Holding the World Bank accountable for leakage of funds from Africa’s health sector

Fatma E. Marouf

Health and Human Rights 12/1

Published June 2010


This article explores the accountability of international financial institutions (IFIs), such as the World Bank, for human rights violations related to the massive leakage of funds from sub-Saharan Africa’s health sector. The article begins by summarizing the quantitative results of Public Expenditure Tracking Surveys performed in six African countries, all showing disturbingly high levels of leakage in the health sector. It then addresses the inadequacy of good governance and anticorruption programs in remedying this problem. After explaining how the World Bank’s Inspection Panel may serve as an accountability mechanism for addressing the leakage of funds, discussing violations of specific Bank policies and procedures that would support a claim related to leakage and examining the relevance of human rights concerns to such as claim, the article explores some of the Panel’s limitations and the positive steps taken to address these concerns.


This article explores the accountability of international financial institutions (IFIs), such as the World Bank, for human rights violations as they relate to the massive leakage of funds from sub-Saharan Africa’s health sector. The actual extent of such leakage has slowly surfaced over the past decade through the use of Public Expenditure Tracking Surveys (PETS) that trace the funds allocated to a specific sector from the central government to frontline providers. While the quantitative results of such surveys are rarely discussed outside the donor community, they reveal that, in many cases, the vast majority of funds never reach clinics and hospitals that actually provide health services. While leakage is not equivalent to corruption, it certainly signals a need for monitoring, evaluation, and investigation. Large levels of leakage also indicate that civil society is not able to participate effectively in demanding the resources to which they are entitled.

In addressing this topic, the article focuses on four related aspects of leakage and its implications for health and human rights. First, it summarizes the quantitative results of public expenditure tracking surveys performed in six African countries, all of them showing disturbingly high levels of leakage in the health sector. Second, it addresses the inadequacy of good governance and anticorruption programs in remedying this problem. Third, it explains how the World Bank’s Inspection Panel may serve as an accountability mechanism for addressing the leakage of funds, discussing violations of specific World Bank policies and procedures that would support a claim related to leakage and examining the relevance of human rights concerns to such a claim. Finally, the article explores some of the Panel’s limitations and the positive steps taken to address these concerns.

While arguments regarding the Bank’s direct or indirect accountability under international human rights law are important, this paper takes a different approach, focusing on how to utilize the Bank’s internal accountability mechanism — the Inspection Panel — as a way to highlight the human rights issues related to leakage and push forward interpretations of the Bank’s internal policies that conform to international human rights law. The main advantage of this approach is that it circumvents many of the legal hurdles involved in attempting to establish direct or indirect accountability. Another benefit of this approach is its potential application to other IFIs, which have very similar internal accountability mechanisms modeled after the World Bank’s Inspection Panel.1

The massive leakage of funds from Africa’s health sector

Historically, lack of information about public spending in key social sectors, such as health and education, has made it difficult to hold governments accountable.2 Poor reporting, highly aggregated data, and discretionary allocations all contribute to the problem, creating nontransparent processes that camouflage how allocated funds are actually being used.3 Traditional methods for assessing outcomes in service delivery include household surveys and social impact assessments, which generally provide qualitative information. More recently, donors, such as the World Bank, have developed the Public Expenditure Tracking Survey (PETS) as a diagnostic and monitoring tool to understand problems in budget execution.4 PETS tracks the flow of resources through various levels of government to the frontline providers (for example, health care clinics and schools) by collecting data at each level and comparing sources, thereby determining where resources are being absorbed and where they are going astray.5 During the past decade, the proliferation of PETS throughout Africa has exposed massive leakage of funds in the health and education sectors. These results, however, have surprisingly received little publicity.

“Leakage” is generally defined as the share of resources intended for, but not received by, frontline providers.6 While “leakage” is not equivalent to corruption, it is often “a good indicator of corruption.”7 Consequently, evidence of substantial leakage in a given sector should invite further investigation into possible corruption. The results of PETS conducted in Uganda, Tanzania, Ghana, Zambia, Chad, and Kenya all provide firm estimates of leakage in the health sectors. Key findings from these PETS are summarized below:

Uganda’s 2000 PETS indicated a leakage rate of 70% for drugs and supplies, with the most high-demand drugs (such as those used to treat malaria) being the least available, while the 1999 PETS indicated that nearly 90% of patients made “informal payments” to providers. PETS performed in 2003–2004 indicated that absenteeism rates among health staff were greater than 35% and that 109 staff out of a total of 465 appeared to be “ghosts.”8

Tanzania’s 1999 PETS found that 41% of non-wage expenditures in health care never reached frontline providers. A follow-up survey in 2001 revealed that decentralized funds sent to the districts were all consumed at the district level and that there was no cash funding below the sector/district level (i.e. only in-kind material reached the health care facilities).9

Ghana’s 2000 PETS found that 80% of non-wage health expenditures never reached frontline providers in 1998–1999 and that the majority of resources reaching the district level health offices were in-kind materials, not cash.10 Consequently, clinics depended mainly on internally generated funds (i.e. user fees and payments for drugs) for all non-salary recurrent expenditures. Indeed, the study found that 40% of clinics did not receive any cash assistance from the government and depended solely on internally generated funds for non-salary recurrent expenditures.11

Chad’s 2004 joint PETS-QSDS found that 73% of allocated non-wage recurrent expenditures never reached the regional health delegations, and over 99% of the non-wage recurrent expenditures allocated to regions never reached local health centers.12 Based on the results of PETS, the study team estimated that, if all public resources officially budgeted for regional delegations had reached the frontline providers in 2003, the number of patients seeking primary health care in Chad would have more than doubled during the year.13

Kenya’s 2004 PETS found that 38% of the total funds allocated to the health centers never reached their destination, 25% of user fees “leaked’ from the facility level, and 37% of the community development funds “leaked” at the facility level.14 PETS also found that provinces and districts were unaware of budgets and programs, the supervisory capacity of provincial and district authorities were insufficient, and Kenya’s financial and accounting systems were inadequate.

