The Role of the Fund in Governance Issues—Review of the Guidance Note—Preliminary Considerations
Author: International Monetary Fund
Publication Date:
08 Feb 2017
eISBN:
9781498346528
Language:
English
Keywords:
PP; Fund engagement; director; IMF mission chief; Policy advice; guidance; staff report; time horizon; fund resource; fund staff; Corruption
Download PDF (2.3 MB)

The Role of the Fund in Governance Issues - Review of the Guidance Note - Preliminary Considerations

Abstract

The Role of the Fund in Governance Issues - Review of the Guidance Note - Preliminary Considerations

Background

1. In its 2016 and 2017 Communiqués, the International Monetary and Financial Committee (IMFC) called on the Fund to review the July 1997 Guidance Note (Guidance Note) on the Role of the Fund in Governance Issues.1 In doing so, the IMFC emphasized the need for global action to tackle the sources and channels of corruption and welcomed proposed work by the Fund on corruption.2 The IMFC’s focus on corruption reflects a growing consensus within the international community regarding the adverse impact that systemic corruption can have on inclusive, sustainable growth. It also takes into account a growing willingness among member authorities to discuss this problem openly. Consistent with the IMFC’s focus, this paper reviews the Fund’s experience with implementation of the Guidance Note, with a particular focus on how the Fund has approached the problem of corruption. In addition to providing an overview of Fund initiatives of specific relevance, this paper distills the findings of an in-depth review of how corruption was addressed—in both the surveillance and the UFR context—in a representative sample of countries over an extended period of time. This paper does not make any specific proposals to amend or clarify the Guidance Note, but rather raises issues for discussion by the Executive Board. Drawing on that discussion, a follow-up paper could identify what modifications, if any, should be introduced with respect to the Fund’s policies and practices in this area.

2. At the outset, it is helpful to have a broad understanding of the concepts of “governance,” “good governance,” and “corruption.” Although the Guidance Note refers to all three, it does not distinguish among them. As is elaborated in the Supplementary Paper, a considerable amount of work has been done outside the Fund on how these terms should be defined. Drawing on this work—and on the Fund’s own application of the concepts—the following observations can be made:3

  • “Governance,” the broadest term, refers to the institutions, mechanisms, and practices through which governmental power is exercised in a country, including for the management of public resources and regulation of the economy. This includes processes at the country level, including institution-level structural arrangements. It is an inherently neutral term, describing the framework for exercising authority, not characterizing its results.

  • “Good governance” is a more normative concept, which recognizes that the quality of governance can impact its effectiveness and efficiency in achieving desired outcomes. Although the Guidance Note encompasses governance and, in particular, “good governance,” it would not be feasible in a single review to examine the Fund’s work in these broader areas. The reason is that the Fund’s work to promote good governance has grown tremendously, as discussed below. Indeed, promoting good economic governance is central to the Fund’s everyday work in surveillance, use of Fund resources, and capacity development. Moreover, work under each of these areas of engagement is assessed through dedicated policy reviews. Accordingly, this stocktaking does not attempt to provide additional perspectives on the Fund’s engagement at the broad levels of governance and good governance. Instead, it focuses on that part of the Fund’s engagement that relates to analyzing and providing advice on corruption issues.

  • Regarding “corruption,” a definition that has been generally accepted is “the abuse of public office for private gain.” This definition has been adopted by a variety of organizations and is consistent with the provisions of the United Nations Convention Against Corruption (UNCAC).4 The definition focuses on abuse by public actors, meaning that fraudulent acts perpetrated exclusively by private citizens are not covered.5 It is recognized, of course, that corruption is often facilitated—and sometimes initiated—by private actors (for example, the offering of a bribe) and, therefore, that any meaningful anti-corruption strategy needs to address behavior within the private sector. It should be emphasized that an act can be corrupt even if it does not result in direct financial gain; an official also engages in a corrupt act if, as a result of political interference, he or she abuses public office. Finally, although corruption is often associated with the misapplication of the law, it can also pervert the law-making process itself, i.e., where the decisions of legislators are motivated exclusively by private interests, often as result of the influence of powerful business networks (“state capture”).

3. The foregoing discussion illustrates that governance and good governance are broader concepts than corruption. It is possible for a country to have poor governance even in the absence of significant corruption, for example, because of ineffective, inefficient, or inequitable policies and institutions. The presence of corruption, however, generally indicates shortcomings in good governance. The premise of the ongoing review—and indeed the IMFC’s guidance—is that systemic corruption has a particularly pernicious effect on economic performance.6 It should be acknowledged, however, that any effective strategy for addressing corruption problems must be more comprehensive than simply anti-corruption measures to have a lasting impact; durable solutions will encompass governance improvements more broadly, including with regard to transparency, accountability, appropriately balanced regulation, and effective institutions. In sum, promoting good governance is the most durable way of addressing systemic corruption.

4. The content of the rest of this stocktaking paper is organized as follows: The macroeconomic impact of corruption is discussed next, followed by a high-level overview of the various initiatives that the Fund has taken to address broader governance issues, including corruption, and a discussion of the standards that govern Fund engagement in these areas. After this, the paper outlines the methodology employed for an in-depth qualitative review of operational experience in addressing corruption in the context of the Guidance Note under both Fund surveillance and use of Fund resources (including program conditionality). This methodology informs two subsequent sections that discuss the detailed findings of the review and lessons learned. The paper ends with Issues for Discussion.

The Macroeconomic Impact of Corruption

5. There is a growing understanding that corruption impacts economic performance by weakening the state’s capacity to perform its core functions and affecting drivers of potential and inclusive growth. As a general matter, the impact of corruption on the state’s ability to carry out its functions increases as corruption becomes more systemic and acute. While all countries experience some level of corruption, in some countries, corruption is so pervasive that it has become the norm, with significant and long-lasting social and macroeconomic impact. However, even where corruption is not systemic, it can still undercut macroeconomic performance. The adverse impact of corruption on economic growth has been generally established in the literature and the Fund’s own experience.7 This logic underlies the Guidance Note. More recent analysis finds that the negative relationship becomes more pronounced when the dependent variable is not simply aggregate or per-capita growth, but is specified as inclusive growth.8

6. The key drivers of potential and inclusive growth include macro-financial stability, public and private investment, human capital accumulation, and total factor productivity (Figure 1).9 The main channels through which corruption impacts the economy reflect key state functions, such as the execution of budgetary policy and the delivery of public services, market regulation, financial sector oversight, and public order and enforcement. Although the precise pattern of transmission varies across countries, the more corruption interferes with these functions, the more it distorts policies and their implementation. Depending on its pervasiveness, corruption can affect some or all of the drivers. Low rates of inclusive growth can also lead to increased incidence of corruption, creating a negative feedback loop that can be long lasting.

Figure 1.
Figure 1.

Corruption-Growth Nexus

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Source: IMF, 2016b.

7. For the Fund, a key channel for corruption is the impact on fiscal performance. Corruption affects core government functions, which can weaken the state’s capacity to tax, leading to lower revenue collection. Widespread corruption harms the culture of compliance, thereby increasing tax evasion.10 Corruption also creates disincentives for taxpayers to pay taxes. When tax exemptions are perceived to be the product of a bribe, the public becomes far less willing to comply with the tax laws, which are perceived as unfair. Tax evasion contributes to perceptions of unfairness, which undermine citizens’ trust in their government. This distrust can potentially discourage entrepreneurs from starting new businesses in the formal economy, further eroding the revenue base.11 High corruption can also reduce the state’s incentive to adopt measures to mobilize additional domestic revenues, particularly through more efficient taxes, such as the VAT.12 In resource-rich countries suffering from corruption, shortfalls in revenues can also result from direct embezzlement and theft. Bribes and kickbacks in public procurement can also inflate costs and distort budget allocation, both in day-to-day operations and especially for large public investment projects.13 Inflated procurement costs have been an issue in advanced economies as well. These shortcomings limit the scope for institution building to support economic activity and negatively impact service delivery in core public sectors, the brunt of which is borne by the poorer members of society.14 The combination of lower revenue and wasteful spending can result in large fiscal deficits and ultimately substantial debt accumulation, at the same time that high levels of perceived corruption can raise a country’s borrowing costs. The net effect of the above can result in a significant adverse macroeconomic impact.

Figure 2.
Figure 2.

Corruption and Tax Revenues 1/

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Sources: Transparency International; and IMF staff estimates.1/ Variables presented are deviations from values predicted by income level.2/ Larger numbers indicate lower perceived levels of public sector corruption.Samples include commodity importers only.Asterisks from the correlation equation imply significance at 1 percent.
Figure 3.
Figure 3.

Public Investment Efficiency and Corruption Perception Index

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Source: IMF staff estimatesNote: The samples are grouped into high, medium and low corruption in equal numbers based on the Corruption Perception Index. The lines show the range of efficiency for each group. The orange and gray boxes are the second and third quartile (from the bottom) of the samples, thus they cover 50 percent of the countries.

8. High levels of corruption significantly impede the state’s ability to carry out other basic functions, which can be no less detrimental in macroeconomic terms. For example, as the revenue base erodes, the government could become more reliant on seigniorage financing or financial repression, undermining sound monetary policy. In the financial sector, corrupt lending practices, weak banking supervision, and regulatory forbearance can threaten the stability of the financial system. Any of these practices may be caused by corruption, which, in turn, may lead to a rising share of nonperforming assets in the banking sector and a diminished ability to intermediate credit. In addition, in several countries, corruption and related fraud in the banking sector has imposed significant fiscal costs for bank recapitalization. At the same time, corruption weakens the ability of the state to address policy distortions that hinder financial development and inclusion. By weakening the enforcement of claims and the recovery of debts, corruption undermines the payment culture, increases market distrust, reduces access to credit, and increases transaction costs. Corruption tends to be associated with excessive and unpredictable public economic intervention, undermining the business climate. Further, corruption can affect the implementation of laws as well as inappropriately influence the processes by which legislation is formed and adjudicated (i.e., through state or regulatory capture); this is the case when private interests supplant the public interest.15

9. More generally, corruption in regulatory functions reduces investment and provides opportunities for rent-seeking, which stifles productivity and growth. As noted earlier, corruption can distort selection of public investment projects, leading to poor outcomes. For example, public investment on education and health is lower in both quantity and quality in corrupt systems because spending allocations are made partly to allow corrupt officials to generate “commissions.” Where resources appropriated for education are stolen, students bear the brunt. In addition, young people may have weaker incentives to invest in education and skills if career success depends largely on who you know, rather than on personal merit. In addition, systemic corruption can promote emigration of skilled persons. These factors suggest a negative relationship between corruption and human capital formation. High levels of corruption discourage private investment and especially foreign direct investment, a strong promoter of growth in recipient countries. More important, corruption can distort resource allocation from productive to rent-seeking activities, leading managers to focus less on increasing firms’ productivity. It can also give advantage to incumbent firms by acting as a barrier to entry, and can negatively affect the ability of firms to enter the marketplace and compete effectively. Corruption can also become systemic and self-sustaining where those benefiting from corruption in turn exert influence to maintain the status quo, protecting rent-seeking opportunities.

10. Corruption is often linked to poor social outcomes. Distortion in the allocation of funding within health and education systems can lead to increased child mortality rates and school dropout rates. Women are more likely to be affected when bribes take the form of sexual activity or when corruption allows illegal practices, such as human trafficking and prostitution.

11. Beyond the short term, poor governance can have more insidious impacts through the erosion of confidence and trust in the state. Economies seriously affected by corruption can sometimes achieve relatively robust growth for sustained periods if corruption remains within limits generally tolerated by businesses and households and economic conditions are otherwise supportive. However, over time, corruption and other forms of poor governance can have a corrosive impact on trust, social capital, and confidence in governance institutions and policies—particularly where there is widespread petty corruption involving demands for bribes from public officials—making the situation precarious. A tipping point may be reached—often when governing institutions are unable to effectively manage an adverse economic shock. Frustration with corruption can then boil over, leading to rejection of governments, institutions, and policies, a process that can be disruptive for growth and living standards. At the extreme, this can result in disorder, civil strife, and conflict, with long-lasting economic and social costs.

12. However, it should also be acknowledged that corruption and its impacts are difficult to quantify. In general, corruption’s inherently concealed nature creates enormous and well-researched challenges in its identification, tracking, and measurement.16 Difficulties include varied understandings of the term “corruption,” differences between perceived and actual levels of corruption, and problems associated with comparing and consolidating different sources of information on corruption.17 In addition, econometric approaches help to identify the “typical” macroeconomic impact of corruption, but this influence may vary depending on the wider circumstances. Countries with high levels of corruption may achieve rapid economic growth, possibly because corruption, although high, is being actively tackled by the government, or because other determinants of macroeconomic performance are more influential. Thus, while there is general acceptance of the relationship between corruption and macroeconomic impact described above, any assessment of the level and impact in a specific country must be approached with humility. In this respect, the Fund will need to continue its analytical work on the macroeconomic impact and other characteristics of corruption.18

Fund Initiatives on Governance and Corruption

13. Beginning with the Guidance Note itself, the Fund has introduced a number of initiatives on governance—and corruption more specifically—since 1997.19 The Guidance Note was developed against the backdrop of growing global attention to governance and corruption by bilateral donors, international organizations and nongovernmental advocacy organizations in the 1990s.20 Following the end of the Cold War and its attendant geo-political exigencies, donor countries faced growing domestic public pressure to more closely scrutinize the use of development assistance. At the same time, an acceleration in global economic, financial, and technological integration increasingly exposed many developing and transition countries to a globalized environment. In this situation, sound macroeconomic policies, transparent, accountable and effective public institutions, and adherence to the rule of law (including protection of property and investors’ rights)—rather than geopolitical alliances—became prerequisites for attracting foreign direct investment.

14. The Guidance Note followed earlier Executive Board papers that explored what role the Fund should have on governance and corruption issues.21 These papers emphasized that, while the Fund had always recognized the importance of good governance, governance issues had become more prominent for the reasons mentioned above. Accordingly, it was necessary to ask whether and how the Fund could do more to help countries deal with governance and corruption problems within the scope of its mandate.

A. Key Features of the Guidance Note22

15. While the Guidance Note acknowledged that the Fund’s existing work helped promote good governance, it sought to introduce a more systematic approach to Fund engagement. Starting from the premise that greater attention to governance issues could contribute to macroeconomic stability and sustainable growth, the Guidance Note recognized that the Fund already provided assistance to strengthen governance in a number of relevant areas. This involvement—through surveillance, UFR, and capacity development (CD)—was largely of a “preventative” nature, helping improve governance, limit the opportunities for corruption, and increase the likelihood of exposing poor governance. For example, Fund advice aimed to create systems that limit the scope for ad hoc decision-making, rent-seeking, and preferential treatment. Further, the Fund encouraged liberalized exchange, trade, and price systems; public sector accountability; transparent financial transactions; and robust accounting, auditing, and statistical systems. In addition, the Fund addressed specific issues of corruption when they were judged to have a significant macroeconomic impact. With this baseline, the Board sought to provide greater attention to Fund involvement in governance issues.

16. Based on Fund experience, the Board provided four guidelines for future involvement in governance issues:

  • a more comprehensive treatment in the context of both Article IV consultations and UFR of governance issues that are within the Fund’s mandate and expertise;

  • a more proactive approach in advocating policies and the development of institutions and administrative systems that aim to eliminate the opportunity for rent-seeking, corruption, and fraudulent activity;

  • an evenhanded treatment of governance issues in all member countries; and

  • enhanced collaboration with other multilateral institutions, in particular the World Bank, to make better use of complementary areas of expertise.

17. While recognizing that the responsibility for good governance lies primarily with the authorities, the Guidance Note identified specific areas of relevance to the Fund. The Board provided that “the IMF’s involvement in governance should be limited to the economic aspects of governance.” The Fund’s contributions, through policy advice and capacity development, are envisaged in two principal spheres: (1) improving the management of public resources through reforms covering public sector institutions; and (2) supporting the development and maintenance of a transparent and stable economic and regulatory environment conducive to efficient private sector activities. Within these spheres, the Board expected the Fund to focus on areas within the Fund’s traditional purview and expertise—e.g., tax administration; central bank operations; statistics; exchange, trade, and price systems; tax regulations; banking sector laws—and to rely on the expertise of other institutions, especially the World Bank, in areas outside the Fund’s expertise. However, while the Fund’s focus is on the economic consequences of poor governance and corruption described above, these consequences can arise from “non-economic” causes, whether state-capture settings and monetized governance, systemic weaknesses in accountability frameworks, data and reporting inadequacies, or institutional weaknesses with wide-ranging impacts (such as judicial weaknesses).

