## pp030918govpaper ## Source details **Canonical URL:** [pp030918govpaper](https://www.imf.org/-/media/files/publications/pp/2018/pp030918govpaper.pdf) ## Other formats - [Markdown version](/-/media/files/publications/pp/2018/pp030918govpaper.pdf.md) - [Structured JSON version](/-/media/files/publications/pp/2018/pp030918govpaper.pdf.json) --- ### EXECUTIVE SUMMARY — Rationale and objectives - Proposes adoption of a Framework for Enhanced Fund Engagement to supplement the 1997 Guidance Note on the Role of the Fund in Governance Issues (the “1997 Governance Policy”). - Objectives of the Framework: - promote more systematic, effective, and candid engagement with members on governance vulnerabilities—including corruption—judged to be macroeconomically critical; - apply systematically to all members to enhance evenhandedness; and - strengthen the global fight against corruption by promoting governmental measures that prevent private actors from offering bribes or providing services that enable proceeds of corrupt acts to be concealed, particularly in the transnational context. - Proposed review: experience with application of the Framework to be reviewed within three years. ### Four elements of the Framework for Enhanced Fund Engagement - Element (A) — Systematic assessment of nature and severity of governance vulnerabilities across state functions most relevant to economic activity: - (i) fiscal governance; - (ii) financial sector oversight; - (iii) central bank governance and operations; - (iv) market regulation; - (v) rule of law; and - (vi) Anti-Money Laundering and Combatting the Financing of Terrorism (AML/CFT). - Element (B) — Assessment of economic impact of identified governance vulnerabilities, taking into account standards for: - surveillance under the Integrated Surveillance Decision (ISD); and - use of Fund resources under the Guidelines on Conditionality. - For surveillance: determination of relevance will be based on assessment of severity, irrespective of evidence of short-term impact. - For program conditionality: assess whether addressing vulnerabilities is of critical importance for achieving program goals. - Element (C) — Guidance for policy recommendations: - Policy advice informed by diagnosis, candid, and discussed with authorities whose views should be reflected in staff reports. - Rely on other institutions’ expertise where areas fall outside Fund competence, particularly the World Bank. - Where corruption is assessed as severe, rely on a multi-pronged strategy: specific anti-corruption measures plus broader regulatory and institutional reforms. - Use diagnoses to prioritize Fund capacity development support. - Element (D) — Assessment of governmental measures to prevent private-sector facilitation of corruption, including transnational facilitation: - Urge members to volunteer to have legal and institutional frameworks assessed in bilateral surveillance to determine whether: - (a) they criminalize and prosecute the bribery of foreign public officials; and - (b) they have an effective AML/CFT system designed to prevent foreign officials from concealing proceeds of corruption. ### Principles and scope of engagement - Engagement continues in bilateral surveillance and in use of Fund resources when governance and corruption issues are sufficiently severe to be relevant. - Macro-criticality standards derive from existing policies: - Surveillance: governance weaknesses should be addressed when they are sufficiently severe to “significantly influence present or prospective balance of payments and domestic stability” as called for by the ISD. - Use of Fund resources: measures will be conditions under Fund-supported programs if they are “of critical importance for achieving the goals of the member’s program” per the Guidelines on Conditionality. - Consider reputational risk where the Fund does not address severe corruption. - The Framework will use third-party indicators consistent with the Board-approved policy on the Use of Third-Party Indicators (approved November 10, 2017). ### Assessing the nature and severity of governance vulnerabilities (Framework element A) - State functions and issues to be assessed (subject to information availability): - (a) Fiscal Governance — revenue administration, public financial management, fiscal transparency. - (b) Financial Sector Oversight — capacity and autonomy of supervisory agency; regulatory capture risks; prevention of connected lending. - (c) Central Bank Governance and Operations — mandate, decision-making structure, autonomy, accountability, transparency, internal control environment. - (d) Market Regulation — complexity and opacity creating rent-seeking opportunities; trade regulation discretion. - (e) Rule of Law — protection and enforcement of property and contractual rights; judiciary quality and independence. - (f) AML-CFT — adequacy of legal framework, institutional capacity, and effective implementation. - Information sources and principles: - Rely on Fund instruments where available: Revenue administration, PFM, fiscal transparency work; FSAP and ROSC for financial oversight and AML-CFT; Safeguards Assessments for central bank governance; staff experience for market regulation. - Rely on other institutions (especially the World Bank) where issues fall outside Fund remit. - Use holistic combination of quantitative and qualitative analysis; third-party indicators to complement but not displace Fund staff analysis. - Corruption definition retained: “abuse of public office for private gain” (1997 Guidance Policy). - The Fund will not publish country rankings of corruption or general governance vulnerabilities. ### Indicators and methodology for assessing corruption severity - Use multiple indicator classes (avoid reliance on any single third-party indicator): 1. Perception-based indicators — broad country coverage; reflect investment/reputational effects. 