Lastly, a 2007 study, which examined leakage and delays of salaries in Zambia’s health sector during 2005–2006, found that 15% of health workers did not receive all the salaries due to them during the past 12 months, 78% of health sector employees experienced delays in receiving their salaries, 16% received less than their full salary, and 10% of health sector employees had to pay an “expediter’s fee” to obtain their salary.15

PETS performed in Rwanda (2000 and 2004), Mozambique (2002), Nigeria (2002), Senegal (2002), Cameroon (2003), and Namibia (2003) likewise indicated significant leakage of funds, although they did not provide quantitative results.16 While PETS have clearly provided a powerful tool that identifies problems in public expenditure management, revealing massive leakage of funds, the World Bank has surprisingly found that “the costs and time demands involved” make the method “impractical for universal application.”17

The high rates of leakage in Africa’s health sectors, particularly with respect to non-wage expenditures like medical supplies, have devastating consequences for the delivery of health services.18 While there are many complex reasons why Africa’s health sector remains poor despite significant increases in donor aid, leakage of funds is one important reason that should not be ignored.19 The World Bank’s failure to monitor the use of funds and its refusal to address directly large-scale leakage render it at least partially responsible for these factors’ negative impact on access to health care.

A recent review of the World Bank’s health work, conducted by the Independent Evaluation Group (IEG), an in-house unit, found that monitoring of health programs “remains weak” and “evaluation is almost nonexistent,” leading to various problems, including “an inability to measure the effectiveness of interventions.”20 Performance in Africa was “particularly weak,” with 73% of projects categorized as unsatisfactory. Moreover, instead of improving over time, IEG states that “outcome ratings in Africa in recent years have shown steady declines.”21 According to a March 2009 review of the health sector strategy approved by the Bank in 2007, only 25% of projects in sub-Saharan Africa had satisfactory outcomes.22 These evaluations confirm that the Bank is failing to take its monitoring obligations seriously, despite the alarming results of PETS. If African countries are to make any real progress towards meeting their Millennium Development Goals, serious efforts must be made to hold accountable not only borrowing governments, but also the donors.

The inadequacy of good governance and anticorruption programs

Ever since former World Bank President James Wolfensohn spoke out against corruption in 1996, the Bank has paved the road for donors to take various measures to try to ensure that aid reaches the intended beneficiaries. These measures include governance and anticorruption initiatives, as well as establishing internal accountability mechanisms. While such efforts represent important strides forward, the IEG’s recent evaluations confirm that the public sector reforms are insufficient for addressing accountability concerns, especially the large-scale leakage described above. In addition, the public sector reforms are all one-sided, focusing exclusively on recipient governments and ignoring the World Bank’s joint responsibility for ensuring that aid money actually benefits the poor. The discussion below focuses specifically on the World Bank’s efforts, since the Bank considers itself a leader among donors who support anticorruption efforts.

The World Bank’s main public sector reforms (PSR) target public financial management, civil service, tax administration, and corruption. While such reforms have their roots in the 1980s, they expanded considerably during the 1990s.23 The reforms related to public expenditure management include, for example, restructuring budgets, implementing strong central controls, computerizing financial information systems, and charging user fees for certain services.24 The Bank’s anticorruption efforts focus primarily on these indirect methods rather than tackling corruption head-on.25 Where the Bank has supported direct anticorruption efforts, these primarily target bureaucratic (i.e. petty) corruption, such as bribe taking, skimming paychecks, and nepotism in appointments.26 The Bank’s petty corruption initiatives include creating an anticorruption commission with a low-level mandate, requiring public officials to disclose their assets, investigating and prosecuting officials who have unaccounted wealth or take bribes, and establishing a code of conduct for public officials.27

These reforms, unfortunately, have not yielded significant results, although their outcomes are “inherently difficult to measure.”28 In terms of both the Country Policy and Institutional Assessment (CPIA) ratings and corruption rankings, low-income countries that receive aid for PSR (such as those in sub-Saharan Africa) have performed little or no better than those that did not receive any PSR funding.29 In other words, the aid provided for PSR reforms had no noticeable impact in the poorest countries. The IEG’s 2008 review of the World Bank’s public sector reforms specifically notes that “[i]mplementation of the budget — procurement and financial management — has still received too little attention” and that “civil service reform has been relatively unsuccessful.”30 Commentators have also asserted that the Bank’s focus on public expenditure management and policy-making capacity remains very technical, and therefore, does not include human rights commitments as a key part of national ownership.31

Making matters worse, the IEG found that most of the Bank’s direct anticorruption initiatives were “rarely invoked except to settle political scores.”32 Only in “rare cases” has the Bank addressed state-capture (i.e. grand corruption), such as embezzlement of public funds.33 The isolated cases where the Bank has addressed state-capture involved “deep political and economic crises” that “exposed the corruption of old regimes and brought in new ones dedicated to a fresh start, such as in Indonesia in the late 1990s and Nigeria after 2003.”34 In failing to address state-capture directly, the Bank essentially turns a blind eye to the massive levels of leakage discussed above, which cannot be due to petty corruption alone.

The IEG’s 2009 review of the Bank’s internal controls further underscores the inadequacy of PSR and anticorruption efforts. This assessment was “the first of its kind, not only for the Bank but also for all international financial organizations.”35 The IEG found that the Bank’s main weaknesses pertain to fiduciary controls, resulting in serious risks of fraud and corruption, especially when combined with significant deficiencies in the areas of risk management, project financial management, and procurement.36 Given that public sector reforms have proven largely unsuccessful and that the Bank is unwilling to tackle state-capture head-on, the current measures for ensuring that aid reaches the intended recipients are clearly inadequate. In what follows below, I advocate using the World Bank’s Inspection Panel as one way to compel the Bank to take the issue of leakage more seriously, raising not only violations of the Bank’s own internal policies, but also violations of closely related human rights norms.

