_management.pdf. Steps to harmonize measures were taken in 2007 with a joint task force on preventing corruption in lending. See Int’l Fin. Inst. Anti-Corruption Task Force, Uniform Framework for Preventing and Combatting Fraud and Corruption (2006), available at http://www.afdb.org/fileadmin/uploads/afdb/Documents/Generic-Documents/Uniform_Framework_for_Combatting_Fraud_and_Corruption.pdf. For an example of steps taken by one of the development banks, see World Bank, Guidelines on Preventing and Combating Fraud and Corruption in Projects Financed by IBRD Loans and IDA Credits and Grants (2011), available at http://www.wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2011/04/15/000356161_20110415011758/Rendered/PDF/611090BR0SecM21Disclosed04113120111.pdf.
• 2For an introduction to these regimes, see Norbert Seiler & Jelena Madir, Fight Against Corruption: Sanctions Regimes of Multilateral Development Banks, 15 J. Int’l Econ. L. 5–28 (2012). An introduction to the development banks’ sanctions regimes and the status of harmonization between them can be found in Frank A. Fariello & Conrad C. Daly, Coordinating the Fight against Corruption among MDBs: The Past, Present, and Future of Sanctions, 45 Geo. Wash. Int’l L. Rev. 253–68 (2013).
• 3The figure reported for 2007–2013 is 239 firms and individuals, but additional suppliers were sanctioned in 2014. See The World Bank Office of Suspension and Debarment, Report on Functions, Data and Lessons Learned 2007–2013, 24 (2014), available at http://siteresources.worldbank.org/EXTOFFEVASUS/Resources/OSDReport.pdf.
• 4For yearly statistics on referrals leading to prosecuted cases in the countries where the corruption has occurred, see Annual Reports, The World Bank, http://web.worldbank.org/WBSITE/EXTERNAL/EXTABOUTUS/ORGANIZATION/ORGUNITS/EXTDOII/0,,contentMDK:22203443~menuPK:5372841~pagePK:64168445~piPK:64168309~theSitePK:588921,00.html (last visited Oct. 4, 2015).  
• 5A historical perspective on the World Bank’s sanctions regime is provided in Dick Thornburgh et al., Report Concerning the Debarment Processes of the World Bank (2002), at 10-12 (2002), available at http://siteresources.worldbank.org/PROCUREMENT/Resources/thornburghreport.pdf. For legal details of the regime, see Sope Williams, The Debarment of Corrupt Contractors from World Bank-Financed Contracts, 36 Pub. Cont. L.J. 277, 289-304 (2007) and Pascale Hélène Dubois, Domestic and International Administrative Tools to Combat Fraud and Corruption: A Comparison of US Suspension and Debarment with the World Bank’s Sanctions System, 2012 U. Chi. Legal F. 195, 216-34 (2012). On legal aspects of the regime, with debate about achievements and reforms from the perspective of the World Bank, see generally Anne-Marie Leroy & Frank A. Fariello, The World Bank Group Sanctions Process and Its Recent Reforms (2012).
• 6The World Bank Group, Initiating Discussion Brief: Review of World Bank Group Sanctions Regime 2011–2014, Phase I Review: Stock-Taking, at 3 (2014), available at https://consultations.worldbank.org/Data/hub/files/consultation-template/consultation-review-world-bank-group-sanctions-systemopenconsultationtemplate/materials/sanctionsreview_initiatingdiscussionbrief.pdf. The document describes a two-part review process. The first part, a stock-taking exercise that collected critical comments from internal and external experts, was completed by 2014. A second part will consider the system’s efficiency and effectiveness. On its sanctions system website, the World Bank publishes reports about the sanctions system, including critical comments provided as part of the review process.
• 7The World Bank refers to countries receiving development support from the institution as client countries.
• 8Int’l Bank for Reconstruction and Dev., Articles of Agreement (as amended effective, art. I(i) (June 27, 2012), available at http://siteresources.worldbank.org/EXTABOUTUS/Resources/IBRD
  ArticlesOfAgreement_links.pdf.
• 9Indeed, although the Bank formally manages funds on behalf of all its members, rich as well as poor, donor governments are the ones that actually transfer funds to the institution and expect them to be spent accountably, while other members are recipients. 
• 10Int’l Bank for Reconstruction and Dev., Articles of Agreement, supra note 8, at art. III(5)(b).
• 11Int’l Bank for Reconstruction and Dev., Articles of Agreement, supra note 8.  