18. Generally, Fund involvement should be guided by an assessment of whether poor governance would have significant current or potential impact on macroeconomic performance in the short and medium term and on the ability of the government credibly to pursue policies aimed at external viability and sustainable growth. To identify governance problems, staff should be alert to inconsistencies or improbabilities in data, the lack of transparency in macroeconomic policy implementation, and other aspects of poor governance that would influence economic performance. The Board recognized that staff faced clear practical limitations in this regard and advised continued reliance on information provided by the authorities, with the possibility of recommending a third-party audit or collecting information from third parties. Staff should raise a given instance of corruption with the authorities where there is a reason to believe it could have significant macroeconomic implications (either because of its monetary value or because it is symptomatic of a broader problem), even if the effects are not precisely measurable. Instances of corruption that do not meet this threshold are best addressed through the Fund’s efforts to promote transparency, improve processes, and remove unnecessary regulations.

19. The Board also provided specific guidance for the three modalities of Fund involvement in governance issues.

  • Surveillance. In Article IV consultations, the staff should: be alert to the potential benefits of reforms that can contribute to good governance; be alert to the potential risks that poor governance can have on the economy—e.g., on market confidence, capital inflows, and investment; and provide policy advice that reflects cross-country experience and is based on broadly agreed international best practices. In terms of corruption involving international transactions, the staff should pay attention to both sides of the transaction. Where poor governance with significant economic impact is evident and brought to staff’s attention, staff should raise it with the authorities.

  • UFR. The above surveillance guidance is relevant for UFR, but the need for adequate safeguards for UFR must also be considered. Weak governance should be addressed early in the reform effort, including through use of Fund conditionality; the Fund should work closely with other institutions in areas where it does not have expertise but that are important for the program. Weak governance that threatens the successful implementation of the program or the safeguards for Fund resources may result in suspension or delay of Fund financing.

  • Capacity development. The Fund may decide, if requested, to perform financial and technical services that are consistent with the purposes of the Fund. The Fund provides technical assistance and training in a wide range of areas that are conducive to promoting good governance. Staff should bring to the attention of the authorities areas in which procedures and practices are found to be falling short of best international practices.

20. The Guidance Note stressed the importance of collaboration. While the Fund should exercise independent judgment in formulating policy advice, collaboration with multilateral institutions—particularly the World Bank—is critical on governance issues in areas outside the Fund’s competence that are important for the achievement of the economic policies advocated by the Fund. The Board also encouraged staff to seek to assist member countries with policy advice and capacity development to help address relevant governance issues raised by bilateral donors and to raise its own governance concerns with donors in certain circumstances.

21. The Board also identified areas in which the Fund should not be involved. First, although it is difficult to separate the economic and political aspects of governance, the Fund should not interfere in politics. However, it is legitimate to seek information about the political situation as an essential element in judging the prospects for policy implementation. Second, the Fund may not adopt the role of an investigative agency or guardian of financial integrity. Third, the Fund should avoid actions prejudicial to domestic legal processes.

22. The standards under the Guidance Note for Fund engagement on governance and corruption issues in the context of surveillance and UFR were not substantially changed by later decisions adopted for surveillance and program work.

  • In bilateral surveillance, the standard was introduced in 2007 and is currently embedded in the 2012 Integrated Surveillance Decision: “In its bilateral surveillance, the Fund will focus on those policies of members that can significantly influence present or prospective balance of payments and domestic stability.”23 The focus on such policies is intended to cover both their “macroeconomic aspects and macroeconomically relevant structural aspects,” which could include corruption.24 The 2015 surveillance staff guidance note indicates that structural polices and issues that aim to raise growth (including in the area of corruption), should be assessed only if they are “macro-critical”—that is, they affect, or have the potential to affect, domestic or external stability.25 26

  • In program engagement, the Guidelines on Conditionality, introduced in 2002, specify that conditionality is to be established only where the measure is critical for achieving program goals, monitoring of program implementation, or implementation of specific provisions of the Articles.27

23. Standards for engagement emphasize the short- to medium-term macroeconomic impact of corruption. This time horizon is specified in the Guidance Note as well as in the Integrated Surveillance Decision, which provides that, to the extent possible, the Fund’s assessment of a member’s policies and its advice to a member will be anchored in the short- to medium-term context, which is normally understood as encompassing a time frame of up to 5 years.28 Similarly, the Guidelines on Conditionality indicate that programs should be directed to “achieving medium-term external viability while fostering sustainable economic growth.”29 In the UFR context, this is generally understood to encompass the repayment period as a matter of safeguards (3-5 years for the credit tranches, up to 4-10 years for the EFF, and 5.5-10 years for use of PRGT resources).

24. For purposes of this stocktaking, the staff team has taken the position that the foregoing standards for assessing the macroeconomic impact of corruption are broadly consistent. Specifically, in most instances there would be a broad equivalence between emphasizing that the Fund should only engage on corruption issues where they have a significant impact on macroeconomic performance (Guidance Note) or are macro-critical (Integrated Surveillance Decision). The standards for using program conditionality apply more narrowly within circumstances where corruption is already judged to be macro-critical, being limited to instances where new measures are critical for meeting program objectives (UFR). However, one would still expect to see more in-depth coverage of corruption (or any given issue) in the UFR context, given the greater staff resources and more frequent engagement through multiple reviews each year that accompany a Fund-supported program as opposed to surveillance-only engagement. Further, greater coverage would also be expected in UFR cases where addressing corruption had been identified as part of the solution to a member’s balance of payments problem.

25. Since its adoption in 1997, the Guidance Note has been reviewed twice by the Executive Board, in 2001 and in 2004, with no modifications. The first assessment focused on the general implementation of the Guidance Note while the second discussed the application of the Guidance Note in the context of surveillance, as part of a broader biennial surveillance review.30 As further detailed in Annex I, both reviews concluded that implementation of the policy was broadly appropriate.

B. The Fund’s Initiatives and Policy Approaches on Governance and Corruption

26. As noted above, any sustainable strategy to address corruption requires efforts to promote good governance more generally. While building effective institutions and increasing transparency and accountability are not specifically directed at addressing corruption and have broader benefits, they do have an important impact on corruption. This section provides an overview of key initiatives of the Fund—some of which have developed since the Guidance Note—that fit into this category. Consistent with the Guidance Note, these initiatives help members to enhance institutional capacity in the management of public resources and in effective policy-making. In general, they focus on advocating appropriate economic reform and regulation, enhancing transparency and accountability, promoting the rule of law, and building institutions.31 The Fund is particularly active in advice on reforming economic regulation; improving public financial management; fiscal and monetary transparency; financial sector assessments; central bank governance and transparency frameworks; the transparency, quality, and timeliness of statistical data; and anti-money laundering/combating the financing of terrorism (AML/CFT). In addition, the Fund provides advice on enhancing the rule of law.

27. As will be seen, many of these activities cut across all of the Fund’s functions: capacity development, surveillance, and use of Fund resources. For example, while AML/CFT represents an important component of the Fund’s CD, it also can be relevant to the Article IV surveillance process (when weaknesses in these frameworks can undermine domestic stability) and Fund conditionality (when AML measures can, for example, help address tax evasion).

Reform of Economic Regulation

28. The Fund often advocates the roll back of excessive public economic intervention. In general, inefficiency, economic distortions, and corruption can flourish where public officials are “gate-keepers” for regulations, permits, contracts, and other tools of bureaucratic power. In this regard, economic reforms designed to reduce excessive regulation may be driven, in the first instance, by a desire to improve economic efficiency and reduce distortive practices. But these reforms also have the benefit of narrowing the scope for rent-seeking and corruption. A wide range of Fund policy advice has these characteristics, including measures to streamline laws and regulations, reduce trade barriers (Figure 4), simplify price and exchange market systems, strengthen tax administration, divest state enterprises, promote free and fair market-entry regulations, and improve natural resource management.

Figure 4.
Figure 4.

Market Depth/Openness and Corruption

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

29. However, the Fund is mindful that premature deregulation can pose its own risks, particularly where institutional frameworks are underdeveloped. The challenge, therefore, is to design balanced regulatory frameworks that maintain the benefits of regulation while minimizing its negative effects. In this regard, the Fund seeks to promote rules, procedures, and criteria that are as targeted, clear, simple, and transparent as possible.32

Public Financial Management (PFM)

30. The Fund provides advice on strengthening fiscal frameworks with the goal of enhancing the budget as the central instrument for allocation of public resources.33 At the budget preparation stage, the Fund promotes the adoption of comprehensive budgets prepared in general government context and strong budgetary institutions including medium-term fiscal and budgetary frameworks, fiscal rules, performance-based budgeting, and explicit fiscal risk management. At the budget execution stage, the Fund advises on commitment and arrears control, payment processes, fiscal reporting and accounting, debt and treasury management, and financial management information systems. The Fund also advises on fiscal institutions, including the legal and regulatory framework, independent fiscal agencies, and the organization of ministries of finance and central agencies. In addition, the Fund provides selective advice on public investment management, intergovernmental fiscal relations, subsidy management and quasi-fiscal activities.34

Fiscal Transparency

31. The Fund promotes fiscal transparency to enhance fiscal management and accountability. It helps governments to have a more accurate picture of their fiscal position and prospects and the long-term costs and benefits of policy changes. It also provides legislatures, markets, and citizens with the information they need to hold governments accountable. The Fund’s Fiscal Transparency Code and Evaluation was first published in 1998, and updated in 2007 and 2014. Fund staff analyze countries’ adherence to the principles and practices advocated by the Code.35

Financial Sector Assessment Program (FSAPs) and Standards and Codes

32. The Fund regularly assesses member countries’ compliance with international standards and codes established by, or developed with, other international institutions. An initiative to prepare Reports on Standards and Codes (ROSCs), launched in 1999, helps to monitor country compliance in 12 areas relevant for the operational work of the Fund and the World Bank.36 37 The reports are prepared and published at the request of the member country and help sharpen policy discussions with national authorities, including on core principles of independence, accountability, resourcing, legal protection for supervisors, and corporate governance.38

Key Fiscal Accountability and Transparency Initiatives

Public Expenditure and Financial Accountability (PEFA). PEFA provides a framework for assessing and reporting on PFM strengths and weaknesses using quantitative performance indicators.

Debt Management Performance Assessment (DeMPA). The DeMPA, developed by the World Bank and the Fund in 2007, is a tool for assessing the effectiveness of institutional arrangements for public debt management.

Fiscal Transparency Evaluations (FTEs). FTEs are the Fund’s principal fiscal transparency diagnostic tool. They provide quantified analyses of the scale and sources of fiscal vulnerability based on a set of fiscal transparency indicators, a summary of country fiscal transparency strengths and reform priorities. Advice on a sequenced fiscal transparency action plan is then given to help address reform priorities.

Tax Administration Diagnostic Assessment Tool (TADAT). TADAT was launched in 2015 after development by the Fund and other organizations. TADAT assessments pinpoint the relative strengths and weaknesses of tax administration.

Transparency in the Extractive Industries. The Fund issued the Guide on Resource Revenue Transparency for application in natural resource-rich countries and expanded its Fiscal Transparency Code to add a pillar on the sector.

33. The standards and codes applied to cover the financial sector represent a key component of FSAPs established in 1999. These standards help countries assess vulnerabilities in the financial sector and identify corrective actions, including on institutional and governance issues.39 They also seek to promote transparency in monetary and financial policies, banking supervision, insurance supervision, securities regulation, and payments systems. FSAPs also go beyond the observance of standards and codes, encompassing a review of the quality of the financial sector policy framework and financial sector safety nets, including governance. Since 1999, many FSAP exercises have been conducted by the Fund. In September 2010, the Fund made it mandatory for 25 jurisdictions with systemically important financial sectors to undergo financial stability assessments under the FSAP every five years; this was later revised to 29 jurisdictions.40 The 2014 FSAP review acknowledged the structural limits of the exercise due to its dependence on counterparts’ level of cooperation and quality of data provided.41

Safeguards Assessment Policy

34. The safeguards assessment policy was introduced in March 2000, in the wake of instances of misreporting and allegations of misuse of Fund resources. A safeguards assessment is a diagnostic review of a central bank’s governance and control framework. While the safeguards policy’s main objective is to mitigate risks of misuse of Fund resources and misreporting of monetary data under Fund-supported programs, it also promotes good governance, accountability, and transparency in central banks. It does this through promoting, inter alia, strong “checks and balances” within board and management structures, as well as sound internal control, auditing, and financial reporting mechanisms. The policy has been subject to reviews every five years by the Executive Board. The last review was concluded in October 2015 and endorsed a sharper focus on governance as an overarching principle of the safeguards assessment framework. Experience has shown that the safeguards framework has helped identify serious governance and misreporting issues at central banks. While the safeguards policy assisted central banks in progressing toward adopting international standards and best practices, it is no panacea against lapses in controls or fraud. The increased emphasis on governance in safeguards work aims to mitigate such incidents by bringing out the importance of independent oversight on bank operations as a counterweight to possible risks of high level overrides of controls. It covers five key areas within central banks of borrowing countries, namely, the external audit, legal structure, financial reporting, internal audit and control mechanisms (the ELRIC framework). To date, 296 assessments covering 96 central banks having been completed.

Experience with Central Bank Governance

35. The Fund assists institution-building in central banks at many levels. It provides support to establish clear mandates through design of objectives, functions, and powers in governing legislation. This includes appropriate governance arrangements, with a clear allocation to decision-making bodies of responsibilities pertaining to daily management, policy setting, and oversight. Ensuring sound internal controls and providing for checks and balances within and between the various decision-making bodies is also critical. This is supported by putting in place appropriate systems, procedures, operating manuals, and rules of conduct and integrity. In addition, functional, personal, and financial autonomy are key to ensuring that political bodies do not unduly hamper the execution of the central bank’s functions, that central bank officials are autonomous from political and private economic interests, and that the central bank has sufficient financial resources to fulfill its mandate. However, in parallel, the Fund also advocates explicit and robust accountability and transparency mechanisms by central banks vis-à-vis the relevant political bodies and the public at large, including through the publication of periodic reports and of audited and certified financial statements prepared according to internationally recognized audit and financial reporting standards. Ultimately, the goal is to assist in developing expert institutions and officials that are independent from private influence and political interference.42

Statistics and Data Dissemination

36. The availability of timely economic and financial statistics promotes good governance by fostering accountability of policy makers and market evaluation and discipline. Quality and timely data dissemination initiatives have been key in providing uniformed references that constrain the discretion of policymakers by offering tools that allow the assessment of their financial and economic policies. Data dissemination also plays a critical role in ensuring efficient allocation of resources by allowing markets to evaluate and impose discipline on government policy.

37. The Fund’s data standards initiatives have been crucial in promoting worldwide data dissemination and transparency. The Fund has taken steps to enhance member country transparency and openness, including setting voluntary standards for dissemination of economic and financial data. The Special Data Dissemination Standard (SDDS) was established in 1996 to guide members that have, or might seek, access to international capital markets in providing their economic and financial data to the public. The General Data Dissemination System (GDDS) was established in 1997 for member countries with less developed statistical systems as a framework for evaluating their needs for data improvement and setting priorities. In 2012, the SDDS Plus was created as an upper tier of the Fund’s Data Standards Initiatives to help address data gaps identified during the global financial crisis. In 2015, the enhanced GDDS (e-GDDS) replaced the GDDS.43 Under the e-GDDS, the Fund has helped 16 developing countries establish online National Summary Data Pages, which serve as a “one-stop shop” for public access to the countries’ main macroeconomic data.