Using internal accountability mechanisms to hold IFIs accountable for funds leakage

Since the World Bank is failing to take adequate measures to address the leakage of funds, how might the Bank be held accountable for its acts or omissions? While some international human rights treaties, as well as highly influential declarations, guidelines, and comments, do stress the importance of international cooperation in protecting human rights, holding the Bank directly accountable for human rights violations remains a challenge, since international human rights treaties do not impose obligations on intergovernmental organizations.37 Moreover, although the Bank, unlike the International Monetary Fund, enjoys only limited immunity, the immunity clause in the Bank’s Articles of Agreement presents another potential hurdle to direct accountability.38

Despite presenting legal hurdles of its own, arguing that the Bank should be held indirectly accountable is one way to circumvent these challenges.39 In 2001, the International Law Commission (ILC) adopted Articles on the Responsibility of States for Internationally Wrongful Acts (ARSIWA). Over the next several years, it developed Draft Articles on the Responsibility of International Organizations (DARIO). The ILC’s comments indicate that the ARSIWA and the DARIO must both be taken “fully into account” in analyzing the attribution of conduct, but gaps and tensions between the two documents remain, which are slowly being developed and clarified through case law.40 Recent decisions by the European Court for Human Rights and the UK House of Lords have interpreted the DARIO only in military context, finding that the acts of a Member State may be attributed to an international organization only if it exercises “effective control” over State organs.41 Courts have yet to address whether the “effective control” doctrine might apply to the actions of an international financial institution, although the ILC has noted that, in many non-military situations, the conduct may have to be attributed to both the State and the international organization.42

Using the Inspection Panel to bring claims based on leakage of funds
Rather than focusing on the Bank’s direct or indirect accountability under international human rights law, areas where the legal doctrine is still being developed, this paper focuses on the Bank’s internal accountability mechanism, the Inspection Panel. The Inspection Panel can be an effective means, I suggest, for practitioners to highlight the human rights issues related to leakage and to advance interpretations of the Bank’s internal policies that conform to international human rights law. As an established, noncontroversial forum for claims against the Bank, the Inspection Panel may serve as a useful, albeit imperfect, proxy to challenge the human rights violations associated with the massive leakage of funds from Africa’s health sector. This approach is also applicable to other IFIs, which have very similar internal accountability mechanisms modeled after the World Bank’s Inspection Panel.

The process for accessing the Inspection Panel is relatively simple. Two or more individuals (or an organization) may submit a written request for inspection to the Panel if they believe that their “rights or interests have been or are likely to be directly affected by an action or omission of the Bank as a result of a failure of the Bank to follow its own operational policies and procedures.”43 The Panel, composed of three members who are randomly selected from a list of experts, submits the request to Bank Management for a response. After reviewing both the request and Management’s response, the Panel makes a recommendation to the Bank’s Board about whether the allegations merit an investigation. If the Board agrees to investigate, the Panel conducts the investigation and then submits its findings to the Board and Bank Management. Management then submits its own recommendations to the Board, and the Board makes the final decision on what actions to take.

Since its creation in 1993, the Panel has received approximately 56 complaints, 18 of which came from Africa.44 While the Panel has addressed various projects that had a negative impact on the right to health (for example, due to environmental contamination or displacement of people), it has not yet received a request to investigate the leakage of funds from the health sector (or any other sector). Such leakage can and should be brought to the Panel’s attention because it stems, at least in part, from the Bank’s failure to implement several of its own policies and procedures. The Bank’s Articles of Agreement provide that “[t]he Bank shall make arrangements to ensure that the proceeds of any loan are used only for the purposes for which the loan was granted.”45 This provision is reiterated in the Bank’s Operational Policy on Project Supervision (OP 13.05), as well as the Operational Policy on Financial Management (OP 10.02). OP 13.05 not only stresses the importance of monitoring and evaluation to identify problems, but also notes the need to prepare implementation completion reports in order to “account for use of Bank resources.”46 Moreover, OP 10.02 requires the Bank to “take… action to rectify the situation” if the borrower fails to maintain acceptable financial management arrangements or to submit the necessary financial reports by their due dates.

Finally, in many cases involving leakage, the Bank may have violated its Disclosure Policy. Unless budget allocations and the results of surveys, such as PETS, are made available to civil society, people cannot effectively demand the resources to which they are entitled. As noted above, in Kenya, PETS found that provinces and districts were unaware of budgets and programs.47 In Chad, PETS demonstrated that “the total lack of transparency of budgetary information at the regional and district levels greatly facilitates the capture of the MOH budget at the central level.”48 This lack of transparency is exacerbated by the Bank’s narrow Disclosure Policy which provides only a limited list of the information that may be disclosed. Recognizing this issue, the Bank recently launched a comprehensive review of its policy and is moving toward disclosing any information that is not on a list of exceptions.49 Under the new policy, which will become effective in July 2010, the public should have access to numerous documents that are relevant to identifying and quantifying leakage, which include audited financial reports. Failure to provide or disseminate these documents could support a request for investigation by the Panel.

Linking internal policy violation to human rights obligations
The internal policy violations discussed above are closely related to both procedural and substantive human rights norms. Thus, in drafting a request to the Inspection Panel, individuals or groups may strengthen their claims by grounding them in the language of human rights, as well as in the texts of internal policies. Recent changes in the Bank’s general attitude toward human rights, as well as some recent decisions by the Inspection Panel, suggest that using a human rights framework to interpret the Bank’s internal policies may be an effective way to hold the Bank accountable for acts and omissions that implicate human rights.

The Bank policies discussed above are particularly linked to procedural obligations, as monitoring, evaluation, supervision, sound management, and investigation are the basic means of preventing violations of human rights, including the right to health and the closely related right to life. While human rights treaties do not specify how rights should be protected and therefore do not explicitly mention the obligation to monitor, evaluate, or investigate, according to Naomi Roht-Arriaza, “[i]t is now widely accepted that references to ‘ensuring’ the full enjoyment of the enumerated rights in comprehensive human rights treaties impose affirmative obligations.”50 In fact, “bodies charged with monitoring compliance with these human rights treaties have insisted that a series of steps are required to ensure the full enjoyment of the rights at issue, at least those — the right to life and to be free of torture — where violations are of the most basic rights and thus are of special concern.”51

The Human Rights Commission (HRC), for example, has directly rejected the argument that State parties have no duty to investigate human rights violations, since there is no explicit provision for such as duty.52 Indeed, the HRC has found that State parties have “a duty to investigate thoroughly alleged violations of human rights, and in particular . . . violations of the right to life, and to prosecute criminally, try and punish those held responsible for such violations.”53 Although the duty to investigate and provide a remedy generally arises in cases involving forced disappearance, torture, or death, the large-scale leakage of funds from Africa’s health sector implicates the right to life by preventing countless individuals from obtaining life-saving medical care. Indeed, the European Commission has interpreted the right to life as not only preventing the unlawful taking of life, but also creating a positive obligation to take appropriate steps to safeguard lives.54 Moreover, some courts have specifically found that the right to life creates positive obligations to provide adequate and competent medical care.55 Consequently, even if the duty to investigate were limited to cases involving violations of the most fundamental rights, the massive leakage of funds from the health sector arguably falls within that class of cases.56

In addition, the Bank’s disclosure policy is closely linked to the right to the information, which is both substantive and procedural in nature.57 Utilizing human rights language may be especially helpful in situations where the Bank’s current disclosure policy or its revised policy falls short of providing the desired information. Groups such as the Global Transparency Initiative point out several shortcomings with the Bank’s proposed revisions to its disclosure policy, including overly-broad exceptions, third party veto power over release of information, lack of detail about how requests for information will be processed, and the absence of an independent appeals body.58 Even the IEG recently recommended “making better information public . . . in ways that stimulate public demand for more efficient and less corrupt service delivery.”59 Combining language about the right to information with language about the Bank’s disclosure policy may therefore be the most persuasive and powerful means of triggering an investigation by the Panel.