• 12The World Bank’s Independent Evaluation Group has reviewed the institution’s governance and anticorruption strategy for the period from 2007–2013 and provided an overview of the many approaches to promoting anticorruption initiatives in dialogue with governments, civil society, and the private sector. See Independent Evaluation Group, World Bank Country-Level Engagement on Governance and Anticorruption: An Evaluation of the 2007 Strategy and Implementation Plan, (2012), available at http://ieg.worldbankgroup.org/Data/reports/gac_eval.pdf.
• 13Kabur and Webb investigate attempts by international financial institutions to include governance-related conditionalities in their lending agreements with developing countries. The results of such attempts are meager; the conditionalities are often too loosely defined for enforcement and are rarely supported with actual cuts in spending. Although the study is more than a decade old, it is still relevant, as there are few developments to suggest that the results would be very different today. Devesh Kabur & Richard Webb, Governance-Related Conditionalities of the International Financial Institutions, in 6 G-24 Discussion Paper Series 1 (2000), available at http://213.154.74.164/invenio/record/16107/files/kupur.pdf.
• 14For details about the World Bank’s anticorruption strategies, see Governance and Public Sector Management, The World Bank, available at http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPUBLICSECTORANDGOVERNANCE/0,,contentMDK:22996457~pagePK:148956~piPK:216618~theSitePK:286305,00.html (last visited Oct. 4, 2015).
• 15For discussion of its justice reform initiatives, see Legal Vice Presidency, The World Bank: New Directions in Justice Reform: A Companion Piece to the Updated Strategy and Implementation Plan on Strengthening Governance, Tackling Corruption (2012), available at http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2012/09/06/000386194_20120906024506/Rendered/PDF/706400REPLACEM0Justice0Reform0Final.pdf. See also Governance, The World Bank, http://www.worldbank.org/en/topic/governance (last visited Oct. 4, 2015).  
• 16This speech was delivered at the World Bank’s annual meeting in October 1996. See James D. Wolfensohn, Annual Meetings Address, The World Bank (Oct 1, 1996), http://go.worldbank.org/PUC5BB8060.
• 17For a review of the process toward establishing the World Bank sanctions regime and how it has developed in parallel with other international anticorruption initiatives, see The World Bank Group Integrity Vice Presidency, Annual Update: Fiscal Year 2014, at 1-4 (2014), available at http://www.-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2014/10/09/000442464_20141009092921/Rendered/PDF/912470BR0SecM2090Box385330B00OUO090.pdf.
• 18Sebastian Mallaby, The World’s Banker: A Story of Failed States, Financial Crises, and the Wealth and Poverty of Nations 176 (2004).
• 19These attempts include external reports and internal papers on governance and anticorruption, as well as quite high-level debates on the balance between interfering too much and too little against corruption. See, for example, the legal note by Anne-Marie Leroy, the World Bank Group’s General Counsel, who defends the anticorruption agenda and underscores the importance of engaging with the criminal justice sector, even when , she seems to imply, high-level political figures are implicated in wrongdoing. Anne-Marie Leroy, Legal Note on Bank Involvement in the Criminal Justice Sector, The World Bank (Feb. 9, 2012), available at http://siteresources.worldbank.org/INTLAWJUSTINST/Resources/CriminalJusticeLegalNote.pdf.
• 20Mallaby, supra note 18, at 180-83.
• 21The World Bank’s ambition to “lead by example” has been stated in several of its publications, recently in the. See, for example, The World Bank, 2015 Sustainability Review, at 1 (2015). World Bank: https://openknowledge.worldbank.org/bitstream/handle/10986/22723/20150sustainability0review.pdf?sequence=1&isAllowed=y), available at https://openknowledge.worldbank.org/bitstream/handle/10986/22723/20150sustainability0review.pdf?sequence=1&isAllowed=y.
• 22See The World Bank Group Integrity Vice Presidency, supra note 17, at 2. See also Thornburg report, et al., supra note 5, at 16-19, and The World Bank, Sanctions Board Statute, art. III (2010), available at http://siteresources.worldbank.org/EXTOFFEVASUS/Resources/SanctionsBoardStatute_9_15_2010.pdf.
• 23See Thornburgh et al., supra note 5, at 1. Dick Thornburgh formerly served as U.N. Under-Secretary-General and as U.S. Attorney General.  