2. Experience-based indicators — surveys of bribery incidence; more objective for bribery but limited in scope. 3. Indicators of anti-corruption institutions’ strength — expert analyses, peer assessments; capture formal features and some implementation judgment. - Recommended measurement strategy: - Holistic approach with multiple indicators; be transparent about what each indicator measures; understand source and methodology; apply judgement for overall view. - Balance caution and attention: - Definitive criminal evidence often elusive; perceptions alone can materially affect economic performance. - Ignoring pervasive perceptions poses reputational risk for the Fund. ### Empirical evidence: corruption, governance, and growth (meta-analysis and panel/cross-section results) - Meta-analysis coverage: - 26 studies on corruption → 149 estimates. - 17 studies on governance → 125 estimates. - Corruption one-quartile deterioration (synthesized estimates): - Overall mean effect: -0.438 - Confidence interval: -0.388 to -0.487 - Distribution bulk: between -0.3 and -1.1 percentage points - Roughly 90 percent of the estimates are negative; 80 percent of those negative estimates are statistically significant. - Panel regression (staff estimates, Annex 4 summary): - A deterioration in the control of corruption index from the 50th percentile to the 25th percentile is associated with a contemporaneous fall in GDP per capita growth in the range of 0.8–1.4 percentage points. - Worsening corruption (50th to 25th percentile) associated with declines in: - Investment: 1.6 percentage points. - FDI: 1.7 percentage points. - Tax-to-GDP ratio: 0.3 percentage points. - Meta-analysis and cross-section synthesized results: - A worsening in the corruption (governance) indicator from the 50th percentile to the 25th percentile is associated with a fall in long-run per capita growth of 0.3 (0.5) percentage point. - Annex 3 panel results (selected): - Macro effects: worsening from 50th to 25th percentile → GDP per capita growth losses ranging from -0.7 to -2 percentage points in some specifications. - Sample sizes reported: N: 3406; 3140; 3406; 3112; 485; 303; 485; 303. Num. Countries: 174; 159; 174; 159; 101; 159; 101; 73. - Meta-analysis subsample point estimates (selected): - AM; EM: -0.336 (standard error 0.121; lower limit -0.573; upper limit -0.099; Z-Value -2.780; p-Value 0.005). - LA: -0.585 (standard error 0.130; lower limit -0.840; upper limit -0.331; Z-Value -4.502; p-Value 0.000). - LICs: -0.879 (standard error 0.100; lower limit -1.075; upper limit -0.682; Z-Value -8.770; p-Value 0.000). - SSA: -0.733 (standard error 0.070; lower limit -0.870; upper limit -0.596; Z-Value -10.469; p-Value 0.000). - Overall: -0.407 (standard error 0.022; lower limit -0.450; upper limit -0.364; Z-Value -18.581; p-Value 0.000). ### General equilibrium analysis: Extended DIG model simulations (Annex 4) - Model features and governance channels incorporated: 1. Waste of output due to weak governance (κt ∈ (0,1)) — reduces effective output and investment incentives. 2. Public investment efficiency (εt ∈ (0,1)) — weak governance reduces public capital accrual. 3. Revenue leakages (ϖt ∈ (0,1)) — reduce effective tax collection and worsen debt dynamics. - Calibration: - Representative EMDE with weak governance. - Tax collection efficiency and public investment efficiency set at the 1st quartile of EMDE distributions. - Reform scenario: country improves governance from the 1st to the 2nd quartile over 10 years. - Specific parameter changes simulated: - Private investment-to-GDP ratio rises from 12 percent to 16 percent. - Public expenditure efficiency rises from 48 percent to 62 percent (εt: 48 percent → 62 percent). - VAT collection efficiency rises from 35 percent to 55 percent. - Simulation results — single-channel reforms: - Reducing output wasted to weak governance: - Real per-capita output rises by 20 percent over 10 years. - Public debt decreases by 4.5 percent of GDP (from an assumed initial level of 30 percent of GDP). - Improving public investment efficiency (εt: 48 percent → 62 percent): - Real per-capita income rises by 3 percent relative to pre-reform era. - Public debt-to-GDP ratio declines by 2 percentage points. - Improving revenue mobilization (VAT collection efficiency 35 percent → 55 percent): - Per-capita output increases by more than 2 percent. - Public debt falls by 9 percent of GDP after 10 years despite lower tax rates. - Simulation results — comprehensive reform (all three channels simultaneously): - Private investment is 70 percent higher in real terms after 10 years. - Output and private consumption increase by about 30 percent. - Public debt falls by almost 15 percent of GDP after 10 years. - Key