Anti-Money Laundering/Combating the Financing of Terrorism

38. In 2000, the Fund responded to calls from the international community to expand its work in the anti-money-laundering area.44 After the events of September 11, 2001, the Fund intensified its AML activities to include CFT. The Fund’s work in this area covers policy advice, capacity development, and the assessment of compliance with international standards. Working closely with the Financial Action Task Force and its regional bodies, the Fund aims to embed AML financial integrity principles and practices into government ministries and central banks across the globe. It does so by building and strengthening institutions across a broad spectrum of activities, including financial intelligence, legislative drafting, national strategies, risk assessments, and the supervision and regulation of bank and nonbank entities. This is sometimes achieved through helping countries develop national AML strategies.45 Since 2000, over 70 AML/CFT assessments have been completed by the Fund. Several of these assessments helped countries make a transition from, or avoid, sanctions and the black- and gray-listing process.46

39. The Fund’s experience shows that the AML/CFT framework is an increasingly important tool for fighting corruption. Because the proceeds of corruption and tax evasion are often laundered to avoid detection or confiscation, an effective AML framework can contribute to both prosecuting and deterring corruption. Under international standards, AML frameworks should require heightened scrutiny of financial transactions conducted by “politically exposed persons,” adequate transparency of company ownership, and credible specialized institutions such as financial intelligence units. With regard to beneficial ownership of companies and assets, the AML/CFT standards have taken on added significance given recent incidents such as the Mossack Fonseca documents. These leaks provided a window into the scale of global financial secrecy and the potential opportunities to hide the proceeds of corruption and related tax evasion.47

Enhancing the Rule of Law

40. The predictable and effective enforcement of a legal framework is critical for an anti-corruption strategy. A credible threat of criminal prosecution of public officials who engage in corrupt acts and confiscation of their ill-gotten assets can serve as a powerful deterrent. Often, however, in the Fund’s experience, where the problem of corruption is particularly acute, a key challenge is that the very institutions charged with enforcing the law—investigative agencies, the police, the public prosecutor, and the judiciary—are themselves corrupt or beholden to powerful interests. They are part of the problem, not the solution. In these circumstances, some countries such as Indonesia, Kenya, and Ukraine have sought to put in place, with Fund support, specialized institutions that can prosecute corruption while longer-term reforms of the traditional law enforcement institutions are implemented. In practice, these institutions have typically been retained—even after improvement of the mainline institutions—because of their specialized skills. In a number of countries—including the three mentioned above—the Fund has also recommended the establishment and implementation of asset declarations frameworks for senior public officials as a means of monitoring income and assets that may potentially derive from corruption.

41. The Fund has also provided broader advice to enhance the rule of law. In this category, measures include developing national anti-corruption strategies; strengthening the legislative and judicial frameworks to effectively criminalize and prosecute corruption; establishing official codes of conduct; and conducting various types of audits, including payroll audits to identify ghost workers.48

Assessing Corruption and its Macroeconomic Impact in Country Cases—Methodology

42. A key expectation established in the Guidance Note is that, where justified based on macroeconomic impact, the Fund would proactively engage in the area of governance and corruption in the context of both surveillance and use of Fund resources. This section sets out the methodology used to assess the extent to which the expectation set forth in the Guidance Note has been met. The assessment was made by distilling the results of an in-depth review of the Fund’s engagement with a representative sample of countries, involving the stocktaking team’s close reading and qualitative analysis of 830 individual surveillance and program staff reports and press releases for the period of 2005 through 2016. Given the IMFC’s own emphasis, the focus of the review has been on engagement on corruption issues, rather than governance more generally, although, as will be seen, the latter term has often been used as coded language for the former. Moreover, as has been noted, reforms designed to address corruption often need to enhance governance more generally. Staff read each sample country’s documents in full and made a qualitative analysis of each document’s discussion (if any) of corruption.

A. Choice of Sample Countries

43. Given the difficulty of conducting an in-depth review of all the surveillance and program staff reports since 2005 for all 189 Fund members, it was first necessary to select a broadly representative sample of countries. Staff judged that 40 countries—just over 20 percent of the membership—would be sufficiently representative to analyze the Fund’s engagement on corruption issues.49 The country selection is illustrative of the diversity of the Fund’s membership in terms of income, region, and type of Fund engagement (both surveillance and program) (Table 1). Specifically, the countries are generally proportionate to the distribution of Fund members across different income groups (8 AMs, 16 EMs, 16 LIDCs). Eight countries have been included from each area department to represent possible differences in approaches across departments. Of the 40 countries, 24 have had program engagement with the Fund at least once over the period 2005–16. The sample was designed to include countries where the Fund’s engagement on corruption is evident and widely-known, cases where the extent of engagement is less evident, and those where some CSOs have viewed the Fund as insufficiently proactive. The design also allows the stocktaking to explore different aspects of the implementation of the Guidance Note, including whether the Fund’s past engagement has varied across countries depending on the risk of corruption as assessed by third-party indicators.

Table 1.

Country Sample for Document Review1/

Countries in bold had Fund program engagement during the period 2005–16.

The low-income developing countries (LIDCs) are a group of 60 countries that were designated as eligible for IMF’s concessional financing under the PRGT in the 2013 PRGT-eligibility review, and had a level of per capita gross national income less than the PRGT income graduation threshold for non–small states (i.e., twice the IDA operational threshold, or US$2,390 in 2011) and Zimbabwe.

B. Assessing Corruption and its Macroeconomic Impact

44. For the stocktaking team to assess whether the level of engagement by mission teams in these countries was appropriate, it was necessary to establish a methodology—independent of the staff reports—that could be relied upon to assess both: (a) how significant a problem corruption was for each of the countries in question; and (b) the extent of the macroeconomic impact. This exercise was designed to establish an “expectation baseline”; i.e., the level of engagement one would have expected from the Fund in each of these countries. On the basis of this assessment, the stocktaking team could then turn to the relevant staff reports and evaluate the extent to which the coverage of corruption was appropriate in the country in question. Ideally, the expectation baseline would have been constructed through an in-depth review—conducted independently of the staff reports—of the level of corruption and macro-economic impact in each of the sample countries. However, given the importance of having a meaningfully large sample of countries, such an approach would have been excessively resource-intensive and would have taken considerable time to complete. As an alternative, a two-step methodology was used to establish the expectation baseline. Of course, it is recognized that one of the problems with establishing any form of expectation baseline, no matter how it is constructed, is that it represents an ex-post measurement that may not accurately capture the full information available at the time of the original engagement.

Step One: Applying an Appropriate Combination of Third-Party Indicators

45. As a first step, the stocktaking team used third-party corruption indicators available for the 40 sample countries. These data established a general baseline assessment of the level of corruption in the sample countries vis-à-vis the rest of the world. Recognizing the limitations of any attempt to measure corruption, staff approached the use of indicators cautiously, referencing multiple sources.50 Consideration was given to the perspectives offered by three corruption indicators, which are not produced by the Fund: (i) the World Bank Worldwide Governance Indicators Control of Corruption Index (CCI); (ii) Transparency International’s Global Corruption Barometer (GCB); and (iii) the Maplecroft Corruption Risk Index (CRI) (see Box 2).51 In addition to being produced by well-recognized institutions and providing broad global coverage, these indicators were chosen to represent each of the three well-accepted approaches to measuring corruption: perceptions of corruption, experience of corruption, and the quality of anti-corruption institutions, respectively. While the indicators provided different approaches to measuring corruption, staff recognizes that these indicators ultimately seek to assess the same phenomenon, and there is some overlap in the underlying sources. Reflecting data availability, staff used CCI indicators for 2005–16; the GCB indicators for 2005–13; and CRI indicators for 2013–16. Each country’s annual global rankings were averaged over the available period to account for movement over the review period. This was important because the stocktaking team did not want to assess corruption-related risks to be low for the survey period based on a favorable corruption ranking in 2016 if that country’s rankings were less positive in earlier years.

46. Corruption rankings are subject to significant margins of error, so the assessment sought only to divide the sample countries on the basis of their position in the global quartiles for each third-party indicator. The average historical rankings of all countries assessed under these indicators are shown in Supplementary Paper, Background Note III, Annex III, Tables 1-3. Each sample country’s average ranking over time was assigned a quartile ranking based on these global rankings (Annex II). For some countries, the three measures of corruption were consistently in the same global quartile, while for others, measures spanned two (or in one case, three) quartiles. For analytical purposes, countries were grouped according to the predominant global quartile of their corruption indicators.52 The predominant global quartile was chosen based on having two or more of the three corruption measures in that quartile.53

Overview of Third-Party Indicators Used in Stocktaking1/

World Governance Indicators—Control of Corruption Index

The Control of Corruption Index (CCI) is one of six governance indicators compiled and published by the World Bank. Data have been compiled annually since 1996, and cover more than 200 countries. The CCI aggregates different sources of information into a single index to assess perceptions of the extent to which public power is exercised for private gain. As with all perception-based measures, it is open to various influences, such as media reporting, and there are risks that the indicator may not track actual patterns of corruption, particularly where these have changed significantly over time (for better or worse). The data also do not capture the nature of corruption, such as distinctions between petty and grand corruption.

Transparency International—Global Corruption Barometer

The Global Corruption Barometer (GCB) draws on survey feedback from more than 100,000 respondents in over 100 countries. Survey results are available for 2005–09, 2011, and 2013. For this review, staff used the subcomponent of the barometer which measures respondents’ feedback on whether they paid a bribe to access various public services in the last twelve months. One shortcoming of the GCB surveys is that they capture the experiences of the population at large, which typically relates to street-level corruption, not grand corruption by high-level officials. Another limitation is that differences in reported bribery might in part be subject to cultural differences in respondents’ willingness to report it.

Maplecroft—Corruption Risk Index

The Corruption Risk Index (CRI) is one component of the governance index compiled by Maplecroft. The governance index measures experts’ assessments of the reported prevalence and persistence of corruption in the public and private sectors, and the efficiency of governments in tackling the issue. The CRI is available annually since 2009 for 198 countries, and is based on a qualitative study of the qualities of institutions in each country. Analysts assess ten key indicators relating to the legislative framework (structure); anti-corruption implementation bodies and the practical applications of the law (process), and the frequency with which various kinds of corruption occur (outcomes). Potential shortcomings in this indicator include its sensitivity to the views of experts (rather than actual corruption experience), and the challenges of comparing the qualities and effectiveness of institutions across countries with different governance systems.

1/ The use of these non-IMF indicators to select a sample of countries to assess the Fund’s treatment of corruption issues does not imply the Fund’s endorsement of the quality of specific indicators; nor do they by themselves represent the Fund’s assessment of the levels of corruption in any country. For more details, see Supplementary Paper, Background Note III, Annex II.

47. Relying exclusively on indicators has important limitations. Importantly, it is static in nature and does not reflect the evolving nature of institutions and differences in countries’ efforts to combat corruption. Where measures are being taken to tackle corruption, this would likely have a positive macroeconomic impact through the investment climate and other transmission channels, independent of the historic level of corruption. The review team sought additional insights into the characteristics of corruption and related policies for the country sample, but given differences in the extent and nature of such information, it was not possible to confidently use it to further refine the above country groupings on a consistent basis. Perhaps most importantly, even if the indicators provided a reliable measure of the level of corruption in a country, they could not be relied upon to assess the specific nature of corruption in each country and the macroeconomic impact of the corruption that exists. While the overall economic theory elaborated above in discussing the macroeconomic impact of corruption is robust, it is general in nature and could not establish an expectation baseline for the macroeconomic impact in each of the sample countries.

Step Two – Review of External Reports

48. Accordingly, as a second step—and to supplement the indicators—information was compiled from a country-by-country survey of publicly-available external reports prepared by international organizations, academia, CSOs, and the general media. These external reports provided important insights into both the nature of corruption in each of the sample countries and the macroeconomic channels of transmission in each country. When the stocktaking team later turned to the review of Fund staff reports, these insights allowed for a more accurate assessment of the adequacy of Fund engagement. It should be noted, however, that external sources did not exist in comparable quantities, quality, or specificity for each of the sample countries.

49. In staff’s view, these two steps helped establish an expectation baseline that, while imperfect, provided a reasonable basis for conducting a qualitative review of each country’s surveillance and UFR staff reports over the review period. Although the external reports confirmed that both the nature of corruption and the nature of its macroeconomic impact varied among countries, some general patterns emerge that are valuable in assessing the Fund’s engagement across the country sample. Specifically:

First Quartile Countries

50. Information for the six countries (all advanced economies) did not point to corruption on a scale commensurate with significant macroeconomic impact. External sources suggested that everyday transactions are largely corruption-free in these countries, benefitting from strong anti-corruption institutions, policies and practices. At the same time, the external reports referred to high-profile corruption cases (e.g., Korea and Spain), most frequently at the intersection between business and politics—revolving door job offers, regulatory capture, and the difficulty in drawing the line between legitimate lobbying/political contributions and undue influence of legislative and regulatory processes. For these countries, inappropriate influence over the legislative, regulatory, and supervisory authorities by private interests could potentially have important macroeconomic effects (for example, by contributing to financial sector instability). Identifying the extent of such risks is difficult, however, and the broad issue of the effectiveness of political institutions is an area where the Fund should not interfere.

Second and Third Quartile Countries

51. External sources point to more significant levels of corruption—and associated macroeconomic impact—in the second and third global quartiles. The group is a large one, comprising, in the review sample, mostly emerging markets, but also two advanced economies and four LIDCs. Generalizing across the group, external reports pointed to relative weaknesses in anti-corruption policies, judicial independence, and overall transparency. For some countries in this group, external reports highlighted the tension between official policies on prosecuting corruption and the fact that it is tolerated in practice. With greater impunity for public officials, petty corruption appears to be more common than in the first quartile. A number of external reports pointed to a significant macroeconomic impact for this group, including with respect to overall growth performance. Similarly, the academic literature also identifies corruption as a key determinant of income differentials relative to wealthier economies.

52. Several countries in this group have experienced significant adverse macroeconomic developments in the wake of political instability. Several countries in the second and third quartiles have seen large scale demonstrations, social unrest, or even popular uprisings linked, at least in part, to a backlash against corruption (e.g., Brazil, Egypt, Georgia, Tunisia), which can also usher in periods of uncertainty, instability, and investor withdrawal. Egypt and Tunisia saw only weak growth in the years after the Arab Spring, and sovereign rating downgrades in Brazil and South Africa were in part caused by corruption-related political instability.

53. A key challenge for staff in analyzing developments in these countries is to assess the potential macroeconomic impact of corruption where it may not be apparent in the immediate conjuncture. For example, growth performance may prove strong, even as corruption is high, reflecting external developments (such as favorable export prices) or domestic factors (such as very high investment rates). Moreover, even where the impact of corruption is currently hidden, it can quickly emerge, such as in the cases of political crises noted above. As is discussed in the subsequent sections, calibrating the Fund’s engagement to reflect both current and potential macroeconomic impact is not straightforward.

Fourth Quartile Countries

54. Countries in the fourth quartile face the most acute levels of corruption (and associated macroeconomic impact) and often present some of the most difficult challenges in terms of gaining traction on the need for reform. External reports confirmed that corruption is generally more ingrained in political and economic life in this quartile; anti-corruption institutions are weak; public officials operate with greater impunity; and citizens have been less successful in challenging governing elites. In some cases, this has undermined both the authority and credibility of the state in core areas of governance, and powerful elites compete over the spoils of corruption, often coopting legislative and judicial processes, and sometimes resorting to violence. Two-thirds of the group are LIDCs, and one-third are classified as fragile states. Several countries in the group have experienced natural resource-related corruption (linked to the exploitation of oil, diamonds, and other minerals).

55. Although circumstances vary, the literature generally points to the potential impact of systemic corruption in terms of fiscal sustainability, growth, and the overall business and investment climate. In addition to instances of political instability and civil strife, economic governance, broadly defined, is also often weak for this group of countries and, in its engagement, the Fund would typically need to consider the phasing of its engagement on corruption issues relative to other priorities for macroeconomic stabilization and reform. As with the second and third quartile countries, macroeconomic performance has varied widely across the group; some countries have achieved strong growth in recent periods, underscoring that the scale of corruption is only one of several factors influencing the pace of economic growth.

56. It should be emphasized that the above methodology is being used here only for the purposes of this retrospective stocktaking review—it is not being proposed as the methodology to be used in determining future Fund engagement. An appropriate approach to guide Fund engagement on corruption issues going forward would be taken up in a future Board paper, providing the Board with the opportunity to offer guidance on the proposed methodology.

Fund Engagement on Corruption Issues—Findings

57. Taking into account the expectation baseline described in the previous section, this section assesses the Fund’s engagement in the country sample, based on an in-depth, qualitative review of Fund country documents on surveillance and use of Fund resources. Supplementary data on overall trends in the Fund’s engagement on corruption issues and patterns of engagement across countries according to measured levels of corruption are provided in Annexes V and VII. The findings below are somewhat generalized and should not be understood to apply to every country in a given quartile.

A. Findings on Quartile 1 Countries

58. The Fund did not normally engage substantively on corruption either in surveillance or program work for countries that fell within this quartile.54 While there were mentions of perceived corruption—including in the context of AML/CFT, regulatory capture, or bribery of foreign officials—these references were episodic and did not include a discussion regarding the actual or prospective macroeconomic impact of the perception of corrupt practices. Further, there was no discussion of collaboration with other organizations or the use of external expertise and indicators.