The right to participation is another right, with both substantive and procedural aspects, that is closely related to the right to information. The World Bank’s own diagnostic studies on governance and corruption show that “external voice” (that is, the ability of non-governmental actors to demand information and accountability) is likely the key determinant in adherence to pro-poor priorities and financial probity.60 In 2008, the IEG confirmed that “corruption can be substantially reduced only when the supply-side reforms are complemented by systematic efforts to increase the citizens’ capability to monitor and challenge abuses of the system and to inform the citizens about their rights and entitlements.”61 Yet “[i]ncreasing awareness of the potential role of civil society in fighting corruption has only materialized in a few of the Bank’s anticorruption lending programs.”62 The Bank’s failure to protect civil society’s right to participation in its anticorruption programs clearly contributes to the problem of leakage and should therefore be raised as an issue in requests for inspection.63

Although raising human rights obligations in this manner may, at first glance, appear irrelevant or futile, recent changes within the Bank and some decisions by the Inspection Panel provide support for this approach. After years of hiding behind its mandate not to interfere in political affairs and resisting a human rights-based approach to development, the Bank has finally accepted that its activities implicate human rights.64 In 2005, Senior Vice President and General Counsel Roberto Danino issued a legal opinion, which stated that “the Articles of Agreement permit, and in some cases require, the Bank to recognize the human rights dimensions of its development policies and activities since it is now evident that human rights are an intrinsic part of the Bank’s mission.”65 The following year, the new Senior Vice President and General Counsel Ana Palacio confirmed that “the Bank can and sometimes should take human rights into consideration as part of its decision making process.”66 She explained that “[t]he challenge we now face is to clarify how these legal concepts should be specifically incorporated into the work of the Bank in order to further its mission of sustainable and equitable development.”67

Palacio recognized not only that the World Bank’s activities have a human rights dimension, but also that the legal principles enshrined in human rights norms provide a “baseline” for assessing development policies and programs. She further acknowledged that, in certain cases, human rights norms generate actionable legal obligations that arise from both international treaties and national laws. In particular, Palacio stressed that human rights principles are relevant to “[a]reas of governance or the legal empowerment of the poor.” The rights violations related to leakage, including lack of monitoring and investigation, failure to disclose information, and inadequate participation are clearly relevant to both of these areas.

The Bank’s Inspection Panel has also opened the door for claimants to “incorporate a wide range of human rights-related concerns into their complaints.”68 A few of its decisions during the past decade explicitly address the relevance of human rights to the Bank’s work. In investigating the Bank’s Chad-Cameroon Pipeline Project (2002), Edward S. Ayensu, the former Chairperson of the Inspection Panel, found “human rights implicitly embedded in various policies of the Bank.”69 Mr. Ayensu called upon the Bank to be “more forthcoming about articulating its role in promoting rights within the countries in which it operates” and encouraged the Bank to study “the wider ramifications of human rights violations as these relate to the overall success or failure of policy compliance in Bank-financed projects.”70 In its report on this project, the Panel stated that human rights considerations are relevant when they “impede the implementation of the Project in a manner compatible with the Bank’s policies.”71 Since leakage clearly undermines the purpose of health sector loans and is incompatible with the Bank’s policies, it represents a situation where human rights implications should be taken into account.

More recently, in investigating the Honduras Land Administration Project (2007), the Panel found that the Bank Policy on Project Appraisal, Operational Manual Statement (OMS) 2.20, required Bank-financed projects to respect international agreements on human rights and indigenous peoples when the project country is a signatory.72 This was “the first time that the Panel explicitly addressed the merits of a claim based on international human rights law.” In the request for inspection, the indigenous Garifuna community claimed that the project violated the Government of Honduras’s obligations under the International Labour Organization (ILO) Convention No. 169. While the Panel stressed that its mandate was limited to questions of internal compliance, it still found the ILO Convention provisions applicable to the Bank through the OMS 2.20, which requires the Bank to ensure that financial activities are consistent with a borrower’s international agreements regarding the environment, as well as the health and well-being of its citizens. The Panel expressed “serious concern” with the General Counsel’s argument that OMS 2.20 should only apply to international obligations that are “essentially of an environmental nature.”

These decisions by the Panel, which confirm that the Bank must engage with human rights concerns in certain situations, should be cited and utilized to support requests to investigate leakage in health sector loans. Such decisions, coupled with the Bank’s nascent recognition of the relevance of human rights norms to its work, provide relevant context for interpreting the Bank’s policies and add a deeper dimension to the analysis of these claims.

Challenges in seeking accountability for leakage of health sector funds through the inspection panel  
While the Inspection Panel has helped ground the dialogue about the relevance of human rights norms to the World Bank’s work by applying these concepts in concrete situations, it also has its own limitations that hamper its use as a true accountability mechanism. First, the role of the Requesters is quite limited, as they do not have an opportunity to rebut Management’s response or propose alternative recommendations for action. Although the Panel clarified in 1999 that Management must consult with the requester and other affected people in preparing an action plan, the Panel acknowledges that such consultation often does not occur.73 To make matters worse, even where Management engages in consultation with the Requesters, only Management has access to the Panel’s report during this process, which, as the Panel recognizes, creates an “unfortunate structural asymmetry.”74 The Panel is currently seeking ways of increasing participation by the Requesters and affected people in the overall process.75

Second, the standard for establishing a violation is quite high. The Panel will address “only those material adverse effects, alleged in the request, that have totally or partially resulted from serious Bank failure of compliance with its policies and procedures,” thereby requiring a causal link between the Bank’s noncompliance and the material harm.76 This link may be difficult to establish where the issue is leakage from an entire health sector. While the Bank clearly violates its own internal policies by failing to perform its monitoring, evaluation, and supervision responsibilities, it may be difficult to show that those violations directly resulted in material harm to specific individuals.