• 24The implications of the report were confirmed by the World Bank in a review of their integrity measures. See generally The World Bank Group Integrity Vice Presidency, supra note 17.
• 25Documents presenting details about the current regime and its historical development are available on the World Bank sanctions website. See Leroy & Fariello, supra note 5, at 2-6, 9-11; The World Bank Office of Suspension and Debarment, supra note 3, at 7-17; The World Bank Group Integrity Vice Presidency, Annual Update: Fiscal Year 2014 (2014), available at http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2014/10/09/000442464_20141009092921/Rendered/PDF/912470BR0SecM2090Box385330B00OUO090.pdf; supra note 17, at 1-4, 18-26.
• 26See The World Bank, Sanctions Procedures, § 1.02(a), Annex A (April 15, 2012), available at http://siteresources.worldbank.org/EXTOFFEVASUS/Resources/WBGSanctions_Procedures_April2012_Final.pdf.
• 27See id. at §§ 9.01, 10.01.
• 28The percentages total more than 100 percent because some cases included several forms of sanctionable conduct. See The World Bank Office of Suspension and Debarment, supra note 3, at 28.
• 29For a review of important World Bank reforms, see Leroy & Fariello, supra note 5, at 22-24.  
• 30See African Development Bank Group, Asian Development Bank, European Bank for Reconstruction and Development, Inter-American Development Bank Group, & World Bank Group, Agreement for Mutual Enforcement of Debarment Decisions, Preamble (Apr. 9, 2010), available at http://siteresources.worldbank.org/EXTOFFEVASUS/Resources/3601045-1377105390925/AgreementforMutualEnforcementofDebarmentDecisions(4.9.2010).pdf. The agreement is confirmed in a World Bank introduction to the sanctions regime from October 2010. See The World Bank Sanctions Regime: An Overview, available at http://siteresources.worldbank.org/EXTOFFEVASUS/Resources/Overview-SecM2010-0543.pdf.
• 31For information and debate, see Fariello & Daly, supra note 2, at 259-69. See also Stephen S. Zimmermann and Frank A. Fariello., Jr., Coordinating the Fight against Fraud and Corruption: Agreement on Cross-Debarment among Multilateral Development Banks, 3 World Bank Legal Rev 189, 189-204 (Eds Hasanne Cissé, Daniel D. Bradlow and Benedict Kingsbury, 2002).  
• 32See Leroy & Fariello, supra note 5, at 2; Thornburgh et al., supra note 5, at 14-15.  
• 33In fact, The Supreme Court of Norway concluded in a Norwegian foreign bribery case that the accused company should not be imposed any (additional) corporate criminal sanctions partly because, inter alia, “the act had resulted in extensive reactions from the World Bank.” Combined with the likely consequence of further debarment from public procurement markets, the court left out corporate penalty on the rationale that “viewed collectively [it] might have disproportionate consequences for the company”—a decision made with 3-2 dissenting votes regarding this rationale. For details, see HR-2013-1394-A, case no. 2012/2114), Norconsult: http://www.domstol.no/en/Enkelt-domstol/-Norges-Hoyesterett/Summary-of-Recent-Supreme-Court-Decisions/Summary-of-Supreme-Court-Decisions-2013/. For a discussion on the concept of a "Criminal justice system" see Linda Gröning, Towards a Theory of the Criminal Justice System in, 99 Nordisk Tidsskrift for Kriminalvidenskab 27, 27-46 (2012).
• 34See Staff Guide to INT (2007), Washington DC: The World Bank Group. See Integrity Vice Presidency, The World Bank Group, Guide to the Staff Rule 8.01: Investigative Process, at 4 (2011), available at http://siteresources.worldbank.org/INTDOII/Resources/INT_GuideToStaffRule801_final_REV.pdf.  
• 35The organization of the sanctions regime and the role of the sanctions board are described in Sanctions Board, The World Bank Group, Law Digest, 17-22 (2011), available at http://siteresources.worldbank.org/INTOFFEVASUS/Resources/3601037-1342729035803/
  SanctionsBoardLawDigest.pdf.
• 36This consequence occurs under public procurement debarment rules as well. See, for example, the point made in Angela B. Styles, et al., How Proposed Debarment Became Equal to Suspension, at 1-3 (Feb 2, 2015), available at https://www.crowell.com/files/How-Proposed-Debarment-Became-Equal-To-Suspension.pdf.