59. There was no discussion of or conditionality on corruption in the reports for Portugal, which was the only program country in the first quartile. Although the program identified significant inefficiencies in the judicial system relating to the enforcement of financial claims, there was no evidence that this inefficiency was due to corruption. The program included conditionality in tackling tax evasion and improving compliance, which by reducing the discretionary powers of tax authorities also aimed at improving governance in tax administration. The reports do cover governance weaknesses and fraud cases in the banking sector, many of which appeared after the program ended. In some cases, these triggered corruption investigations.

60. Overall, the treatment of corruption generally in Quartile 1 appears consistent with the specified standards for Fund engagement, with at least one important exception. For both surveillance and program staff reports, treatment of corruption issues was consistent with the country context and expected macroeconomic impact discussed above. That said, the 2004 review called for the “supply side” of corruption to be covered more effectively (e.g., through more reporting on implementation of the OECD convention on bribery of foreign officials in Article IV consultations or cross-country reports); this has not happened in practice. A further question is whether the limited analysis of regulatory capture in Q1 countries was commensurate with its potential macroeconomic impact.

B. Findings on Quartile 2, 3, and 4 Countries

61. Consistent with the country context, there was much more discussion of corruption in countries in these quartiles relative to those in the first quartile. A number of broad trends run through the quartiles.

  • Even where corruption was raised explicitly, the macroeconomic impact and time horizon within which the impact was assessed were rarely discussed, although, from the context, the impact could typically be discerned. Staff analysis most often focused on the impact on the business climate or, more rarely, on fiscal performance (lower revenue yields, higher procurement costs, etc.). However, coverage was typically brief, assuming a generic transmission channel with no discussion of specific country characteristics. In Q4, there were more instances in which the macroeconomic impact was discussed in depth.55 On the time horizon, staff generally did not address the potentially destabilizing longer-term impact of corruption in cases where it undercuts confidence in the economic governance regime, even in Q4 countries. Staff may have considered that this required speculation about the dynamics of the political economy that are outside the expertise of the Fund or that the possible longer horizon for this transmission channel was not consistent with the coverage of either surveillance or program conditionality.

  • The Fund’s engagement was more often responsive than anticipatory, suggesting that greater weight was attached to publicly-established cases of corruption than to corruption indicators discussed above. It often engaged more extensively and explicitly in the aftermath of political and economic crises that were perceived to have arisen because of high-level corruption scandals.56 This resonates with mission chiefs’ views that engagement on corruption was often influenced by high profile corruption cases and media coverage. Crises and scandals may have helped to justify Fund engagement, as the macroeconomic costs of corruption became more evident; they also provided firm evidence of corruption that could not be dismissed as conjectural or anecdotal.

  • Coverage of corruption issues tended to be episodic, rather than sustained, and was often quite generic in nature. Events were reported in isolation (a corruption scandal or launch of an anti-corruption initiative) without subsequent follow up. Important setbacks in the fight against corruption were rarely mentioned (e.g., when anti-corruption institutions came under fire in Indonesia). The authorities’ views on combating corruption were rarely described or discussed in detail. And where Fund policy advice was documented, it tended to be quite generic in nature. Examples included encouraging the government to streamline its role in the economy, and improving governance and transparency to strengthen the business climate.57 This may have reflected a lack of expertise among staff regarding specific advice to address corruption. This was not universal, however, with some Board documents offering detailed advice on policies (Box 3).

  • The language used in discussing corruption issues was often indirect, and required some “interpretation” by outsiders. A significant proportion of staff reports were judged to be discussing corruption-related themes, without explicitly using the term “corruption”; instead, indirect language was used (Box 4).58 Including indirect references, almost one-half of country reports covered corruption issues, at least briefly (Table 2). Based on feedback from mission chiefs, official sensitivities were likely a factor discouraging explicit references to corruption in some cases. Also, while some of the Fund’s engagement on efficiency, accountability, and transparency issues may have been motivated in part by corruption concerns, this is not always evident from Fund documents alone.

    Table 2.

    Coverage of Corruption in Selected Board Reports, 2005–161/

    Article IV consultation staff reports, selected issues papers, UFR and SMP reports for 40 countries.

    This captures both explicit text references to corruption as well as euphemistic references to governance and other topics that were interpreted by the review team as implicitly addressing issues of corruption. It excludes the larger proportion of countries that discuss governance issues in general, without any intended references to corruption.

    World Bank, Worldwide Governance Indicators, Control of Corruption Index.

    Transparency International, Global Corruption Barometer.

    Maplecroft, Good Governance Survey, Corruption Risk Index, 2016.

  • The analytical use of third-party indicators has been sporadic and limited. They were cited in only about 16 percent of country reports, with more frequent usage in Q3 and Q4 countries (Table 2 and Annex V, Figure 3).59 Moreover, their use was not well integrated into the general staff analysis. When indicators were used, practice varied significantly across departments, with MCD being the most frequent user (25 percent of reports) in comparison to other area departments (12-18 percent).

  • Collaboration with, and use of the expertise of, other agencies was discussed in some cases. References of this type appeared more frequently in program documents than in surveillance reports; and such references became more frequent in program documents as one moved from Q2 to Q3 to Q4 countries.

Examples of Detailed Fund Policy Advice

Detailed policy advice on strategies for reducing corruption was provided in several Article IV staff reports (SRs) and selected issues papers (SIPs). Advice in these cases was often developed with a view to informing ongoing or prospective Fund-supported programs, and reflected the findings of Fund capacity development missions and benefitted from collaboration with the World Bank and other partners.1/ Examples include:

Brazil: Advice on strengthening governance standards and anti-corruption safeguards, public procurement, banking and fiscal data transparency, conflict of interest legislation, and AML/CFT (2016 SR).

DRC. Measures to address corruption in the mining sector and associated loss of budget revenues—including tax policy and tax administration reforms (2007 SIP); state-owned enterprise (SOE) reforms (2012 SR); and a range of other governance measures (2014 SR).

Greece. Measures to strengthen revenue administration to help address tax evasion—an issue linked to public corruption (2013 SIP).

Kyrgyz Republic. Principles for the provision of tax incentives to minimize the scope for discretion-based corruption (2013 SIP).

Ukraine. Very specific and detailed advice both in surveillance and UFR on a broad range of issues: procurement, tax and customs administration, energy sector (in particular, the gas sector), privatization of SOEs, weak law enforcement framework (including anti-corruption and AML/CFT laws and institutions), ineffective judiciary, and regulatory simplification (2014-16 SRs).

Zimbabwe. Options for strengthening governance of the land reform process, drawing on a World Bank assessment (2005 SIP).

1/ In program documents, approaches for reducing corruption are typically part of the authorities’ policy commitments. While these policies are endorsed by the Fund, it is not straightforward to establish whether they reflect policy advice from the Fund or other partners.

Describing Corruption: The Use of Indirect Language

Mission chiefs indicated they had often opted to use indirect language when discussing corruption challenges. Examples cited included: need for a level playing field; efficiency of public spending; enhancing expenditure framework; implementation problems; governance issues; transparency and accountability; enhancing business climate/environment; and improving AML/CFT (see Annex IV and Supplementary Paper, Background Note VI).

The review of staff reports yielded additional examples of coded language used to refer to corruption: difficult or uncertain investment climate; notable weaknesses in the business climate; weak governance; pervasive governance deficiencies/challenges; vested interests; red-tape, rent-seeking; uneven administrative decisions; uneven implementation of the law; partial and discriminatory enforcement of laws; implementation bottlenecks; “pressures” under the tax system and “selective decisions” by officials; predictability of the tax regime; transparency and fairness of privatization; transparency of budgetary process; and connected lending.

62. Beyond these general observations, Fund engagement on corruption-related issues became more intensive as one moved into higher quartiles.

  • For Q2, in more than half of the country sample, corruption was severe enough that it contributed to major political crises (Brazil, South Africa, Tunisia, Turkey).60 This often hampered economic management, sometimes resulting in economic and financial stress. However, corruption was typically given limited attention, other than in Georgia and Greece, which were both program countries. More recently, in 2014-2016, there has been a trend of more engagement on corruption issues in Brazil, South Africa, and Tunisia. As an example, in 2016 the Fund engaged on a broad anti-corruption platform in Brazil, covering transparency measures, anti-corruption legislation, and targeted measures on anti-money laundering and against corruption.

  • In general, there was more pronounced discussion of corruption in Q3 countries compared to those in Q2—consistent with the view that less favorable corruption indicators pointed to greater likelihood of adverse economic fallout from corruption.61 62 However, as with Q2 countries, the Fund’s engagement in Q3 cases was broadly sustained only for a few of the countries—Mali and Indonesia. More recently, reports for some of the other countries have included a more open and in-depth discussion of corruption issues (Mexico, Peru).

  • As would be expected from the country context, compared to the other quartiles, the Fund engaged more actively on corruption issues in Q4 countries.63 As with Q3, Fund engagement with Q4 countries on broader issues of governance was also extensive, making it difficult to separate out the degree of focus on corruption issues. One half of the 18 countries showed a broadly sustained and intensive engagement on corruption issues. In another five of the countries, engagement may have been less over the review period but was sometimes quite in-depth. Coverage in the remaining four countries was more limited or episodic. Also in some cases, as in Paraguay, for example, the Fund provided support for capacity development on corruption-related issues, such as anti-money laundering and increased transparency through access to information.

63. More specifically, there was significant variation in engagement with countries in Q2-Q3. This variation in engagement may have been consistent with the Fund standards for engagement, differences in country circumstances, and the need to prioritize Fund engagement; it is difficult to draw sharper conclusions. Some countries were delivering relatively strong per capita income growth; Fund teams may not have assessed corruption to be a critical constraint on macroeconomic performance warranting continuous Fund engagement or may have seen the authorities’ anti-corruption efforts as broadly appropriate given the circumstances. Similarly, the more explicit discussion of corruption seen in some cases (e.g., Greece and Italy after the Euro crisis, and Tunisia after the Arab Spring) was likely warranted, given the greater urgency of addressing a more evident current macroeconomic impact of corruption in the context of weak underlying growth.64 The more substantive focus in Georgia was also appropriate, given the assessed high levels of corruption at the time reforms started in 2004.

64. It is reasonable to ask whether the Fund should have engaged earlier and more proactively in some countries, considering the potential impact of corruption even as growth was strong. In this connection, CSOs have argued that the Fund has been reluctant to tackle corruption in certain “politically-strategic” emerging markets. While earlier Fund engagement could possibly have helped to mitigate crisis risks, one challenge was to identify and assess the latent macroeconomic risks from corruption to justify a higher level of Fund engagement. While differing country contexts and the corresponding prioritization of the Fund’s work likely explain a significant part of the variation in the Fund’s engagement, significant differences between generally continuous engagement in some countries and more limited engagement with others raises questions of evenhandedness.

65. For Q4 countries, there was also considerable variation in engagement, which can only partly be explained. As with Q3 countries, some were delivering robust per capita income growth through much, if not necessarily all, of the review period. Fund teams may not have assessed corruption to be a currently binding constraint on macroeconomic performance, prioritizing other areas of Fund engagement. In some cases, the authorities’ anti-corruption policies may also have been assessed as broadly adequate, reducing the priority for Fund engagement. Also, some countries undertook major reforms; Fund staff may have allowed time for the reforms to bear fruit—such as the constitutional and major legal reforms undertaken by Kenya from 2010. At the same time, given the relatively high level of corruption in these countries and its potential impact on macroeconomic performance, it is more difficult to justify the limited or episodic Fund engagement in cases such as Libya and Zimbabwe. Given the risks that high corruption posed for governance institutions in Q4 countries, more consistent attention might have been expected on its potential for triggering economic instability—consistent with the identification of corruption in some of the staff reports as a source of insecurity and political instability with detrimental economic effects, for example, in DRC and Iraq.

66. A significant number of Fund-supported programs included corruption-related conditionality. For the sample of country reports covering the 2005–16 period, potentially almost half of program documents included such conditionality, depending on how the conditionality is interpreted (Table 3). Much of the identified conditionality was designed to strengthen governance, with only one in three reports referring explicitly to corruption as the motivating consideration. As noted above, durable solutions to corruption problems often require a broader approach to strengthening governance. However, the link between a diagnosis of corruption and a broader solution was often difficult to draw because of a lack of clear documentation of the diagnosis. A number of trends were evident:

Table 3.

Coverage of Corruption in Selected UFR/SMP Reports, 2005–161/

UFR and SMP reports for 40 countries, including combined Article IV/UFR reports.

  • Consistent with the expectations noted in paragraph 24, above, corruption was typically addressed with more breadth and depth when a country had a program than when it did not (e.g., Brazil, Greece, and Peru). Among the Q4 countries, three of the four with the least coverage of corruption issues were non-program cases. It is not entirely clear whether this reflects greater attention to corruption in a program context, or that staff assessed corruption to be more macro-critical in the context of the economic adjustment and reform needed to address macroeconomic imbalances found in countries seeking Fund support. Likely, both considerations were relevant.

  • It was often difficult to discern whether conditionality was designed to target corruption specifically. Given the limited specific references to corruption, the conditionality may have been designed to strengthen financial integrity, transparency, and accountability in general, or to address specific concerns regarding corruption.

  • Conditionality was initially over-ambitious in a few cases. Specifically, a few programs with extensive anti-corruption conditionality failed to deliver on the envisaged time-scale, possibly reflecting an overly optimistic assessment of prospects for the pace of institutional change. Over time, more gradual approaches to reform were adopted (e.g., in DRC, Haiti, Kenya, Kyrgyz Republic, Zimbabwe). In some cases, this evolution of conditionality involved a narrowing of focus to fiscal governance (PFM, tax administration, budget transparency), and a move away from explicit references to corruption. Given that the context for this change was not always explicit, it could have given a misleading impression that the Fund was downgrading its engagement on corruption.

  • Program traction may have been stronger for designing general anti-corruption frameworks than for achieving outcomes that address actual issues that arise. Insights into program conditionality are available from an unpublished 2015 study of cases of financial integrity failure authored by the African (AFR) and Fiscal Affairs (FAD) Departments (see Annex III and Box 5).

Examples of “Framework” versus “Outcome” Conditionality

Greece: Framework conditionality included the establishment of a single public procurement authority. Examples of outcome conditionality were the setting of specific targets for audits of asset declarations of tax administration officials, or the publication of the number of cases referred by the financial intelligence unit to the revenue administration to improve the use of AML tools to tackle tax evasion.

Honduras: Conditionality requiring the implementation of an anti-corruption strategy was an example of framework type, while conditionality requiring a census of public employees aimed at cancelling redundant positions is an outcome type.

Ukraine: The adoption of enabling legislation for establishment of an independent anti-corruption commission (NABU) and the preparation of a revenue administration reform plan to overhaul state fiscal service are examples of framework measures. Outcome measures were, for instance, the appointment of the Head of NABU Anti-Corruption Prosecutor and two deputies by a specific date under a specified procedure and the adoption and publication of a decision to unify and increase retail gas tariffs to a specified amount.

67. Of the seven countries in Q2, five had at least one program each during the review period. The use of conditionality varied among the countries. It seemed appropriate in Georgia, where conditionality was quite substantial at first when the fight against corruption was initiated. In subsequent programs, there was little conditionality due in part to an assessment that corruption had been significantly reduced. While there were no explicit references to corruption in the Turkey reports under the 2005–08 SBA, some of the conditionality was likely directed, at least partially, to addressing corruption. In Brazil, Greece, and Tunisia, while there were commitments from the authorities to institute various anti-corruption measures, the measures were not established as specific Fund conditionality, except for one measure in Tunisia’s current program. While the programs had different objectives, and the role of anti-corruption measures in meeting these goals differ across the countries, these considerations and the reasons for abjuring conditionality are not clear in the reports, leaving open the issue as to whether treatment was evenhanded across countries.

68. In the third quartile, five out of the nine countries had programs, but corruption-related conditionality was included only in one. In Egypt, conditionality on broader governance issues may also have been intended to address underlying corruption concerns important for program success, but this is not clear from the staff reports alone. The paucity of conditionality explicitly focused on conditionality in Bolivia and Mongolia is perhaps surprising given that the respective prior Ex-Post Assessments (EPAs) had concluded that corruption was a significant issue. There is also no clear explanation for why there was no corruption conditionality in two successive SBAs for Peru.

69. All except four of the 18 Q4 countries had at least one Fund-supported program during the review period. The extent of corruption-related conditionality varied across Q4 countries. Conditionality relating to corruption was significant in just over half of the program countries in the group—eight out of 14 cases. This appears consistent with the standards of engagement and the country context identified by the stocktaking. In the remaining six countries, conditionality focused on corruption issues was more limited, and their potential role for achieving program outcomes was unclear (Box 6). The varied treatment of conditionality in these countries could be seen as indicating lack of evenhanded treatment across countries—but it is important to underscore that what constitutes criticality for program success (which is the basis for choice of conditionality) varies markedly across countries and (for a given country) across different points of time.