Third, the Panel will not accept a request for inspection if at least 95% of the loan financing has already been disbursed.77 Given that PETS take a long time to complete and require the collection of a significant amount of data before reaching any meaningful quantitative results, the results of these surveys may not be available until long after disbursement is complete. Under the Bank’s new disclosure policy, however, other forms of information should be made available that may assist in identifying leakage at an earlier stage.

Moreover, even if the Panel finds that the Bank has violated its own policies, it does not have the power to make the final decision about what actions the Bank should take, if any, to remedy the situation. The Bank’s Board of Directors, not the Panel, adopts the final action plans, which often do not respond adequately to the Panel’s findings.78 In addition, the Panel lacks powers of enforcement, restitution, and oversight over the implementation of remedial measures.79 The Panel is, however, taking steps to address this concern. For example, the Panel has made it a practice to return to affected areas in order to give briefings on the results of the investigation, and on a few occasions, has played a more substantive role in post-investigation activities.80

Thus, while the Panel certainly has its limitations as an accountability mechanism, it has been finding new ways to address some of these shortcomings. The Panel is a dynamic entity that is being shaped and molded by the claims it is asked to address. As more and more challenging cases are being brought before the Panel, including cases that raise serious human rights concerns, the Panel will be forced to explore the outer edges of its mandate and perhaps even push those boundaries further.


The international community can no longer afford to ignore the massive leakage of donor funds from critical social sectors, such as health, particularly when this leakage occurs in the poorest places on the planet. Instead of blaming only the recipient governments for such leakage, solutions should focus on holding both donors and borrowers accountable for the use of funds. Although the actions of IFIs, such as the World Bank, helped create the impoverished situations that plague sub-Saharan Africa today through devastating structural adjustment programs, these IFIs have largely escaped any true call to accountability.

Holding donors accountable is a legally challenging task, but it is not impossible. The World Bank’s Inspection Panel is one potential venue for holding the Bank accountable for leakage of funds from the health sector. The problem of leakage should be framed both in terms of the Bank’s violations of its own policies and in terms of closely related violations of basic human rights obligations. Organizations concerned with the leakage of health sector funds should therefore consider filing a request for inspection with the Panel, especially in the absence of other available channels to bring complaints against the Bank.

Concerns with aid effectiveness and accountability are especially critical at present due to an aid climate that is shifting from project-specific loans to general budget support. Many poverty reduction strategies, which cut across multiple sectors, including health, are now funded through general budget support, whereby the government simply disburses the money through its own financial management system. While general budget support helps simplify and unify procurement, disbursement, and management procedures, it may also amplify the risk of leakage by making it harder for donors to trace how the funds are being used.81 Pledges of greater funding for the health sector may have little meaning if the Bank does not take the necessary steps to ensure that the funds actually reach the intended beneficiaries.

Fatma E. Marouf, JD, is Associate Professor of Law at the William S. Boyd School of Law, University of Nevada Las Vegas, where she teaches international human rights and immigration law. She received her BA from Yale University and her JD from Harvard Law School.

Please address correspondence to the author at William S. Boyd School of Law, 4505 Maryland Parkway, Box 451003, Las Vegas, NV 89154, email: fatma.marouf@unlv.edu.


1.    E. Suzuki and S. Nanwani, “Responsibility of international organizations: The accountability mechanisms of multilateral development banks,” Michigan Journal of International Law 27/1 (2005), pp. 177–225.

2.    R. Reinikka and J. Svensson, “Survey techniques to measure and explain corruption,” World Bank Development Research Group, Policy Research Working Paper 3071 (Washington, DC: World Bank, 2003), p. 1. Available at http://siteresources.worldbank.org/INTPEAM/Resources/PETS2.pdf.

3.    Ibid.

4.    During the past decade PETS and QSDS have been implemented in approximately two dozen developing countries. See B. Gauthier, PETS-QSDS in sub-Saharan Africa: A stocktaking study, (Washington, DC: World Bank, 2006), p. 1. Available at http://siteresources.worldbank.org/INTPRS1/Resources/383606-1212502456058/061108_PETS-QSDS_Gauthier.pdf.

5.    R. Reinikka and N. Smith, Public expenditure tracking surveys in education (Paris: UNESCO International Institute for Educational Planning, 2004), p. 33. Available at http://www.unesco.org/iiep/PDF/pubs/Reinikka.pdf.

6.    Gauthier (see note 4), p. 27. Strictly defined, leakage is the share not received by a facility with respect to the resources intended for the facility, whereas the more narrow definition of resource measures the share of resources received by a facility with respect to the resources disbursed to that facility by a higher level.

7.    “Using PETS to fight corruption” (Bergen, Norway: U4 Anti-Corruption Resource Centre, n.d.). Available at http://www.u4.no/themes/pets/petstool.cfm.

8.    K. Kaiser, I. Kishnarova, and M. Lindelow, Measuring corruption in the health sector: What we can learn from public expenditure tracking and service delivery surveys in developing countries (Washington DC: World Bank, 2005); see also Gauthier (see note 4); M. Lindelow, R. Reinikka and J. Svensson, Health care on the frontlines: Survey evidence on public and private providers in Uganda, Africa Region Human Development Working Paper 27212 (Washington, DC: World Bank, 2003). Available at http://www-wds.worldbank.org/external/default/WDSContentServer/IW3P/IB/2003/11/11/000090341_20031111113920/Rendered/PDF/272120PAPER0AFRHD0wp0no1038.pdf; M. Lewis, “Governance and corruption in public health systems,” Working Paper 78 (Washington, DC: Center for Global Development, 2006), pp. 21, 30.

9.    G. Sundet, “Public expenditure tracking surveys: Lessons from Tanzania,” U4 Brief 14 (Bergen, Norway: U4 Anti-Corruption Resource Centre, 2007). Available at http://www.cmi.no/publications/file/2812-public-expenditure-tracking-surveys.pdf.