• 37The World Bank has improved its compliance-incentivizing mechanisms by adjusting the length of debarment in inverse proportion to the debarred actor’s collaboration with investigators and enforcement of a convincing compliance system. Once an offender is found to have regained its position as a trustworthy candidate for World Bank–financed operations, it should no longer be excluded from such opportunities. Hjelmeng and Søreide consider this provision of settlements to be an efficient aspect of the Bank’s debarment strategy. Erling Hjelmeng & Tina Søreide, Debarment in Public Procurement: Rationales and Realization, in Integrity and Efficiency in Sustainable Public Contracts 215–32 (Gabriella M. Racca & Christopher R. Yukins eds., 2014).
• 38The formal rules of the system are proposed through a process coordinated by the Legal Vice Presidency, which plays a role similar to that of a justice department in a national government system. The proposals are approved by the Executive Directors of the World Bank. This is a group of country representatives with oversight responsibility and authority to appoint and remove the World Bank president; it is the nearest the World Bank gets to a parliament. This is, however, also the body with power to interpret the IBRD Articles of Agreement (a form of authority usually vested in a different institution in countries, such as the Higher Court). The day-to-day implementation of “the law,” and thus the development of practical law enforcement processes, is managed by the Evaluation Officers. Their daily judgments are influenced by the body of case law from the Sanctions Board and, to some extent, by relevant cases from courts in member countries.
• 39For a comparison of the World Bank sanctions regime with debarment in the US public procurement system, see Dubois, supra note 5, at 195.
• 40See Leroy & Fariello, supra note 5, at 12.
• 41Despite the confidentiality around settlements, the World Bank operates the regime transparently, and publishes statistics about the sanctions regime. See The World Bank Office of Suspension and Debarment, supra note 3, at 23-31.
• 42The World Bank, supra note 26, at § 1.01(a).
• 43See Frank A. Fariello, Jr. & Giovanni Bo, Development-Oriented Alternatives to Debarment as an Anticorruption Accountability Tool, 6 World Bank Legal Rev., 415, 429 (2015).
• 44After the Thornburg report’s recommendation, the World Bank operates as if it has the burden of proof, see The World Bank Group, supra note 6. For a sanction to be imposed, the World Bank must be able to document the offense. In cases documented by a court or in cases of an alleged offense, the INT will be inclined to investigate the facts.
• 45See generally The World Bank Office of Suspension and Debarment, supra note 3.  
• 46See World Bank Sanctions Procedures (as adopted by the World Bank as of January 01, 2011), The World Bank (Jan. 2011), available at http://siteresources.worldbank.org/EXTOFFEVASUS/Resources/WBGSanctionsProceduresJan2011.pdf;. See also Seiler & Madir, supra note 2, at 13-15, 28-29.  
• 47See Public Sector Governance - Governance and Anti-Corruption Diagnostics, The World Bank (Oct. 4, 2015), available at, http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPUBLICSECTORANDGOVERNANCE/0,,contentMDK:23410781~pagePK:210058~piPK:210062~theSitePK:286305,00.html.
• 48Criminal justice systems around the world are subject to severe challenges in terms of insufficient capacity and various forms of corruption. See generally Richard E. Messick & Sofie A. Schutte, Corruption Risks in the Criminal Justice Chain and Tools for Assessment, 6 U4 1 (Mar. 2015), available at http://www.u4.no/publications/corruption-risks-in-the-criminal-justice-chain-and-tools-for-assessment/. For data on the performance of criminal justice systems and risk of corruption, see World Justice Project (Oct. 4, 2015), http://worldjusticeproject.org.
• 49By intuition, the bidders have more to gain by sharing cartel profits when there are fewer bidders, and with fewer bidders it is also easier to organize an illegal cartel. For an explanation, see Jean-Jacques Lafont & Jean Tirole, A Theory of Incentives in Procurement and Regulation 536-550 (1993).
• 50See Fariello, Jr. & Bo, supra note 43, at 419-23 (describing similar concerns with the sanctions regime).