70. In several cases, the Fund suspended financing until new corruption issues were addressed or because of missed program conditionality. For example, in February 2016, the Managing Director issued a statement indicating that concerns over Ukraine’s slow progress in improving governance and fighting corruption would make it difficult to continue the Fund-supported program. Additionally, the first review of Mali’s 2013 ECF arrangement was delayed pending corrective actions for serious lapses in PFM practices. Similarly, the fifth and sixth reviews under the 2002 PRGF arrangement with DRC were not completed because of delays in structural reforms, including in the transparency and accountability of SOEs operating in the corruption-affected minerals sector. Mongolia’s 2001 PRGF arrangement went irrevocably off track after the second review due to governance concerns. Outside the country sample for this review, lending was suspended to Mozambique, pending completion and publication of an audit of previously undisclosed loans (see footnote 48). Similarly, a 2013 staff press release at the conclusion of a mission to Malawi, which announced the delay of the completion of the third review under the ECF arrangement due to, inter alia, corruption.

Variation of Corruption-Related Conditionality in Countries with High Corruption Scores

Significant Conditionality

  • Afghanistan, DRC, Haiti, and Iraq are highly fragile countries where extensive conditionality relating to corruption was required. With Haiti, Iraq, and Zimbabwe, many of the conditions established on broader governance issues were probably also partly concerned with addressing underlying corruption.

  • During the review period, in Kenya, Nigeria, and Ukraine, the authorities have consistently acknowledged the adverse impact of corruption on the realization of significant growth potential. In Nigeria, for example, the authorities have made significant efforts to start addressing corruption, particularly in the oil industry. Regarding Ukraine, see Box 7, below.

Less than Expected Conditionality

  • Neither of Bangladesh’s two programs contained conditionality explicitly linked to corruption, but macroeconomic outcomes were generally strong, suggesting that conditionality was not critical for meeting program objectives. In Cameroon and Paraguay, most of the measures under the program were general commitments rather than specific conditionality. Given that Cameroon’s 2005 EPA highlighted the general lack of progress in earlier programs in strengthening governance and tackling corruption, specific conditionality could have been warranted in subsequent arrangements.

  • In the Kyrgyz Republic, conditionality targeted at corruption was limited and did not seem to match the apparent macro-criticality of corruption. In Uganda, corruption was robustly handled in the 2002-06 PRGF arrangement, consistent with the assessment in the 2005 EPA that corruption had cut into government revenues and led to wasteful spending, thereby affecting macroeconomic performance. However, with limited progress in implementing program measures to combat revenue agency corruption through 2005, the reduced attention to corruption-related conditionality in subsequent programs is noteworthy.

Ukraine: The Fund’s Engagement on Corruption—2014-2016

The Fund’s recent engagement with Ukraine on corruption issues is noteworthy for its openness, depth, and consistency. In 2014, in the context of discussions for a Fund-supported program, the authorities requested Fund staff to conduct a comprehensive diagnostic study of governance issues pertaining to corruption, the business climate, and the judiciary. The study found that corruption was pervasive and oppressive; that the business climate was hampered by an over-bearing and opaque regulatory framework; and that the judiciary was ineffective in prosecuting corruption as well as in resolving contractual disputes. It also emphasized that “state capture” of public policy and key economic sectors by oligarchs and other elites was entrenched. To the credit of the authorities, they agreed to publish the diagnostic report.

The Fund’s engagement on corruption issues has been guided by the July 2014 diagnostic study. This engagement also builds on the Fund’s successful experience establishing anti-corruption institutions in Indonesia during the Asian Financial Crisis.1/ Given corruption’s impact on Ukraine’s fiscal capacity, the business climate, growth, and overall trust in government, staff has recommended ambitious institutional reforms—several of which were part of program conditionality—in four main areas:

  • Legislative reforms to strengthen the laws on procurement, anti-corruption, asset disclosure by public officials, and anti-money laundering;

  • The establishment, with the active engagement of civil society, of a new, independent anti-corruption investigative agency (the National Anti-Corruption Bureau of Ukraine) and the appointment of an anti-corruption prosecutor;

  • The streamlining and simplification of the regulatory framework, including permits and administrative procedures, and inspections of businesses, to strengthen the business climate; and

  • Judicial reform to enhance independence, integrity and effectiveness. A new law on the judiciary was adopted in 2016, setting judicial reform in motion, starting with the Supreme Court.

While important progress has been made, the process of anti-corruption reforms has not always been easy, and it takes time for tangible results to emerge. The authorities’ willingness to expose the extent of the problem and to outline a reform strategy informed by the diagnostic study has allowed the Fund to work with them in a strategic and consistent manner to make fighting corruption a key component of their reform program. This program has also provided a framework of engagement with NGO’s, who have embraced and reinforced efforts to combat corruption, as well as with other stakeholders, including the European Union and bilateral creditors. In terms of outcomes, relatively fast progress was made in adopting legislation, streamlining regulations, setting up and operationalizing new institutions, and preparing and carrying out the first asset disclosure by public officials. However, fighting corruption, improving governance, and carrying out judicial reform are lengthy and complex processes, and the authorities’ continued determination and support is critical to root out corrupt practices and to foster a new culture of transparency and good governance.

1/IMF, 2004c

Conclusions and Lessons Learned

71. This section distills the key findings set out in the previous section and identifies possible lessons for the Fund’s future engagement on corruption issues. To put these findings in context, it is helpful to highlight some of the challenges that the Fund faces in engaging in this area—challenges that will need to be factored into guidelines and principles for future engagement.

  • A first consideration is the difficulty of assessing the extent and nature of corruption, given that corrupt activities are typically concealed. Judgment is needed on the sources of information to be considered, and their respective reliability for informing analysis and policy recommendations. For example, information that comes to the attention of staff may be anecdotal and general in nature, and therefore difficult to pursue.

  • A second challenge is to understand the transmission channels between corruption and macroeconomic performance, which is likely to depend both on the nature and extent of corruption, and on the wider context, including the short- to medium-term impact of other influences on macroeconomic outcomes. Some authorities remain reluctant to engage with the Fund on corruption issues, regarding it as primarily a political, rather than economic matter, or may be reluctant for discussions with Fund staff to be made public (see Annex IV and Supplementary Paper, Background Note VI).

  • A further challenge is that, because corruption is often hidden, the effectiveness of anti-corruption or other governance-improvement policies are difficult to quantify, and have their largest impact over the medium to long term.

  • Finally, there is the challenge of prioritization. Although the problem of corruption may be significant in a country, it will often compete with other policy challenges that lie within the Fund’s domain. Given its limited resources, the Fund often places priority—whether in the context of surveillance or use of Fund resources—in areas where the macroeconomic impact is the most immediate and predictable. In addition, when prioritizing the coverage of corruption, the level of the Fund’s engagement would depend on the level of support being provided by other organizations, notably the World Bank.

While many of these factors will remain a consideration, the Fund can help staff meet these challenges through a clear policy framework and operational guidance. This paper raises these issues for discussion by the Executive Board. Drawing on that discussion, a follow-up paper would identify what modifications should be introduced with respect to the Fund’s policies and practices in this area.

72. As a general matter, the review finds considerable progress in implementing the Guidance Note as regards the Fund’s engagement on corruption issues. The breadth and depth of coverage of corruption in many country reports and the Fund’s contributions to the corruption literature demonstrate a keen awareness that corruption can have significant macroeconomic effects relevant for Fund engagement. This engagement is most extensive, appropriately, in countries with the highest reported levels of corruption. For half of the countries in the highest global corruption quartile, the review found broadly sustained and intensive Fund engagement. Where these countries implemented Fund-supported programs, this contributed greater specificity, depth, and granularity to the engagement. Readiness to suspend financing helped achieve traction where corruption issues arose or were not adequately addressed within the programs.

73. The review reveals scope to further strengthen this engagement, however, as elaborated below. As already discussed, this exercise was only a stocktaking to identify areas that may warrant further guidance. If the Executive Board wishes to follow up on any of the lessons learned, the Board could request a follow-up Board paper elaborating on forward-looking principles or operational guidance for Fund engagement in governance and corruption.

A. Assessing the Scale of Corruption

74. It would appear important that the Fund develop its own consistent method for assessing the scale of corruption in a country. As a preliminary matter, staff’s discussions with the authorities would provide an important basis for the assessment. This could be complemented by the use of both: (a) third-party indicators; and (b) a set of structural characteristics shown to be of significance in affecting the level of corruption in a country. This set could include, for example, the robustness of the member’s public financial management, the extent—and transparency—of its economic regulation, the efficiency of its tax administration, and the effectiveness of its judiciary. Identifying these characteristics would be a key feature of a possible follow-up Board paper based on input from functional and area departments with relevant expertise.

75. A key issue is the identification—and appropriate role of—third-party indicators. As noted above, such indicators would be expected to play an important, but not exclusive, role in assessing the scale of corruption. Similarities across countries in a single corruption indicator may not accurately reflect potentially important divergences in corruption activity that warrant distinct engagement approaches. Economic circumstances and governing institutions vary across countries, implying potentially different transmission channels and levels of macroeconomic impact. Against that backdrop, staff could make greater use of the indicators not just to provide standalone snapshots, but to integrate them into staff’s own analysis, including by tracking indicators across time and making peer comparisons. Consideration should also be given to using a range of third-party indicators, since corruption can take many forms, and is not amenable to measurement using a single indicator.65 In practice, indicators vary considerably in methodology and focus, and their outcomes for a country do not always coincide. (For more detail, see Supplementary Paper, Background Note III.)

76. Coincidentally, staff guidance on the use of third-party indicators is being developed. This responds to recent questions from some Directors about the quality of several indicators and the lack of systematic guidance for their use in Board documents. A forthcoming Board paper will address issues relating to the transparency of the use of indicators (for example, being explicit about their characteristics, methodological shortcomings, and measurement uncertainties), the value of robustness checks (such as using multiple indicators or comparing with other inputs), and the importance of presenting the views of the authorities and/or other stakeholders, to the extent that they have a different interpretation.

B. Assessing Macroeconomic Impact—Understanding the Relevant Time Horizon

77. Further guidance will be needed on the method to be used to assess the macroeconomic impact of corruption (Annex IV and Supplemental Paper, Background Note VI). The review found that staff papers rarely discussed the degree or nature of macroeconomic impact. Arguably, this assessment was implicit in staff’s choice of when to engage on issues of corruption. However, the inferential link from corruption indicators to macroeconomic impact is not always evident or useful. Assuming that there is acceptance of the general analysis set forth in the discussion of the macroeconomic impact of corruption, it would be expected that, in the context of both surveillance and use of Fund resources, analysis of the nature of the macroeconomic impact of corruption would be more likely as the assessed level of corruption increases.

78. Developing guidance on how to assess the macroeconomic impact of corruption will require a discussion of the appropriate time horizon that should be used for this purpose. While the findings show that Fund staff adhered to the existing standard set forth in the Guidance Note (as also reflected in the Integrated Surveillance Decision and the Guidelines on Conditionality), there is a need to reconsider whether the short- to medium-term horizon is a sufficient time frame for assessing the impact of corruption and the remedies to address it. For some countries with severe corruption, the macroeconomic impact of corruption may be easily disentangled from the role of other factors that influence economic performance; in other cases, particularly where economies are growing rapidly, this may be less easy, particularly over a short time horizon. The full impact of corruption can unfold only over a significant period of time; remedial measures may also take an extended time to yield results.

79. The necessity of using a longer time horizon to assess economic impact is not unique to corruption. It is an issue of broader relevance for macroeconomic and macro-structural policies—including inequality, gender issues, and climate change—and, indeed, has been raised in those contexts. Thus, any further guidance in this area may need to be coordinated as a broader policy matter.

C. Providing Policy Advice in the Context of Surveillance

80. In addition to the problems identified above regarding diagnosing the problem of corruption, there also appears to be issues regarding formulating effective policy advice in the surveillance context once a problem is diagnosed. The review shows a difference between Fund engagement on corruption issues in surveillance relative to programs. Analysis and advice on corruption issues in program cases tends to involve greater specificity, depth, and granularity than under surveillance, where policy advice on strengthening frameworks and addressing corruption concerns is often generic in nature. For long-time surveillance-only countries, Fund engagement on corruption markedly increased at a program’s inception, while engagement on corruption issues often waned once a country exited a program. Where corruption is addressed as part of surveillance, attention is also commonly episodic, which can give the impression that poor governance is an occasional event rather than well-established, even where other information, including external sources and third-party indicators, point to a more endemic problem.

81. As already discussed, to a degree, a difference in coverage between surveillance and UFR cases should be expected. This likely reflects, in part, the fact that measures to address corruption are more critical when macroeconomic performance has deteriorated to the point that a Fund-supported program is necessary. In addition, staff resources dedicated to programs, including the provision of targeted technical assistance, are typically much more extensive than for surveillance, allowing for more in-depth coverage of a range of policy issues, not limited to corruption. Feedback from staff does suggest, however, a need for more guidance on framing surveillance policy advice (Annex IV and Supplementary Paper, Background Note VI). Staff has indicated they would welcome advice on how to tailor the Fund’s policy advice to most effectively combat corruption. Such guidance could help distill the central elements of policies to address corruption as well as the general components of effective anti-corruption strategies. In addition to issues of policy design, guidance could also discuss how to achieve traction. The overall approach to addressing corruption would no doubt be informed by the choice of those characteristics of an member’s economy that are relied upon for purposes of assessing whether corruption is systemic. Thus, to the extent that weak frameworks for public finance management are relevant when assessing the level of corruption, strengthening these frameworks would be an important element of any robust anti-corruption strategy.

D. Use of Program Conditionality

82. When conditionality was used, it generally appeared to be well justified and streamlined. In this regard, it appears that the Fund learned the lessons presented in the 2008 Kenya EPA, in which the Fund found that the program had been over burdened with governance measures with a tenuous link to program criticality. Other EPEs and EPAs presented similar lessons, stressing the importance of conditionality that is streamlined, macro-relevant, and tailored to implementation capacity (Annex VI). While this lesson is applicable to conditionality generally, corruption issues present a particular challenge given the sensitivity of the topic and vested interests that may oppose reform. Mission chiefs welcomed clearer guidance on designing policy advice and conditionality in this regard (Annex IV and Supplementary Paper, Background Note VI).

83. Where program ownership has been very strong, good progress has been made in combating corruption, even with limited Fund conditionality on corruption. For example, Georgia made great strides in reducing everyday corruption since 2004, even though corruption was not a major focus of Fund-supported programs after 2007. Where corruption is embedded in the political economy and resistance to change is strong, progress can be slow, even with extensive conditionality (e.g., Afghanistan, DRC, Iraq, Ukraine, Zimbabwe). In these cases, the Fund sought to progressively strengthen institutional practices, often starting with fiscal governance. The AFR-FAD review found that, in some cases, only prospective or actual suspension of the Fund’s financial assistance was sufficient to galvanize the authorities to focus on and address the underlying causes of financial integrity failures (e.g., Malawi 2013, Mozambique 2016).

84. As is the case in the surveillance context, greater guidance on how to tackle specific corruption challenges would be welcomed in framing policy advice in the context of Fund-supported programs. Such guidance could discuss realistic goals and phasing of reforms, particularly given the limited time horizon of Fund-supported programs. It could also weigh the respective merits of process- and outcome-oriented goals and the documentation of the goals and outcomes of program conditionality.