10.  X.Ye and S. Canagarajah, Efficiency of public expenditure distribution and beyond: A report on Ghana’s 2000 public expenditure tracking survey in the sectors of primary health and education, Africa Region Working Papers 31(Washington, DC: World Bank, 2002), pp. 5, 24. Available at http://www.worldbank.org/afr/wps/wp31.pdf.

11.  Ibid., p. 25, note 12.

12.  Gauthier (see note 4).

13.  Ibid., p. 39.

14.  Republic of Kenya, Public expenditure tracking survey (PETS) 2004, Preliminary Report (Nairobi, Kenya: Ministry of Planning and National Development, Ministry of Finance, Ministry of Health, Ministry of Education Science and Technology, 2004).

15.  O. Picazo, S. Kagulura, and the PET/QSDS Team of the Ministry of Health and the University of Zambia, “State of human resources for health in Zambia: Findings from the public expenditure tracking and quality of service delivery survey, 2005/06” (presentation at the Human Resources for Health Research Conference, Mulungushi International Conference Center, Lusaka, Zambia, June 7–8, 2007). Available at http://www.hrhresourcecenter.org/hosted_docs/State_HRH_Zambia_PET_QSDS.pdf; see also O. Picazo and F. Zhao, Zambia health sector public expenditure review: Accounting for resources to improve effective service coverage (Washington, DC: World Bank Publications, 2008).

16.  Gauthier (see note 4), pp. 27–29.

17.  IEG-World Bank, Public sector reform: What works and why? An IEG evaluation of World Bank support (Washington, DC: World Bank, 2008), p. 50. Available at http://siteresources.worldbank.org/EXTPUBSECREF/Resources/psr_eval.pdf.

18.  Gauthier (see note 4), p. 32; R. Reinikka and J. Svensson, “Using micro-surveys to measure and explain corruption,” World Development 34/2 (2006), pp. 359–370

19.  For example, the World Bank’s spending on health programs increased from US$6.8 billion in 1997 to US$16 billion in 2006 without much improvement in health outcomes. IEG-World Bank, Achieving sustainable development: Annual review of development effectiveness 2009 (Washington, DC: World Bank, 2009); see also L. Elliot, “World bank admits most health aid fails,” The Guardian (May 1, 2009). Available at http://www.guardian.co.uk/business/2009/may/01/world-bank-health-aid-poverty.

20.  IEG-World Bank (see note 19); Elliot (see note 19).

21.  IEG-World Bank, Review of IDA internal controls: An evaluation of management’s assessment and the IAD review (Washington, DC: World Bank, 2009), p. 40. Available at http://siteresources.worldbank.org/EXTOED/Resources/ida_controls_full.pdf.

22.  Global Policy Forum, “World bank health work flawed still pushing privatization of services” (July 10, 2009). Available at http://www.globalpolicy.org/social-and-economic-policy/the-three-sisters-and-other-institutions/the-world-bank/47923.html.

23.  IEG-World Bank (see note 17), pp. 12–17.

24.  Ibid., p. 45.

25.  Ibid., p. 60.

26.  Ibid., pp. 62–63.

27.  Ibid., p. 62.

28.  Ibid., p. 36.

29.  Ibid., p. 60.

30.  Ibid., p. 73; Ibid., p. 53.

31.  L. H. Piron, “The right to development: Study on existing bilateral and multilateral programmes and policies for development partnership,” Submission to the Commission on Human Rights, E/CN.4/Sub.2/2004/15 (2004).

32.  IEG-World Bank (see note 17), p. 62.

33.  Ibid., p. 63. Other examples of state-capture include corrupt awards of big contracts, kickbacks from big international corporations, and privatization to insiders at bargain prices.

34.  Ibid.

35.  IEG-World Bank (see note 21), p. xviii.

36.  Ibid.

37.  For example, the Maastricht Guidelines state that “[i]t is crucial for the elimination of violations of economic, social and cultural rights for international organizations, including international financial institutions, to correct their policies and practices so that they do not result in deprivation of economic, social and cultural rights.” Maastricht Guidelines on Violations of Economic, Social and Cultural Rights, Maastricht, January 22–26, 1997, at para. 19 (emphasis added). Available at http://www.uu.nl/uupublish/content/20-01.pdf. See also Committee on Economic, Social and Cultural Rights, General Comment No. 14, The Right to the Highest Attainable Standard of Health, UN Doc. No. E/C.12/2000/4 (2000), para. 64 (stating that “international financial institutions, notably the World Bank and International Monetary Fund, should pay greater attention to the protection of the right to health in their lending policies, credit agreements and structural adjustment program”). Available at http://www.unhchr.ch/tbs/doc.nsf/0/40d009901358b0e2c1256915005090be?Opendocument. Various scholars have also forcefully argued that the World Bank has at least some human rights obligations. See, for example, S. I. Skogly, The human rights obligations of the World Bank and the International Monetary Fund (London: Cavendish, 2001).

38.  World Bank, International Bank for Reconstruction and Development: Articles of agreement (Washington D.C: World Bank, 1944), Section 3, Article VII(3), “Position of the Bank With Regard to Judicial Process,” and World Bank, International Development Association: Articles of agreement (Washington DC: World Bank: 1960), Article VII (3), “Position of the Bank With Regard to Judicial Process”; see also Mendaro v. World Bank, 717 F.2d 610, 617-18 (D.C. Cir. 1983) (interpreting the Bank’s Articles of Agreement as waiving immunity from actions arising out of the Bank’s external relations, but preserving immunity as to “suits arising out of the Bank’s internal operations, such as its relationships with its own employees,” since the latter do not relate to the Bank’s ability to fulfill its purpose).

39.  See, for example, R. Hammonds and G. Ooms, “World bank policies and the obligations of its members to respect, protect and fulfill the right to health,” Health and Human Rights: An International Journal 8/1 (2004), pp. 26–60 (arguing that the World Bank should be held liable through its Member States).

40.  See, for example, ILC 55th Sess. at para. 5 (2004), UN Doc. A/CN.4/541 (noting that while the ARSIWA “are not immediately relevant to international organizations, they have to be fully taken into account when discussing issues relating to the attribution to international organizations that are parallel to those concerning States”).