mechanisms and policy implications: - Reducing distortions that discourage firms directly raises private investment and employment, generating large output gains and improving public debt positions. - Raising public investment efficiency crowds in private demand and modestly improves growth and debt dynamics. - Improving tax administration widens the tax base, permits lower tax rates, and significantly strengthens public debt dynamics while supporting growth. - Comprehensive reform yields substantially larger gains than isolated reforms due to reinforcing general equilibrium effects. ### Use of Fund resources, conditionality, and policy advice - Conditionality standard: - Measures to address governance and corruption weaknesses will be conditions under Fund-supported programs if they are “of critical importance for achieving the goals of the member’s program” per the Guidelines on Conditionality. - Where vulnerabilities are severe, they will be judged “of critical importance” when addressing them is required to reverse the member’s underlying balance-of-payments problem within the time frame needed to repay the Fund. - Repurchase periods referenced: 3¼–5 years for the credit tranches; extended facilities: up to 4½ -10 years for the EFF, and 5½-10 years for the ECF. - Policy advice and strategy: - Advice guided by diagnosis; systematic application of assessment elements expected to increase coverage of governance and corruption issues in surveillance and UFR. - Where corruption is macro-critical: anti-corruption measures plus broader institutional and regulatory reforms, elimination of excessive regulation, promotion of transparency and accountability, capacity-building, and attention to political economy and communication. - The Fund should avoid interfering in individual enforcement cases and domestic or foreign politics, but may condition further access on independent audits where corruption-related allegations create program uncertainty. - Capacity development: - Governance assessments will identify priority areas for Fund technical assistance and capacity development, tailored to surveillance and UFR priorities. ### Addressing transnational facilitation, organized crime, and sectoral focus - Members urged to implement measures preventing private actors from offering bribes or providing services enabling concealment of proceeds of corrupt acts, with emphasis on transnational context. - The Fund urges members to volunteer to have legal and institutional frameworks assessed to determine: - (a) whether they criminalize and prosecute bribery of foreign public officials; and - (b) whether they have an effective AML/CFT system designed to prevent foreign officials from concealing proceeds of corruption. - Organized crime nexus (Box 5): - OCGs use bribery and intimidation to co-opt public officials, launder proceeds, and ensure immunity from law enforcement. - Global revenues of organized crime estimated at above a trillion dollars a year; illicit proceeds (including “proceeds of grand corruption”) often end up in international financial centers. - Extractives sector: - OECD 2014 finding: extractive industry contributes 19 percent of bribery of foreign government officials and 21 percent of the total value of bribed transactions. - The Fund is partner of EITI; supports transparency, technical assistance on tax policy, revenue administration, PFM, statistics, and disclosure of payments by extractive firms where applicable. ### Implementation, institutional arrangements, and review - Management intends a centralized, institutional process to ensure similarly-situated countries are treated similarly. - A standing Working Group on Governance will implement the centralized assessment process, comprising representatives of all Area Departments and relevant Functional Departments. - Assessments will be member-by-member with key input from country desks; Working Group to make final judgements in group reviews to ensure uniformity. - Timing of engagement determined by: - (a) timing of Article IV consultation and program approval/review; and - (b) availability of information needed to apply the Framework effectively. - Resource implications: - Management will assess resource implications in context of the consideration of Administrative Budget for FY 2020. - Interim implementation within existing departmental budgets. - Executive Board review: within three years of adoption to assess effectiveness and uniformity. _italic_ Source: pp030918govpaper (IMF staff paper excerpts). ### EXECUTIVE SUMMARY ### EXECUTIVE SUMMARY ### Rationale and objectives - The paper proposes adoption of a Framework for Enhanced Fund Engagement to supplement the 1997 Guidance Note on the Role of the Fund in Governance Issues (the “1997 Governance Policy”). - While the 1997 Governance Policy remains an appropriate basis, further Board guidance is needed to ensure objectives are achieved given experience over the past 20 years. - Evidence indicates corruption can have a pernicious effect on a country’s ability to achieve sustainable, inclusive economic growth. - The Framework for Enhanced Fund Engagement is designed to: - promote more systematic, effective, and candid engagement with members on governance vulnerabilities— including corruption—judged to be macroeconomically critical; - apply systematically