E. Transparency of Fund Engagement

85. While the use of indirect language may have helped address sensitivities, heavy reliance on such language can cloud staff’s analysis and policy recommendations. In some cases, a parallel review of external sources such as reports of other international organizations, academia, CSOs, and the media by the stocktaking team helped to reveal the Fund’s underlying, implicit engagement on corruption issues.66 However, this was not always possible and should not, in any case, be necessary for interpreting Fund reports. In some cases, “interpreting” staff reports would likely be difficult beyond “insiders” familiar with the language and tone of Fund staff reports. It should be acknowledged that the use of coded language could, in part, be interpreted as adherence to the Guidance Note’s admonition to avoid action prejudicial to any related domestic legal processes—or to avoid commenting on specific cases under dispute. Further, coded language could have been strategically used in certain cases to increase traction with the authorities on sensitive issues and thereby aid adoption of the Fund’s policy advice. Nevertheless, there is significantly more room for staff reports in general to be more direct and explicit in discussing corruption-related issues, particularly in explaining the Fund’s diagnosis of corruption-related challenges. As noted above, however, proposed solutions may be appropriately framed as increasing transparency and improving governance more broadly. It is interesting to note that explicit references to the word “corruption” have varied over time, though it is not clear whether this reflects changing attitudes toward direct language or changing patterns of overall engagement (see Annex V).67

F. Evenhandedness of Fund Engagement

86. The stocktaking raises questions about the evenhandedness of the Fund’s engagement on corruption issues, in the sense of treating similarly situated countries in a similar manner. The principle of uniformity of treatment does not mean that all members must be treated equally, but rather that any decision by the Fund to differentiate between members must be based on the application of criteria that are relevant to the objective of the exercise being conducted by the Fund. Differences in the level of Fund engagement across countries assessed to have equivalent levels of corruption may be warranted at a point in time when there are higher priority concerns in one country than another (e.g., an impending currency crisis). Differences in the level of engagement can also be expected in surveillance cases vis-à-vis program cases, given the intensity of Fund engagement in the latter context. That said, the review makes clear that the variation in treatment of corruption issues across countries with high levels of corruption was usually not clearly explained by reference to specific country circumstances. Differences in treatment may be explained in large part by the above-described challenges facing staff in assessing the nature and extent of corruption, understanding the transmission channels, and prioritizing engagement (see paragraph 71). Ensuring greater uniformity of treatment will be further enhanced if additional guidance is provided in many of the areas discussed above; i.e., how to measure corruption; how to assess macroeconomic impact; the nature of the advice to be provided, etc.

87. The framework put in place in 2016, in response to the Triennial Surveillance Review’s findings about the evenhandedness of Fund surveillance, can also help address these concerns in surveillance. This framework articulated principles for evenhanded surveillance, focusing on “inputs.” These principles guide the allocation of resources to reflect countries’ individual or systemic risk factors and emphasize that policy advice reflects sound, objective analysis tailored to country circumstances.68 Ongoing application of these evenhandedness principles, including to coverage of corruption and governance issues, should help deliver a more evenhanded treatment of these issues.

G. Collaboration with Other Institutions

88. Collaboration with other institutions, especially the World Bank, is important for sharing information on country circumstances and developing policy approaches for addressing corruption risks. The contributions of each institution would reflect their areas of comparative expertise (see Supplementary Paper, Background Note II). The 2004 review noted that external cooperation is strongest in a program context, and this remains the case. That said, explicit information in country documents on the nature of this collaboration is relatively scarce, except where the anti-corruption agenda is very active, in which case, Board documents sometimes refer to collaborative work with the World Bank and some regional developments banks. In general, there appears to be scope to collaborate more with other international organizations.

Issues for Discussion

  • Do Directors agree with the summary of the main areas of macroeconomic impact of corruption (paragraphs 5-12)?

  • Do the findings on Fund engagement on corruption issues (paragraphs 57-70) resonate with Directors?

  • Do Directors agree with the “Conclusions and Lessons Learned” regarding the scope for more effective Fund engagement on issues of corruption (paragraphs 71-88)?

  • Do Directors support the development of principles and approaches that could inform updated staff guidance on governance issues including, in particular, corruption?

References

  • Aidt, Toke and Jayasri Dutta, 2008, “Policy Compromises: Corruption and Regulation in a Democracy,” Economics and Politics, Vol. 20, No. 3, pp. 335-360.

    • Search Google Scholar
    • Export Citation

  • Aidt, Toke, 2011, “Corruption and Sustainable Development,” International Handbook on the Economics of Corruption, Vol. 2, pp. 3-51.

    • Search Google Scholar
    • Export Citation

  • Akai, Nobuo, Yusaku Horiuchi, Masayo Sakata, 2005, “Short-run and Long-run Effects of Corruption on Economic Growth: Evidence from State-Level Cross-Section Data for the United States,” CIRJE-F-348, available at http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.691.6238&rep=rep1&type=pdf [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • Alam, M. Shahid, 1989, “Anatomy of Corruption: An Approach to the Political Economy of Underdevelopment,” The American Journal of Economics and Sociology, Vol. 48, Iss. 4, pp. 441-456.

    • Search Google Scholar
    • Export Citation

  • Arnone, Marco and Eleni Iliopulos, 2007, “The Cost of Corruption: Economic, Institutional, and Social Effects”, Milan: Vita e Pensiero.

    • Search Google Scholar
    • Export Citation

  • Blackburn, Keith, Niloy Bose, and M. Emranul Haque, 2006, “The Incidence and Persistence of Corruption in Economic Development,” Journal of Economic Dynamics and Control, Vol. 30, No. 12, pp. 2447-67.

    • Search Google Scholar
    • Export Citation

  • Crivelli, Ernesto, and Sanjeev Gupta, 2016, “Does Conditionality in IMF-Supported Programs Promote Revenue Reform?” International Tax and Public Finance, Vol. 23, Iss. 3, pp. 550-579.

    • Search Google Scholar
    • Export Citation

  • Huntington, Samuel P., 1968, Political Order in Changing Societies (New Haven: Yale University Press).

  • IMF, 1996, “The Role of the Fund in Governance Issues,” December.

  • IMF, 1997a, “The Role of the Fund in Governance Issues—Guidance to the Staff,” April.

  • IMF, 1997b, “Statement of the General Counsel on the Role of the Fund in Governance Issues—Guidance Note,” July.

  • IMF, 1997c, “The Role of the Fund in Governance Issues—Guidance Note,” July, available at: http://www.imf.org/external/pubs/ft/exrp/govern/govindex.htm [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2001a, “Enhancing Contributions to Combating Money Laundering: Policy Paper,” April, available at: http://www.imf.org/external/np/ml/2001/eng/042601.PDF [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2001b, “Review of the Fund’s Experience in Governance Issues,” March, available at: https://www.imf.org/external/np/gov/2001/eng/report.htm [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2001c, “The Acting Chairman’s Summing Up on the Review of the Fund’s Experience in Governance Issues,” February.

  • IMF, 2002, “Guidelines on Conditionality,” Decision No. 12864, September 25, as amended, available at: https://www.imf.org/external/pubs/ft/sd/index.asp?decision=12864-(02/102) [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2004a, “Biennial Review of the Implementation of the Fund’s Surveillance and of the 1977 Surveillance Decision,” July, available at: https://www.imf.org/external/np/pdr/surv/2004/082404.htm [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2004b, “The Chairman’s Summing Up of the Biennial Review of the Implementation of the Fund’s Surveillance and of the 1977 Surveillance Decision,” August, available at: https://www.imf.org/external/np/pdr/surv/2004/082404.pdf#pin [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2004c, “Legal, Judicial and Governance Reforms,” Indonesia: Selected Issues, IMF Country Report No. 04/189, available at: http://www.imf.org/external/pubs/ft/scr/2004/cr04189.pdf [last visited June 13, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2011, “Review of the Standards and Codes Initiative,” February, available at: https://www.imf.org/external/pp/longres.aspx?id=4550 [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2013, “Review of Facilities for Low-Income Countries,” March, available at: http://www.imf.org/external/np/sec/pn/2013/pn1345.htm [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2014a, “Press Release: IMF Executive Board Reviews Mandatory Financial Stability Assessments Under the Financial Sector Assessment Program,” January, available at: [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2014b, “Review of the Financial Sector Assessment Program—Further Adaptation to the Post-Crisis Era,” August, available at: http://www.imf.org/~/media/Websites/IMF/imported-full-text-pdf/external/np/pp/eng/2014/_081814.ashx [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2014c, “Update on the Fiscal Transparency Initiative,” June, available at: http://www.imf.org/external/pp/longres.aspx?id=4888 [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2015, “Guidance Note for Surveillance under Article IV Consultations,” March, available at: https://www.imf.org/~/media/Websites/IMF/imported-full-text-pdf/external/np/pp/eng/2015/_031915.ashx [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2015, “Options for Low Income Countries’ Effective and Efficient Use of Tax Incentives for Investment,” October, available at: https://www.imf.org/external/np/g20/pdf/101515.pdf [last visited May 18, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2016a, “Evenhandedness of Fund Surveillance—Principles and Mechanism for Addressing Concerns,” February, available at: https://www.imf.org/~/media/Websites/IMF/imported-full-text-pdf/external/np/pp/eng/2016/_012716.ashx [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2016b, “Corruption: Costs and Mitigating Strategies,” Staff Discussion Note 16/05, May, available at: http://www.imf.org/external/pubs/ft/sdn/2016/sdn1605.pdf [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMF, 2017, “Building Fiscal Capacity in Fragile States”, April.

  • IMFC, 2000, “Communiqué of the International Monetary and Financial Committee of the Board of Governors of the International Monetary Fund,” September, available at: https://www.imf.org/external/np/cm/2000/092400.htm [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMFC, 2016a, “Communiqué of the Thirty-Third Meeting of the International Monetary and Financial Committee,” April, available at: https://www.imf.org/en/News/Articles/2015/09/28/04/51/cm041616a [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMFC, 2016b, “Communiqué of the Thirty-Fourth Meeting of the International Monetary and Financial Committee,” October, available at: https://www.imf.org/en/News/Articles/2016/10/08/AM16-cm100816-Communique-of-the-Thirty-Fourth-Meeting-of-the-IMFC [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • IMFC, 2017, “Communiqué of the Thirty-Fifth Meeting of the International Monetary and Financial Committee,” April, available at: https://www.imf.org/en/News/Articles/2017/04/22/sm2017-communique-of-the-thirty-fifth-meeting-of-the-imfc [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • Lin, Shuanglin and Wei Zhang, 2009, “The Effect of Corruption on Capital Accumulation,” Journal of Economics, Vol. 97, Iss. 1, pp. 67-93.

    • Search Google Scholar
    • Export Citation

  • Mauro, Paolo, 1995, “Corruption and Growth,” Quarterly Journal of Economics, Vol. 110, No. 3, pp. 681-712.

  • Mauro, Paolo, 1996, “The Effects of Corruption on Growth, Investment, and Government Expenditure,” IMF Working Paper, No. 96/98. Available at: http://www.imf.org/en/Publications/WP/Issues/2016/12/30/The-Effects-of-Corruptionon-Growth-Investment-and-Government-Expenditure-2042

    • Search Google Scholar
    • Export Citation

  • Mauro, Paolo, 2004, “The Persistence of Corruption and Slow Economic Growth,” IMF Staff Paper, Vol. 51, No. 1. Available at: http://www.imf.org/external/Pubs/FT/staffp/2004/01/pdf/mauro.pdf

    • Search Google Scholar
    • Export Citation

  • Méndez, Fabio and Facundo Sepúlveda, 2006, “Corruption, Growth and Political Regimes: Cross Country Evidence,” European Journal of Political Economy, Vol. 22, pp. 82-98.

    • Search Google Scholar
    • Export Citation

  • Méon, Pierre-Guillaume and Khalid Sekkat, 2005, “Does Corruption Grease or Sand the Wheels of Growth?” Public Choice, Vol. 122, Iss. 1, pp. 69-97.

    • Search Google Scholar
    • Export Citation

  • Méon, Pierre-Guillaume and Laurent Weill, 2010, “Is Corruption an Efficient Grease?” World Development, Vol. 38, Iss. 3, pp. 244-259.

    • Search Google Scholar
    • Export Citation

  • Neeman, Zvika, M. Daniele Paserman, Avi Simhon, 2008, “Corruption and Openness,” The B.E. Gournal of Economic Analysis & Policy, Vol 8, Iss. 1, Art. 50.

    • Search Google Scholar
    • Export Citation

  • Rose-Ackerman, Susan, 1975, “The Economics of Corruption,” Journal of Public Economics, Vol. 4, Iss. 2, pp. 187-203.

  • Rose-Ackerman, Susan, 1978, Corruption: A Study in Political Economy (New York: Academic Free Press).

  • Svensson, Jakob, 2005, “Eight Questions about Corruption,” Journal of Economic Perspectives, Vol. 19, No. 3, pp. 19-42.

  • Ugur, Mehmet and Nandini Dasgupta, 2011, “Evidence on the Economic Growth Impact of Corruption in Low-Income Countries and Beyond,” EPPI-Centre Report 1914, Social Science Research Unit, Institute of Education, University of London.

    • Search Google Scholar
    • Export Citation

  • United Kingdom, 2016, “Against Corruption: A Collection of Essays,” May, available at: https://www.gov.uk/government/publications/against-corruption-a-collection-of-essays/against-corruption-a-collection-of-essays [last visited May 8, 2017].

    • Search Google Scholar
    • Export Citation

  • United Nations General Assembly, 2003, United Nations Convention Against Corruption, 31 October, A/58/422.

Annex I. Past Reviews of the Fund’s Experience with Governance1

1. 2001 Review. This review found that the 1997 Guidance Note remained generally appropriate as the framework for guiding the Fund’s approach in the area of governance and corruption. Directors agreed with its focus on prevention as the Fund’s main strategy, the exercise of judgment within relatively broad boundaries as to when to get involved in a specific issue, the importance of country ownership, the need for evenhandedness, and the need for active collaboration with other international organizations with relevant expertise (IMF, 2001b, 2001c). Directors stressed that the Fund’s involvement should be limited in line with the Guidance Note to economic aspects that could have a significant macroeconomic impact. In this regard, Directors called for further efforts to apply the “significant macroeconomic impact” test more explicitly and they encouraged Fund staff to do additional analysis and research on how to apply it more meaningfully. Directors were also of the view that the Fund should explore ways to pay more attention to the “two-sided” nature of corruption, namely both the countries affected by public corruption as well as those that are sources of bribery. They considered that engagement on the latter issue could potentially be addressed by assessing the status of the implementation of OECD Bribery Convention and similar initiatives.

2. 2004 Review. The second review also found that implementation of the Guidance Note was broadly satisfactory, though references to governance were largely descriptive and of a general nature. It was recommended that further efforts be concentrated on: (a) greater use of World Bank expertise and outside information; (b) developing new information where needed (such as through ROSCs on fiscal transparency or monetary and financial transparency); and (c) greater attention to policy recommendations (including staff views on the adequacy of steps taken and the needs for further action) (IMF, 2004a, 2004b). In particular, Directors called for staff to explore ways to refine coverage of governance issues in surveillance, such as through greater use of governance indicators. While Directors concluded in the 2001 review that governance and corruption issues should be addressed in future reviews of surveillance, technical assistance, and conditionality, the 2004 biennial surveillance review was the last to explicitly do so.

3. Good progress has been made on some of the Board’s recommendations, though open issues remain, and there is scope for further improvement on all fronts (see Conclusions and Lessons Learned above).

  • In general, staff is comfortable relying on World Bank expertise and other outside information. About half of mission chiefs report at least annual consultations with the World Bank, and a third consult with other organizations (Annex IV and Supplementary Paper, Background Note VI). Similarly, about half of mission chiefs report relying on external information with respect to governance and corruption issues. However, as noted in the Conclusions and Lessons Learned, there remains scope for improvement in this area. Not only could cooperation with the World Bank and other organizations be further strengthened, third-party indicators could be better employed. As noted above, staff guidance on such indicators is being developed in parallel.

  • The Fund has also had success in developing new information, where needed. Various Fund initiatives have been adopted or reviewed, and their governance focus sharpened, since the 2001 and 2004 reviews. For example, the Fund’s involvement in the global AML/CFT efforts has intensified since the September 11, 2001 terrorist attacks; the Safeguards Assessment framework has been reviewed and the Board endorsed the sharper focus on governance as an overarching principle of the framework; and the SDDS/GDDS initiatives to promote data transparency have been enhanced. The ROSCs initiative to promote fiscal and monetary policy transparency has been reviewed, the fiscal transparency framework revamped, and new codes adopted. Mandatory financial sector assessments have been extended to 29 jurisdictions with systemically important financial sectors.

  • One major shortcoming remains defining and applying the macroeconomic significance test. As detailed in the Conclusions and Lessons Learned and raised by mission chiefs (Annex IV and Supplementary Paper, Background Note VI), staff is still seeking more guidance on making this difficult assessment.

  • As noted in the Findings section of the paper, staff engagement often remains largely descriptive and of a general nature, with little explicit follow-up on earlier staff advice.

  • The two-sided nature of corruption has not been fully explored. While there have been discussions of the OECD’s Bribery Convention, the FCPA, and other measures, engagement on corruption issues is still weighted toward the countries affected by public corruption rather than the members that may be the sources of bribes.

Annex II. Groupings of Sample Countries based on non-IMF Indicators of Corruption 1/2/

Colors represent quartiles of countries, subdivided by corruption indicators (light blue=lowest corruption quartile through dark blue=highest corruption quartile).