41.  In Behrami v. France and Others, 22 BHRC 477 at para. 133, (2007) Appl. No. 78166/01 & 71412/01, the European Court for Human Rights held that the detention of an individual by French forces, acting as part of the international force in Kosovo pursuant to a mandate from the United Nations Security Council, was attributable to the United Nations rather than to France, but this decision has been criticized for applying the wrong legal test and being driven primarily by policy concerns. See, for example, A. Sari, “Jurisdiction and international responsibility in peace support operations: The Behrami and Saramit cases,” Human Rights Law Review 8/1 (2008), 169–70. In Al-Jedda v. Secretary of State for Defence [2007] UKHL 58, the House of Lords applied the correct standard in assessing whether the actions of British troops operating as part of a multi-national force in Iraq were attributable to the United Kingdom or to the United Nations. The Lords properly framed the legal issue in terms of whether “the UN exercise effective control over the conduct of UK forces” (Ibid., para. 22, emphasis added), concluding that the conduct was attributable to the United Kingdom since the coalition forces were not dispatched by the UN and had no UN mandate. Available at http://www.publications.parliament.uk/pa/ld200708/ldjudgmt/jd071212/jedda.pdf.

42.  See ILC, 61st Sess, para. 25 (2009), UN Doc. A/CN.4/610.

43.  International Bank for Reconstruction and Development, International Development Association/“The World Bank Inspection Panel,” Resolution No. IDA 93–6 (Sept. 22, 1993), para. 12. Available at http://siteresources.worldbank.org/EXTINSPECTIONPANEL/Resources/ResolutionMarch2005.pdf.

44.  World Bank, “The Inspection Panel: Cases — by country.” Available at http://web.worldbank.org/WBSITE/EXTERNAL/EXTINSPECTIONPANEL/0,,contentMDK:21680496~menuPK:64129250~pagePK:64129751~piPK:64128378~theSitePK:380794,00.html.

45.  World Bank, International Development Association Articles of Agreement, Article III, Section 5(b). Available at http://siteresources.worldbank.org/IDA/Resources/ida-articlesofagreement.pdf.

46.  World Bank, World Bank Operations Manual, OP 13.05 — Project Supervision (Washington, DC: World Bank, 2001) and World Bank, World Bank Operational Manual, OP 10.02 — Financial Management (Washington, DC: World Bank, 1997). The Bank’s Operational Policy on Monitoring and Evaluation (OP 13.60) would also be relevant in requesting an inspection based on leakage of health sector funds, as it requires the Bank to evaluate the “relevance, efficacy and efficiency” of its loans. An online index to all World Bank Operations Policies is available at http://web.worldbank.org/WBSITE/EXTERNAL/PROJECTS/EXTPOLICIES/EXTOPMANUAL/0,,menuPK:4564185~pagePK:64719906~piPK:64710996~theSitePK:502184,00.html.

47.  Republic of Kenya (see note 14).

48.  Gauthier (see note 4), p. 37.

49.  For the World Bank’s new access to information policy, see Operations Policy and Country Services, Toward greater transparency through access to information: The World Bank’s disclosure policy (Washington, DC: World Bank, 2009. Available at http://siteresources.worldbank.org/EXTINFODISCLOSURE/Resources/R2009-0259-2.pdf?&resourceurlname=R2009-0259-2.pdf.

50.  N. Roht-Arriaza, “Special problems of a duty to prosecute: Derogation, amnesties, statutes of limitation and superior orders,” in N. Roht-Arriaza (ed), Impunity and human rights in international law and practice 29 (1995), p. 29.

51.  Ibid.

52.  Rodríguez v. Uruguay, Comm. No. 322/1988, 9.8.1994, CCPR/C/51/D/322/1988 (1994), para. 12.3.

53.  Bautista de Arellana v. Colombia, Comm. No. 563/1993, 10/27/1995, CCPR/C/55/D/563/1993 (1995), para. 8.6. Likewise, in a murder case, the European Commission of Human Rights (“European Commission”) found that Article 2, protecting the right to life, created positive obligations to investigate and prosecute the offenders. Roht-Arriaza (see note 50), p. 32. Such cases also reinforce the right to a remedy set forth in various human rights treaties. See, for example, Universal Declaration of Human Rights (UDHR), G.A. Res. 217A (III) (1948), Art. 9. Available at http://www.un.org/Overview/rights.html.See also International Covenant on Civil and Political Rights (ICCPR), G.A. Res. 2200A (XXI) (1966), Art. 2(3). Available at http://www2.ohchr.org/english/law/ccpr.htm.

54.  Association X v. United Kingdom, (7154/75) DR 14, 31.

55.  See, for example, Savage v. South Essex Partnership NHS Foundation Trust [2008] UKHL 74 (December 10, 2008) (holding that, pursuant to the right to life, health authorities have an overarching obligation to protect the lives of patients in their hospitals, including a duty to ensure that staff are highly trained, professional and competent, and that the policies, procedures and systems in place at the hospital adequately safeguard life).

56.  See, for example, I. Byrne, “Enforcing the right to health: Innovative lessons from domestic courts,” in A. Clapham and M. Robinson (eds), Strategies for realizing the right to health, Swiss Human Rights Book 3 (Zurich: Rüffer & Rub, 2009), pp. 525–557. Available at http://www.swisshumanrightsbook.com/SHRB/shrb_03_files/37_453_Byrne.pdf. Byrne notes (p. 525) that “courts have attempted to give practical meaning to notions of indivisibility and interdependence of rights recognizing that the right to health has clear links to many other rights, not just economic and social but also civil and political rights — e.g. the right to life and, the right not to be subjected to torture or cruel, inhuman or degrading treatment, and the right to information.”

57.  The UN General Assembly has long recognized that “[f]reedom of Information is a fundamental human right . . . and the touchstone of all the freedoms to which the United Nations is consecrated.” Calling of an International Conference on Freedom of Information, G.A. Res. 59(I) (1946); see also ICCPR (see note 53), Art. 19: “[e]veryone shall have the right to . . . seek, receive and impart information and ideas of all kinds”). See also the African Charter on Human and Peoples’ Rights, OAU Doc. No. CAB/LEG/67/3 rev. 5 (1981), Art. 9 (“[e]very individual shall have the right to receive information”), rreprinted in International Legal Materials 21 (1982), p. 58. Available at http://www1.umn.edu/humanrts/instree/z1afchar.htm.