• 51According to Iimi (2011), who studied worldwide infrastructure projects financed by aid or development loans, the average number of bidders is 5.2 in the water and sewerage sector, 6.2 in the roads sector, and 4.6 in the electricity sector. In the majority of electricity works and water auctions only two or three firms were competing for the contracts. Atsushi Iimi, Public Procurement: Learning from the Experience of Developing Countries, in Emerging Issues in Competition, Collusion, and Regulation of Network Industries 119, 129-30 (Estache ed., 2011). See also Antonio Estache & Atsushi Iimi, Quality or Price? Evidence from ODA-Financed Public Procurement, 40 PUB. FIN. REV. 435 (2012) (these details were first described by Emmanuelle Auriol & Tina Søreide, infra note 53). The World Bank’s observation of higher risk of collusion when there are few bidders is described in the Fraud and Corruption Awareness Handbook: A Handbook for Civil Servants Involved in Public Procurement, The World Bank 33-36 (2013), available at https://openknowledge.worldbank.org/bitstream/handle/10986/18153/877290PUB0Frau00Box382147B00PUBLIC0.pdf?sequence=1. For a discussion about the risk of cartelization due to few bidders on World Bank financed projects, see Curbing Fraud, Corruption, and Collusion in the Roads Sector, The World Bank Integrity Vice Presidency 12-15 (June 2011), available at http://siteresources.worldbank.org/INTDOII/Resources/Roads_Paper_Final.pdf. Hence, debarment applied as an anticorruption strategy in markets with few suppliers can easily increase the risk of cartel collaboration between remaining suppliers, and it is not necessarily a welfare-enhancing strategy even if procurement agents are corrupt. The mechanisms at play are analyzed in Emmanuelle Auriol & Tina Søreide, An Economic Analysis of Debarment, NHH Dep’t of Bus. and Mgmt. Sci. Discussion Paper No. 2015/23 (Sept. 11, 2015), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2662374.
• 52On this point, the World Bank regime is better designed than many national systems when it comes to incentivizing compliance. Protection of markets may require placing more emphasis on debarment of individuals involved in the listed forms of crime, for example, a group of leaders, rather than sanctioning the entire organization that they represent.
• 53See, for example, the emphasis given to anticorruption by the U.N., the OECD, and the G20. Over the past 15 years, we have witnessed a sharp increase in the amount of various governance and business indicators, many of them including estimates of corruption.  
• 54Press release - Corrected: World Bank Debarments and Suspensions Play Important Role in Fighting Corruption, Transparency International (June 26, 2014), https://www.transparency.org/news/pressrelease/world_bank_debarments_and_suspensions_play_important_role_in_fighting_corru.
• 55For explanation of how the size of bribes depends on shifts in the supply and demand functions, see generally Susan Rose-Ackerman, Corruption: A Study in Political Economy (1978). The argument follows from the analysis of bribes as a price that clears the market between those involved in corruption. For an explanation, see Tina Søreide & Susan Rose-Ackerman, Corruption in State Administration, in Research Handbook on Corporate Crime and Financial Misdealing, 6-7 (Jennifer H. Arlen ed. (forthcoming 2016)).  
• 56An analysis of the market consequences of competition between firms with different risk of facing sanctions as a consequence of detected corruption (for example, if home country governments monitor and enforce foreign bribery regulations differently) is provided by Kjetil Bjorvatn and Tina Søreide, Corruption and Competition for Natural Resources, 21 Int’l Tax & Pub. Fin. 997, 999-1001 (2014).
• 57On trust as a functional condition, see Russell Hardin, Trust (2006); Niklas Luhmann, Trust and Power: Two Works By Niklas Luhmann (1979); Margaret Levi, A State of Trust, in Trust and Governance 77 (Valerie Braithwaite and Margaret Levi eds., 1998); Niklas Luhmann, Familiarity, Confidence, Trust: Problems and Alternatives, in Trust: Making and Breaking Cooperative Relations 94 (Diego Gambetta ed., 1988). For the functionalist accounts of trust, see James S. Coleman, Foundations of Social Theory (1990); Anthony Giddens, The Consequences of Modernity (1990); Piotr Sztompka, Trust: A Sociological Theory (1999); Diego Gambetta, Can We Trust Trust?, in Trust: Making and Breaking Cooperative Relations 213 (Diego Gambetta ed. 1988).
• 58See, for example, Monika Bauhr et al., Does Corruption Cause Aid Fatigue? Public Opinion and the Aid-Corruption Paradox, 57 Int’l Stud. Q. 568, 570 (2013).  
• 59See Coleman, supra note 57, at 180-96 (on trust building in networks of organizations).
• 60“Comprehensiveness” refers to the regime’s criminal justice characteristics, especially with respect to the significant weight placed on due process.
• 61See Fariello, Jr. & Daly, supra note 2, at 257-59; Fariello, Jr. & Bo, supra note 43, at 415-17.