These corruption indicators were used to allocate countries to specific quartiles for the purpose of assessing the Fund’s engagement on corruption across countries. They do not represent the Fund’s assessment of the current levels of corruption in any country.

Annex III. Summary of Unpublished 2015 AFR/FAD Study of Financial Integrity Failures

The six countries covered by the study (Ghana, Guinea, Malawi, Mali, Mozambique, and Uganda) had Fund-supported programs both before and after the episodes of fraud. This study explored the contributory role of underlying weaknesses in PFM systems; the focus, nature, and impact of prior Fund capacity development; the modalities of Fund-supported programs surrounding the events; and possible steps to enhance the Fund’s engagement in such cases. The study found that the pre-existing emphasis on strengthening PFM was reinforced in the post-fraud period. Implementation of conditionality was mixed, improving in some instances only with the actual or prospective suspension of the Fund-supported program. Implementation was better for “up-stream” legislative and strategic benchmarks focusing on the establishment of frameworks and plans than it was for “down-stream” measures to strengthen specific PFM practices in order to achieve a particular outcome, even though the latter were judged more important for addressing vulnerabilities to fraud (Table 1 of this annex).1 Despite weaknesses in SOE oversight and procurement that contributed to the frauds, these did not feature prominently in post-fraud conditionality—possibly reflecting a relative lack of staff familiarity with good practice in these areas.

Annex III. Table 1.

Compliance with PFM-Related Structural Benchmarks

Source: Fund staff calculations.

Annex IV. Views of Country Authorities, Organizations, and Mission Chiefs

1. This annex summarizes the views of key stakeholders on the issues addressed in this stocktaking exercise. Questionnaire and on-line feedback was obtained from 38 countries, 70 mission chiefs, 8 CSOs, and 8 international organizations; meetings were also held with staff from the World Bank and OECD; and seven CSOs participated in a conference call. The information obtained was inevitably partial in coverage, and should be regarded as indicative.1 While the main analysis rests on the review of staff reports set out in the main paper, the views of stakeholders provide a valuable cross-check on the documentary review. The supplement to this paper provides more comprehensive feedback from these stakeholders (see Supplementary Paper, Background Notes V, VI and VII).

Country Authorities

2. Country authorities recognize and value the Fund’s expertise and influence on corruption issues. They emphasized that the Fund’s global engagement provides it with unique perspectives on how corruption impacts macroeconomic outcomes. Countries highlighted the Fund’s comparative expertise in fiscal issues (fiscal transparency, PFM, revenue administration, subsidy reform), AML/CFT, safeguards assessments, trade and capital account liberalization, and general data and transparency issues. The policy leverage associated with the Fund’s lending activities on these and other areas was noted by several respondents.

3. Countries saw scope to strengthen implementation of the Guidance Note. There was general recognition that the macroeconomic impact test is not easy to apply, not least in the absence of clear operational guidance for staff. In this regard, some country respondents saw the Fund’s coverage of corruption as too general—failing to engage sufficiently forcefully with detail and follow-up where necessary. Others saw the Fund as engaging on corruption more forcefully in programs than in surveillance, or focusing mostly on grand corruption cases rather than routine corruption. There was also concern that corruption issues arising in emerging market and advanced economies such as regulatory capture are not adequately captured. Consistent with these varying views, two-thirds of respondents found shortcomings on the question of evenhandedness of the Fund’s engagement.

4. Authorities generally agreed that the time horizon for assessing macroeconomic impact should be over the medium- to long term. Two-thirds of respondents supported a time-horizon of up to 5 years. This was seen as helping to focus attention on areas where the impact was most imminent and also as being aligned with the time horizon of Fund-supported programs. One-third of respondents saw a five-year horizon as too short, suggesting that discussion of the risks of corruption in Fund surveillance warranted a longer-term perspective, given the time required to frame legislation, implement strengthened governance arrangements, and achieve an impact on outcomes.

5. Many members noted the challenges of using third-party indicators. While some found their existing use to be broadly appropriate, a somewhat larger group favored expanded use, informed by more regular staff discussions with the World Bank and other sources. Only a few supported the Fund developing its own corruption indicators.

6. Country authorities saw scope to increase collaboration with external counterparts. A significant group called for use of more external expertise, for example, on the nature of corruption and its political context, while maintaining the Fund’s focus on the macroeconomic aspects.

International Organizations

7. International organizations concurred on the importance of the Fund’s engagement in combating corruption.2 The Fund’s advisory role on sound economic and financial management was seen as giving it a strong position to influence and support ministries of finance, central banks, and other economic governance bodies. They highlighted the Fund’s conditionality as important leverage for difficult reforms. The Fund was also credited for its analytical work on the costs of corruption. Some organizations looked to the Fund for a deeper and more explicit engagement on corruption issues, and for this engagement to be sustained, given the medium- to long-term impact of corruption.

8. International organizations looked for stronger collaboration and communication, notably at a strategic level. While operational collaboration between Fund country teams and counterparts is viewed as generally effective, some looked for the Fund to play a more visible role in global anti-corruption programs and initiatives.

Civil Society Organizations (CSOs)

9. CSOs credited the Fund for increasingly addressing issues of governance and corruption, and looked for an expanded, albeit focused engagement. In this regard, the Fund’s value-added was in assessing and addressing macroeconomic impact, while concentrating on areas where its policy expertise provides leverage. However, they underlined the time required to make progress in fighting corruption, and advised the Fund to aim for continuous, steady engagement, rather than phases of elevated engagement and goals that may be beyond the capacity of the authorities to deliver. They further viewed more regular discussion of corruption during Article IV consultations as laying the foundation for effective Fund engagement in the event of a program request. CSOs also suggested that further work and collaboration (including with CSOs) is needed in measuring the extent and impact of corruption through indicators and other information. In the latter context, some CSOs saw merit in some Fund initiatives, such as the fiscal transparency evaluations, becoming regular, mandatory, and public. Finally, like country counterparts, CSOs saw scope for the Fund to improve on evenhandedness.

Fund Mission Chiefs

10. Mission chiefs noted that not all Fund engagement on corruption is documented in country documents. While the majority reported discussing corruption, only 40 percent indicated that these discussions were explicitly documented. Mission chiefs indicated that they sometimes used indirect language in staff reports when discussing corruption. These included: need for a level playing field; efficiency of public spending; enhancing expenditure framework; implementation problems; asset recoveries; governance issues; transparency and accountability; enhancing business climate/environment; and improving AML/CFT. Where corruption was not discussed, this was variously attributed to lack of information or staff expertise, sensitivity or disinterest on the part of the authorities, or corruption issues that fell outside the Fund’s macroeconomic mandate.

11. Many mission chiefs would welcome clearer guidance on implementing the key Guidance Note principles. About two-thirds either lacked confidence or were only somewhat confident in deciding whether to tackle corruption in a given country. They saw merit in strengthened guidance on how to assess macroeconomic impact. While most mission chiefs point to the use of macro data in determining engagement on corruption, a significant number noted the influence of high profile corruption cases and media coverage. About half of mission chiefs also rely on third-party corruption indicators and other external information, but also noted familiarity with shortcomings in some indicators.3 Mission chiefs would also welcome more guidance on the design of policy advice and conditionality (see Figure 1 of this annex). Reflecting these uncertainties, many mission chiefs saw shortcomings in the evenhandedness of engagement across countries.

Annex IV. Figure 1.
Annex IV. Figure 1.

Mission Chiefs’ Views on Priorities for the Staff Guidance Note

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Source: Survey of IMF mission chiefs, January 2017.1/ Based on responses from 37 of 70 mission chiefs—the same proportion that “agree” or “strongly agree” on a need for more specific guidance on how Fund staff should engage on issues of corruption.

12. Mission chiefs report generally high levels of engagement with other organizations. Almost half consult with the World Bank at least annually in developing Article IV consultation advice or program conditionality for combating corruption. About one third also consult with a wide variety of other organizations. Where mission chiefs held back from additional consultation, it was most often because corruption was not a major concern, they had doubts about the capacity of other institutions to provide clear and realistic solutions, or staff believed that the Fund already had sufficient expertise.4 Although Fund staff are familiar with the publications of some CSOs, few mission chiefs volunteered that they consult with these institutions (Table 1 of this annex).

Annex IV. Table 1.

Fund Staff Consultation with Other Organizations

Source: Survey of IMF mission chiefs, January 2017 (68 out of 70 respondents addressed this question).

Annex V. Overall Trends in Explicit References to “Corruption”

Since 2005, a little over one-third of country documents explicitly referred to corruption (Table 3 of main paper). Coverage varied over the sample period, with a decline during the period 2009-2010 (Figures 1-2 of this annex). This likely reflected a reprioritization of the Fund’s work during the global financial crisis. Other institutional considerations influencing the allocation of resources for work on corruption may also have played a role. Specifically, the 2012 guidance note on the inclusion of AML/CFT and financial integrity issues in surveillance contributed to corruption issues being more directly addressed in a number of subsequent Article IV consultations, with missions benefitting in some cases from LEG participation.1 On average, 16 percent of country reports cited third-party corruption indicators. The latter were used more frequently for countries in the higher corruption quartiles, though practice varied across countries (Figure 3 of this annex).

Annex V. Figure 1.
Annex V. Figure 1.
Annex V. Figure 1.

Frequency of References to “Corruption” in Fund Documents

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Source: Word search of surveillance and UFR country reports (SM and EBS series) in the IMF Institutional Repository for the 40 sample countries (see Table 1 of main paper).
Annex V. Figure 2.
Annex V. Figure 2.

Text References to Corruption

(Percentage of reports with references, 2005-2016) 1/

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Source: Fund staff calculations.1/ Blue bars show explicit references; red bars show implicit references.
Annex V. Figure 3.
Annex V. Figure 3.

Use of Third-Party Corruption Indicators

(Percentage of reports with corruption indicator, 2005–2016 1/

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Source: Fund staff calculations.1/ Horizontal axis shows percentage of reports including corruption indicator.

Annex VI. EPA and EPE Findings on the Fund’s Engagement on Corruption Issues

1. Ex-Post Assessments (EPAs) and Ex-Post Evaluations (EPEs) provide useful insights into the Fund’s engagement on corruption issues. Both EPAs and EPEs review holistically performance under completed Fund-supported programs.2 This annex draws on the findings of EPAs and EPEs completed during 2005-2016 for the group of countries examined in this stocktaking. Specifically, it covers EPAs for Cameroon, Mongolia, and Uganda (all 2005), Kenya (2008), Georgia (2011), and Afghanistan (2015), as well as EPEs for Ukraine (2013 and 2016).3 These EPEs and EPAs raised issues concerning the treatment of corruption issues in Fund programs.

2. EPAs/EPEs were commonly frank in describing the channels of corruption and its macroeconomic and program effects. In some cases, this candor reflects the focus of the underlying program documents (Afghanistan, Kenya, Ukraine), while in others the directness of the EPAs/EPEs provide a contrast to the approach of the underlying staff reports, where discussions were less direct (Mongolia, Uganda). The occasional difference in candor between the EPAs/EPEs and the underlying program documents may, in part, reflect the interdepartmental authorship and the nature of the exercise—i.e., one-off interaction versus long-term relationship with the authorities; seeking reactions of the authorities on a specific report versus collaboratively reaching recurrent understandings on program design.

3. A common finding is that addressing corruption requires a longer time horizon than provided by a Fund-supported program. EPAs and EPEs recognize that the authorities often make important progress on anti-corruption measures under Fund-supported programs, with attendant positive effects on growth and perceptions. However, EPEs/EPAs were also fairly forthcoming in recognizing that addressing corruption issues can be difficult and requires a long time horizon, given vested interests, entrenched behaviors, and other complicating factors. As a result, the full objectives of programs were often not realized. In Cameroon, an action plan to reduce corruption was adopted, but only a few concrete measures were taken and problems persisted. In Kenya, the authorities made substantive progress with structural reforms under the programs reviewed, and initially a perception emerged that the authorities were finally focused on addressing corruption. In the event, problems persisted and there were often long lags in implementation. In Mongolia, despite broad achievement of the macroeconomic objectives of the program, the Fund was not able to complete reviews under a Poverty Reduction and Growth Facility arrangement after significant governance concerns were highlighted that had not been addressed, leading to the conclusion that any successor program would have to address such governance issues. In Uganda, there were repeated attempts to curb corruption within the revenue authorities but reforms faced constant setbacks. And in Ukraine, the authorities produced a candid and comprehensive diagnostic study to lay the foundation for anti-corruption reforms, but specific measures faced attempts at dilution and delay, raising questions about the influence of vested interests.

4. Ownership by the authorities is key. A predictive factor in several countries for achieving success in the area of anti-corruption was buy-in from the authorities (Georgia, Kenya, Ukraine). The Georgia EPA states bluntly: “Key to the success of the program was the authorities’ strong ownership and determination,” noting that the new leadership in 2004 “undertook forceful measures to fight corruption, increase public sector efficiency, improve governance, and create a business-friendly environment.” By contrast, the two EPEs for Ukraine attributed delays in program implementation to waning ownership over time and the continued influence of vested interests.

5. EPA/EPEs underlined the importance of conditionality that is streamlined, macro-relevant, and tailored to implementation capacity. Where the Fund engages on corruption issues, EPA/EPEs considered it crucial that conditionality should clearly prioritize macro-critical goals and, while they should be ambitious, they need to be tailored to the authorities’ capacity to deliver. Excessive prior actions, over-burdened programs, and shifting goalposts driven by donor pressure could set unrealistic expectations and undermine ownership. In Kenya, for example, the EPA suggested that Fund conditionality in an early program was overly-reactive, driven by donor priorities. As a result, the program was overburdened with 19 structural governance measures and 70 anti-corruption measures in the MEFP. The EPA found that some of these specific measures “had tenuous links with macro- and program-criticality,” and resulted in a feeling the Fund was micromanaging the anti-corruption process. This arrangement was accompanied by slow progress, cyclicality, multiple setbacks, and off-track programs. After 2003, arrangements with a greater focus on macroeconomic measures designed to indirectly help address corruption achieved greater program ownership. By contrast, the most recent EPE for Ukraine found that conditionality was critical, well-articulated, and in the areas of the Fund’s competency. Ultimately, reports recommended that conditionality on corruption issues, as in all areas, should be tied to Fund expertise and tailored to country circumstances.

6. Collaboration with other organizations and donors is useful. The Kenya EPA highlighted the important role that collaboration with the World Bank played in changing the Fund’s thinking on governance and increasing traction. Following an increased focus on anti-corruption at the World Bank in 2005, the Bank sent a mission to Kenya to identify governance concerns and map a way forward. This involvement “helped the Fund move to a forward-looking institutional approach on governance” that resulted in better progress by the authorities and to completion of program reviews.

Annex VII. Cross-Country Patterns in Fund Engagement

1. This annex explores cross-country patterns of Fund engagement on corruption issues, notably in relation to information on levels of corruption. This can provide initial insights into “uniformity of treatment” (members in similar circumstances should be treated similarly).1 However, a more rigorous evaluation than has been achieved in this review would need to consider the complex ways in which corruption can differ across countries (even where third-party indicators might otherwise be similar), the macroeconomic impact of corruption in different countries, the resources available to the Fund for its engagement, and nuanced differences in the Fund’s engagement across countries, an aspect that is not typically documented in country reports. To this extent, the following evaluation is only a rough first-cut, prompting areas for further analysis of possible gaps in evenhandedness.

Overall Patterns of Engagement

2. Explicit references to corruption are generally more common for countries with worse corruption rankings. Looking at the full sample of 830 surveillance and program documents, explicit references to corruption were found in 57 percent of reports for countries in the fourth global corruption quartile, compared to 2 percent for countries in the first quartile (Table 2 of the main paper).2 On a regional basis, explicit references are most common in AFR and MCD (where countries typically have weaker corruption rankings) and least common in EUR (where country rankings are strongest). At a country level, the frequency of explicit treatment is correlated with corruption levels—whether measured using corruption perceptions or staff’s broader country groupings (Figure 1 of this annex). However, there is considerable cross-country variation that cannot be explained solely by corruption differences. Relative to peers, coverage of corruption is lower in some countries (e.g., Libya, Mongolia, Turkey) and higher in others (Cameroon, Italy, Mali).3

Annex VII, Figure 1.
Annex VII, Figure 1.

Coverage of Corruption Relative to Perceptions Rankings and Relative to Staff Groupings 1/

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Source: Fund staff Calculations.1/ Staff grouping of countries as described in Note 1.Asterisks on the correlation coefficient indicate significance at the 1 percent level.