58.  Global Transparency Initiative, “Comments on the World Bank’s approach paper ‘Toward greater transparency: Rethinking the World Bank’s disclosure policy” (Washington, DC: Global Transparency Initiative, 2009). Available at http://www.ifitransparency.org/uploads/7f12423bd48c10f788a1abf37ccfae2b/GTI_Comments_on_WB_Disclosure_Approach_Paper_final.pdf.

59.  IEG-World Bank (see note 17), p. 75.

60.  D. Kaufman, The World Bank, “Rethinking governance: Empirical lessons challenge orthodoxy,” unpublished discussion draft (March 11, 2003). Available at

61.  IEG-World Bank (see note 17), p. 64.

62.  Ibid., p. 63. One example is Ghana, where “reforms … to strengthen good governance and social accountability have to an important extent been demand driven from civil society. Ibid., p. 63.

63.  Uganda provides a classic example of how participation, combined with access to information, can effectively stop the leakage of funds. Uganda’s 1996 PETS revealed that between 1991 and 1995, 87% of the annual capitation grant (per student) never reached the schools. The Ugandan government responded to these grim statistics by publicizing transfers of funds from the central government to the districts in newspapers and on the radio, requiring primary schools and district offices to post transfer information for everyone to see, and training school committees about how to use the information to hold the government accountable. By 2001, schools received 82% of non-wage funds. The information campaign explained two-thirds of this huge drop in leakage. Thus, the extent to which funding reached the intended beneficiaries depended heavily on schools’ opportunity to voice their claims for the funds. See Reinikka and Svensson (see note 2), pp. 3, 6, and 13. Available at http://wwwwds.worldbank.org/external/default/WDSContentServer/IW3P/IB/2003/07/08/000094946_03062104301451/Rendered/PDF/multi0page.pdf. By comparison, Tanzania’s experience shows that information alone, without active participation by civil society, does not produce the same results. See G. Sundet, “Public expenditure and service delivery monitoring in Tanzania: Some international best practices and a discussion of present and planned Tanzanian initiatives” (Washington, DC: USAID, March 2004), p. 4. Available at http://www.worldbank.org/socialaccountability_sourcebook/Regional%20database/Case%20studies/Africa/Tanzania%20-%20PETS.pdf.

64.  World Bank (see note 45), Article IV, Section 10; see also I. Shihata, “Prohibition of political activities in the bank’s work” Legal opinion of the General Counsel (July 11, 1995), later published and expanded in I. F. I. Shihata, The World Bank legal papers (The Hague, Netherlands: Martinus Nijhoff Publishers, 2000), pp. 219–244, with further elaboration in I. Shihata, The World Bank in a changing world: Selected essays and lectures, vol. 3 (The Hague, Netherlands: Martinus Nijhoff Publishers, 2000), pp. 156–187.

65.  R. Danino, Senior Vice-President and General Counsel, The World Bank, Legal Opinion on Human Rights and the Work of the World Bank (January 27, 2006). Available at http://www.ifiwatchnet.org/sites/ifiwatchnet.org/files/DaninoLegalOpinion0106.pdf.

66. A. Palacio, “The way forward: Human rights and the world bank: Special report,” Development Outreach (October 2006), pp. 35–37. Available at http://siteresources.worldbank.org/EXTSITETOOLS/Resources/PalacioDevtOutreach.pdf.

67.  Ibid.

68.  S. Herz and A. Perrault, “Bringing human rights claims to the World Bank inspection panel” (Washington, DC: Center for International and Environmental Law / Bank Information Center and San Francisco: International Accountability Project, 2009). Available at www.bicusa.org/admin/Document.101841.aspx.

69.  World Bank Inspection Panel (see note 43), p. 9; E. Ayensu, “Remarks of the Chairman of the Inspection Panel to the Board of Executive Directors on the Chad-Cameroon Pipeline projects” (Washington, DC: World Bank Inspection Panel, September 12, 2002). Available at http://web.worldbank.org/WBSITE/EXTERNAL/EXTINSPECTIONPANEL/0,,contentMDK:20227264~menuPK:64129469~pagePK:64129751~piP:64128378~theSitePK:380794,00.html.

70.  World Bank Inspection Panel (see note 43), p. 74.

71.  World Bank Inspection Panel, Investigation report: Chad-Cameroon petroleum and pipeline project (Loan No. 4558–CD); Petroleum sector management capacity building project (Credit No. 3373–CD); and Management of the petroleum economy (Credit No. 3316–CD) (Washington, DC: World Bank, 2002), para. 215 on pp. 62–63. Available at http://siteresources.worldbank.org/EXTINSPECTIONPANEL/Resources/ChadInvestigationReporFinal.pdf.

72.  World Bank Inspection Panel (see note 43), pp. 74–75.

73. Ibid., p. 56.

74.  Ibid.

75.  Ibid, p. 57.

76.  World Bank Board of Directors, BP 17.55, Annex C: Conclusions of the board’s second review of the inspection panel (“1999 Clarifications”) (Washington, DC: World Bank, (April 20, 1999), para. 13, emphasis added. Available at http://web.worldbank.org/WBSITE/EXTERNAL/PROJECTS/EXTPOLICIES/EXTOPMANUAL/0,,contentMDK:20065440~menuPK:64701771~pagePK:64709096~piPK:64709108~theSitePK:502184,00.html. Also cited in Suzuki and Nanwani (see note 1), pp. 213–214.

77.  World Bank Inspection Panel (see note 43), para. 14(c).

78.  Herz and Perrault (see note 68), p. 2.

79.  E. Suzuki and S. Nanwani (see note 1), p. 219.

80.  World Bank Inspection Panel (see note 43), pp. 57–59.

81.  But see Transparency International, “Poverty, aid and corruption,” Policy Paper No. 1 (2007), p. 4. Available at http://www.reliefweb.int/rw/lib.nsf/db900sid/ASAZ-7HEDBS/$file/TI_Apr2008.pdf?openelement (asserting that general budget support can curb corruption by aligning aid with national development plans and helping ensure that aid priorities are actually based on the needs of the poor); See also C. Dom, “The Joint Evaluation of General Budget Support 1994–2004: Thematic Briefing Paper 1: What are the effects of General Budget Support?” (Glasgow: Department for International Development (DFID), and Paris: Organisation for Economic Co-operation and Development (OECD), 2007). Available at http://www.oecd.org/dataoecd/14/35/38339258.pdf; this document reported that general budget support can increase aid effectiveness by making public expenditure more efficient.