• 62The emphasis on procedures is explained in John Coogan et al., Combatting Fraud and Corruption in International Development: The Global Impact of the Multilateral Development Banks’ Sanctions Regimes, 22 J. Fin. Crime 228, 228-231, 238 (2015). See also Pascale H. Dubois & Aileen E. Nowlan, Global Administrative Law and the Legitimacy of Sanctions Regimes in International Law, in Anti-Corruption Policy: Can International Actors Play a Constructive Role? 201, 212 (Susan Rose-Ackerman & Paul D. Carrington, eds., 2013).
• 63See Fariello, Jr. & Bo, supra note 43, at 420.
• 64See Hardin, supra note 57; Sztompka, supra note 57, at 116-17; Rasmus H. Wandall, Trust and Legal Governance: A Case Study of Ethiopian Criminal Justice, 42 J.L. & Soc’y 283, 285-87 (2015).
• 65As emphasized above, due process in sanctions decisions is necessary. What we highlight is that proper procedures will not diminish other problems associated with the system. For an economic perspective on the effectiveness of integrity systems, controls and compliance systems in state institutions, see Søreide & Rose-Ackerman, supra note 55, at 6-11.
• 66For further debate on how the World Bank can better complement the sanctions regime, see Fariello, Jr. & Bo, supra note 43, at 423-35.
• 67Paul Collier, Learning from Failure: The International Financial Institutions as Agencies of Restraint in Africa, in The Self-Restraining State: Power and Accountability in New Democracies 313, 322-30 (Andreas Schedler et al. eds., 1999).
• 68A certain independence from the World Bank task managers who oversee the spending is clearly needed, according to the basic principle of checks and balances, since task managers are mandated to complete projects and may find dealing with corruption an obstacle to their aims. Many of them will report their suspicions, but there are risks that these reports will be ignored, especially if they are made orally and not in writing. For a review of such issues in the development community, using Norwegian development aid as a case in point, see Eirik G. Jansen, Don’t Rock the Boat: Norway’s Difficulties in Dealing with Corruption in Development Aid, in Corruption, Grabbing and Development: Real World Challenges 186 (Tina Søreide & Aled Williams eds., 2014).
• 69The term referrals in this context refers to the package of evidence sent from World Bank investigators to investigators at the national level, so that the latter will have a better starting point for prosecuting their civil servants and elected leaders. Few referrals are acted upon, however, and demanding law enforcement is critically important.  
• 70INT investigators have several times conducted ex ante assessments of corruption risks prior to funding. For examples, see The World Bank Group Integrity Vice Presidency, supra note 17, at 21-23. In a vast infrastructure program in Nepal, for example, where funds were transferred to local communities, the local decision makers’ discretionary authority over spending was determined based upon an assessment of the corruption risk. Such ex ante assessments are now conducted more systematically. The World Bank Group Integrity Vice Presidency: Annual Update Fiscal Year 2015, The World Bank Group Integrity Vice Presidency 40-41 (2015), available at http://web.worldbank.org/WBSITE/EXTERNAL/EXTABOUTUS/ORGANIZATION/ORGUNITS/EXTDOII/0,,contentMDK:23691324~pagePK:64168445~piPK:64168309~theSitePK:588921,00.html.
• 71The story of the Chad-Cameroon pipeline is well told in Steve Coll, Private Empire: ExxonMobil and American Power 154-76 (2012). The Padma Bridge project in Bangladesh is another case where attempts were made to insist on integrity and secure development results despite corruption-related challenges at the political level. See Statement on Padma Bridge, The World Bank (June 29, 2012), http://www.worldbank.org/en/news/press-release/2012/06/29/world-bank-statement-padma-bridge.
• 72Substantial variation in governments’ willingness to react against corruption and enforce their anti-bribery legislation was revealed by an OECD review of completed. See OECD Foreign Bribery Report: An Analysis of the Crime of Bribery of Foreign Public Officials, The Organisation for Economic Co-operation and Development 15-22 (2014), available at http://www.oecd-ilibrary.org/docserver/download/2814011e.pdf?expires=1446334104&id=id&accname=guest&checksum=DDCFAC12D47DC0827B89FB3D2BF3DE14.
• 73See, for example, Tina Søreide, Drivers of Corruption: A Brief Review, The World Bank (2014), available at https://openknowledge.worldbank.org/bitstream/handle/10986/20457/916420PUB
  0Box30UBLIC009781464804014.pdf?sequence=1.