Program-Related Engagement

3. For countries implementing Fund-supported programs, corruption-related conditionality was more frequent in cases with worse corruption rankings. However, the extent of corruption-related conditionality varied widely across programs, and the relationship with the World Bank corruption perceptions index is relatively loose (Figure 2 of this annex). In some cases, programs included a series of corruption-related benchmarks, even though corruption indicators are in the moderate range (Egypt, Turkey), while in other cases, programs lacked any explicit conditionality on corruption-related issues, even though indicators pointed to high corruption risks (Bangladesh, Mongolia, Paraguay).

Annex VII. Figure 2.
Annex VII. Figure 2.

Corruption Conditionality Relative to Corruption Perceptions

Citation: Policy Papers 2017, 004; 10.5089/9781498346528.007.A001

Source: Fund staff calculations.Asterisks on the correlation coefficient indicate significance at the 5 percent level.
3

See Supplementary Paper, Background Note I on Distinguishing Governance, Good Governance, and Corruption for a more extended discussion.

4

Article 19 of UNCAC on Abuse of Functions states that “Each State Party shall consider adopting such legislative and other measures as may be necessary to establish as a criminal offence, when committed intentionally, the abuse of functions or position, that is, the performance of or failure to perform an act, in violation of laws, by a public official in the discharge of his or her functions, for the purpose of obtaining an undue advantage for himself or herself or for another person or entity.” UN General Assembly, 2003.

5

For example, tax evasion and illicit cross-border flows engaged in solely by private actors are generally outside the scope of this paper except to the extent they relate to the proceeds of corruption.

6

IMF, 2016b defines “systemic corruption” as circumstances where “corruption is no longer a deviation from the norm, but is manifested in a pattern of behavior so pervasive and ingrained that it becomes the norm.” “Systemic corruption” has also been defined as corruption that “is both pervasive and organized, affecting different levels of government, and practiced by bureaucrats and politicians alike in nearly all government departments” Alam, 1989.

7

Most studies using perceptions-based measures of corruption identify a negative impact on growth. See Mauro (1995, 1996, 2004); Akai, et al. (2005); Blackburn, et al. (2006); Arnone and Iliopulos (2007); Neeman, et al. (2008); Ugur and Dasgupta (2011); see also Rose-Ackerman (1975, 1978), and Lin and Zhang (2009). Literature ascribing positive results to some forms of corruption tends to be older (Huntington, 1968), less grounded in empirical research, and limited to certain conditions. For example, Méon and Sekkat (2005) and Méon and Weill (2010) find that corruption is less detrimental where institutions are less effective. For the role of institutions in mediating the economic impact of corruption, see Méndez and Sepúlveda (2006), Aidt et al (2008). Another useful discussion involves the distinction between short-term effects, which can be sometimes positive, and long-run impacts, which are detrimental (Svensson, 2005; Méndez and Sepúlveda, 2006; Neeman, et al. 2008).

8

Aidt, 2011 (using the variable “genuine growth per capita”).

10

Recent data compiled by the IMF for 108 countries show that there is a strong correlation between a low C-efficiency (a measure of actual to potential VAT revenue) and high corruption. Fund staff’s analysis shows that an improvement in corruption perception from the median to the 75th percentile is associated with higher revenues of 0.8 percentage point of GDP.

11

The Mossack Fonseca documents bring into focus the scope of global financial secrecy and the potential it creates for tax evasion and other criminal activities. It also highlights the fact that tax evasion is an international, as well as domestic, issue that cuts across countries and governance frameworks.

12

Research by IMF staff suggests that even conditionality under IMF-supported programs—which generally has a positive impact on revenue performance—becomes ineffective in countries with high corruption. Crivelli and Gupta, 2016.

13

Corruption can also distort the selection of such projects and the composition of public expenditure more generally, while poor payroll controls can permit the proliferation of “ghost workers,” whose salaries are embezzled.

14

Inclusive growth can also be impacted where corruption undermines the buildup of human capital as well as the scope to use fiscal policy for meeting society’s redistributive goals.

15

Certain forms of regulatory capture in advanced economies may have played a role in the systemic failures of oversight, regulation, and disclosure during the 2008 global financial crisis. More generally, powerful financial companies can bend the regulatory, policy, and legal institutions for their private benefit through high-level bribery and influence peddling.

16

See IMF, 2016b at pp. 5, 15.

17

See Supplementary Paper, Background Notes I (on the meaning of “corruption”) and III (on indicators).

18

See Supplementary Paper, Background Note IV on Analytical Work of the Fund on Governance and Corruption.

19

The Guidance Note (IMF, 1997c) represents Board policy set out in a Board decision. This decision was labelled a “guidance note,” a misnomer that has sometimes caused confusion. It is not a staff guidance note approved by management to guide implementation of Board policy; it is itself the Board’s policy. In a future amendment of the policy, this misnomer will be corrected.

20

For further detail, see the Supplementary Paper, Background Note II on “Governance and Anti-Corruption Initiatives in International, Regional, and Other Fora.”

23

Integrated Surveillance Decision, Decision No. 15203-(12/72), 07/18/2012 (“ISD”), at paragraph 6.

24

Id.

25

“Macro-critical” is not used in the Integrated Surveillance Decision and is intended in the staff guidance note only as a shorthand reference to the standard in the Integrated Surveillance Decision described above. See IMF, 2015, at Box 2.

26

For issues that are macro-critical and where the Fund has in-house expertise, both staff analysis and advice are required. For those issues that are macro-critical but for which Fund expertise is lacking, staff should analyze the issues, drawing on expertise of other organizations, but is not expected to provide policy advice. IMF, 2015, at Paragraphs 7, 75, 76, 81.

27

IMF, 2002, at Paragraph 7(a).

28

ISD, at Paragraph 17. See also IMF, 2015, at Paragraph 4 and Box 1.

29

IMF, 2002, at Paragraph 6(b).

31

As an example, in the case of tax incentives in LICs, Fund papers shows the importance of good governance, including the need for a proper statutory basis and minimized discretion to combat corruption. It also describes progress that has been made in some countries following Fund engagement. IMF, 2015.

32

For example, a recent IMF regional project in the eastern Caribbean promoted a streamlining of discretionary tax exemptions, which could also help curb corruption. In Bolivia, Cote d’Ivoire, and Uruguay, automated risk management systems are being implemented to strengthen the integrity of customs services.

33

Fund TA in this domain increased significantly in recent years. It notably covered fragile states and focused on strengthening governance, including by enhancing transparency and accountability, as well as by combating corruption. A recent report found that this translated in improved revenue outcomes, as well as significantly improved expenditure and accountability (PEFA) scores. IMF, 2017. Thus, following TA, Kosovo and Timor-Leste improved along a wide set of indicators, while Liberia and Mali showed improvement in targeted areas.

34

Recent examples concerning PFM support include Mongolia and El Salvador, where the Fund helped improve transparency by developing and publishing medium-term fiscal frameworks, including information on fiscal risks. Also, the Baltic states, the Czech Republic and Poland have significantly improved their corruption perception scores by, among other things, strengthening their public PFM systems.

35

Fiscal Transparency Evaluations (FTE) produce useful information. For example, in Albania, the FTE found that over 250 public corporations with liabilities equivalent to 30 percent of GDP were not included in published fiscal reports. FTEs can also help improve country transparency perceptions; the Philippines’ scores have improved since it implemented the FTE recommendations. For a recent update to the Board, see IMF, 2014c. Following the Fund FTE in Guatemala, which was part of a broader anti-corruption engagement, the government started publishing the list of the earmarked funds as well as the net worth, assets, cash available, and amounts paid to the banks to manage these funds. In addition, also in line with IMF recommendations, the Ministry of Finance proposes to introduce a 5-year medium term budget framework. All ministries now prepare user-friendly presentations, available online, on the budget proposed for the next year, including a detailed calendar of its preparation.

36

These comprise accounting and auditing; anti-money laundering and combating the financing of terrorism (AML/CFT); banking supervision; corporate governance; data dissemination; fiscal transparency; insolvency and creditor rights; insurance supervision; monetary and financial policy transparency; financial market infrastructures; crisis resolution and deposit insurance; and securities regulation. The last Executive Board Review of the Standards and Codes Initiative was conducted in 2011 (IMF, 2011). The next review is expected in July 2017.

37

The Fund’s ROSCs help to assess the quality of economic governance institutions but do not provide direct evidence on the extent or nature of corruption.

38

On FSAP, see further information on the IMF website and the 2014 review (IMF, 2014b).

39

Id.

42

Recent Fund advice in these areas has been provided to Banque des Etats de l’Afrique Centrale (BEAC), Burundi, Iraq, Macedonia, Rwanda, Tunisia, and Ukraine.

43

More than 97 percent of Fund member countries participate in the e-GDDS, SDDS, or SDDS Plus. There are 110 participants in the e-GDDS, 66 SDDS subscribers, and 8 SDDS Plus adherents. See the IMF’s website (http://dsbb.imf.org/).

44

IMFC, 2000, at Paragraph 17; IMF, 2001a.

45

Recent examples include Costa Rica, Paraguay, Peru, and Uruguay.

46

For instance, the Fund helped Ghana, Myanmar, Nepal, and Sudan to move off the gray list.

47

The Fund has recommended greater transparency in this area in a wide range of countries, including Belize, Cyprus, and the United States.

48

In Afghanistan, the Fund supported the authorities’ anti-corruption efforts broadly, including in the context of Fund-supported programs—for example, the current ECF arrangement contains structural benchmarks aimed at criminalizing corruption in line with the United Nations Convention Against Corruption (UNCAC). In Mali’s program, audits were required to be conducted and a report published following procurement irregularities. In Mozambique, the Fund has suspended its lending until an audit is completed and published regarding the circumstances that led to the contracting, without the required parliamentary approval, of previously undisclosed loans.

49

The survey of Board documents covered staff reports and press releases for Article IV consultations, use of Fund resources, Policy Support Instruments (PSIs), and staff-monitored programs.

50

These comparative indicators are developed by external organizations. While each offers insights on the extent of corruption, they also have inherent limitations and do not by themselves, or collectively. represent the Fund’s assessment of corruption in any country. For further information on the nature and shortcomings of indicators, see Supplementary Paper, Background Note III.

51

Only 109 countries had data available at least for one year over the period 2005–16 for each of the three indicators. 37 of the 40 were selected from this group of 109. Three other countries were selected to achieve income and regional balance, even though data for these countries was available for only two of the indicators.

52

The relatively larger size of the fourth quartile group of countries reflects the selection sample, which was over-representative of countries with the highest corruption indicators.

53

The one country that had corruption measures in each of the first three quartiles was given a second quartile position based on the “average” ranking.

54

The countries in the first quartile are France, Japan, Korea, Portugal, Spain, and the United States.

55

Some staff reports for Cambodia, DRC, Nigeria, Uganda, and Ukraine, for instance, had an in-depth analysis of macro-economic impact. Another good example was Mali in Q3.

56

Thus, engagement appeared to pick up after a change of government (Libya, Kyrgyz Republic), launch of government anti-corruption program (Nigeria), unrest over corruption issues (Iraq, Kyrgyz Republic, Nigeria, Ukraine), and major corruption scandals (Afghanistan, Uganda).

57

This echoes the 2004 surveillance review, which called for more attention to policy recommendations on governance, noting that references to governance were of a general nature, describing actions or intentions, without providing staff endorsement, an assessment of adequacy, or views on possible additional measures.

58

An even larger proportion of reports discussed aspects of economic governance that can help reduce risks of corruption (e.g., PFM, tax administration, fiscal reporting), but these were not considered to be indirect references to corruption except where there was evidence of specific corruption-related concerns.

59

See Supplementary Paper, Background Note III for a discussion of the range of available third-party indicators on corruption and their use in Fund reports. In addition to indicators of corruption, a large number of reports cited indicators of competitiveness and the general business climate.

60

The countries in the second quartile are Brazil, Georgia, Greece, Italy, South Africa, Tunisia, and Turkey.

61

The countries in the third quartile are Bolivia, China, Egypt, Indonesia, Mali, Mexico, Mongolia, Papua New Guinea, and Peru.

62

Assessment here is complicated by the fact that staff reports had much more discussion of broader governance issues than in Q2 countries, which may at least in part have been intended to address corruption issues.

63

The fourth quartile is comprised of Afghanistan, Bangladesh, Cambodia, Cameroon, Democratic Republic of Congo, Haiti, Honduras, Iraq, Kazakhstan, Kenya, Kyrgyz Republic, Libya, Nigeria, Paraguay, Russia, Uganda, Ukraine, and Zimbabwe.

64

The relationship between growth and coverage of corruption issues is not consistently a close one, however. For example, income growth slowed in South Africa from around 2009 but corruption issues were not addressed in depth in Fund staff reports until a number of years later.

64

Indicators attempting to provide generalized measurements may be inadequate to capture corruption issues in a particular sector, for example, and could be supplemented by sector-specific indicators or administrative data.

66

It should be noted, however, that, given the stocktaking team’s focus on corruption, some language that actually referenced general governance concepts—such as increasing transparency or improving efficiency—could have been misinterpreted as an indirect way to talk about corruption.

67

It should be noted that the supplemental exercise regarding the overall trends in explicit references to the word “corruption” did not form part of the main qualitative review of staff reports.

68

See IMF, 2016a. A mechanism was also established for reporting and assessing specific concerns by country authorities and for ensuring that lessons or policy changes emerging from evenhandedness assessments are disseminated to Fund staff to guide future work practices.

1

The study therefore recommended focusing remedial PFM conditionality on a smaller number of outcome-oriented structural benchmarks or prior actions. Given that these outcomes take longer to deliver, conditionality may need to be set in a medium-term timeframe with interim milestones.

1

For example, relative to the Fund membership, the country responses over-represent advanced economies and under-represent low-income developing countries.

2

A questionnaire was sent to the ADB, AfDB, EBRD, IADB, FATF, OECD, UNDP, UNODC and World Bank. Responses were received from ADB, AfDB, EBRD, IADB, FATF, OECD, UNODC and World Bank.

3

About half of mission chiefs view indicators as too blunt to be useful in formulating country-specific advice. About 45 percent of respondents noted that indicators were based on limited respondents. A significant number of mission chiefs also saw corruption indicators as subject to potential methodological and data sourcing problems (36 and 25 percent of respondents, respectively).

4

70 percent of mission chiefs report internal consultations on developing Article IV consultation advice and conditionality, with LEG, FAD, and SPR being the lead contact points (in descending order of importance). Less frequently, staff consult with MCM, FIN, and STA.

1

In June 2011, the Executive Board discussed a report reviewing the evolution of the Fund’s AML/CFT program, and endorsed the inclusion of financial integrity issues in surveillance on a mandatory basis in specific circumstances. The 2012 guidance note (IMF, 2012) provides a framework to deal with cases where money laundering, terrorism financing, and related crimes (including corruption) are so serious as to threaten domestic stability, balance of payments stability, or the international monetary system—in the case of Article IV surveillance, or the stability of the domestic financial system—in the case of Financial Stability Assessments (FSAs). LEG staff visited 9-10 countries annually during 2014-2016 to cover financial integrity issues in Article IV and UFR missions.

2

EPAs, recently replaced by Ex-Post Peer Reviewed Assessments, reviewed performance under Fund-supported programs for countries with longer-term program engagement. EPEs review performance under programs supported by exceptional access in the GRA within a year after the end of the arrangement.

3

See Cameroon EPA, Uganda EPA, Mongolia EPA, Kenya EPA. Georgia EPA, Ukraine EPE, Afghanistan EPA, and Ukraine EPE.

1

See IMF, 2016a, “Evenhandedness of Fund Surveillance—Principles and Mechanism for Addressing Concerns.”

2

Includes Article IV consultation staff reports and selected issues paper reports, UFR, and SMP reports; excludes press notices, public information notices, and short supplementary reports.

3

In the case of Georgia, the coverage includes backward-looking discussion of the country’s good progress in improving its corruption ranking.

1

An explanation of any qualifiers used in summings up can be found here: http://www.imf.org/external/np/sec/misc/qualifiers.htm.

Same Series

Other IMF Content

Other Publishers

Cover Policy Papers
The Role of the Fund in Governance Issues - Review of the Guidance Note - Preliminary Considerations
Author: International Monetary Fund
Volume/Issue:
Volume 2017: Issue 004
Publisher:
International Monetary Fund
ISBN:
9781498346528
ISSN:
2663-3493
Pages:
76
DOI:
https://doi.org/10.5089/9781498346528.007