to all members to enhance evenhandedness; and - strengthen the global fight against corruption by promoting governmental measures that prevent private actors from offering bribes or providing services that enable proceeds of corrupt acts to be concealed, particularly in the transnational context. ### Four elements of the Framework for Enhanced Fund Engagement - Element (A): Systematic assessment of the nature and severity of governance vulnerabilities, including corruption, through assessment of state functions most relevant to economic activity: - (i) fiscal governance; - (ii) financial sector oversight; - (iii) central bank governance and operations; - (iv) market regulation; - (v) rule of law; and - (vi) Anti-Money Laundering and Combatting the Financing of Terrorism (AML/CFT). - Element (B): Assessment of the economic impact of identified governance vulnerabilities, taking into account standards for: - surveillance under the Integrated Surveillance Decision (ISD); and - use of Fund resources under the Guidelines on Conditionality. - For surveillance, determination of relevance will be based on assessment of severity, irrespective of evidence of short-term impact. - For program conditionality, the Fund will assess whether addressing vulnerabilities is of critical importance for achieving the goals of the member’s program. - Element (C): Guidance for policy recommendations: - Policy advice will be informed by the diagnosis of vulnerabilities, be candid, and discussed with authorities whose views should be accurately reflected in staff reports. - For areas outside the Fund’s competence, the Fund will rely on expertise of other international institutions, particularly the World Bank. - Where corruption is assessed as severe, the approach will rely on a multi-pronged strategy promoting specific anti-corruption measures and broader regulatory and institutional reforms, taking the member’s circumstances into account. - Diagnoses for surveillance and conditionality will help prioritize the agenda for Fund capacity development support in the country. - Element (D): Assessment of governmental measures to prevent private-sector facilitation of corruption, including transnational facilitation: - The Fund would urge members to volunteer to have their legal and institutional frameworks assessed in bilateral surveillance to determine whether: - (a) they criminalize and prosecute the bribery of foreign public officials; and - (b) they have an effective AML/CFT system designed to prevent foreign officials from concealing the proceeds of corruption. ### Principles and standards guiding engagement - Engagement will continue in bilateral surveillance and in the use of Fund resources when governance and corruption issues are deemed sufficiently severe to be relevant under those activities. - The macro-criticality standards derive from existing policies: - Surveillance: governance weaknesses should be addressed when they are sufficiently severe to “significantly influence present or prospective balance of payments and domestic stability” as called for by the ISD. - Use of Fund resources: measures to address governance and corruption will be conditions under Fund-supported programs if they are “of critical importance for achieving the goals of the member’s program” per the Guidelines on Conditionality. - Consideration should also be given to reputational risk when the Fund does not address corruption in surveillance or use of Fund resources; severe corruption raising reputational concerns is likely to meet the macro-criticality standard. - The Framework sets out principles to guide selection and use of information for assessments, including the use of third-party indicators consistent with the Board-approved policy on the Use of Third-Party Indicators in Fund Reports (approved November 10, 2017). ### Governance scope and approach - The Fund will maintain focus on broader governance issues and institution-building (e.g., public financial management, financial sector assessments, central bank governance, AML/CFT), recognizing these reforms often limit opportunities for corruption even if not targeted specifically at corruption. - Coverage of corruption in Fund engagement has varied; the Framework aims for more systematic and evenhanded coverage across members. - The Fund will seek increased collaboration with the World Bank, other institutions, the private sector, and civil society organizations to leverage expertise on governance issues. - The Framework does not propose publishing country rankings of corruption or general governance vulnerabilities. ### Process, consultations, and review - The approach reflects guidance from Executive Directors during the July 21, 2017 Board discussion and subsequent informal discussions, including an informal session on February 23, 2018 and bilateral/group discussions. - The paper draws on the Board-approved policy on Use of Third-Party Indicators (November 10, 2017). - It is proposed that experience with application of the Framework for Enhanced Fund Engagement be reviewed within three years. *Prepared and approved by IMF staff and management as described in the Executive Summary.* ### 9. Second, when they become relevant for surveillance and use of Fund resources, the ### pp030918govpaper - 9. Second, when they become relevant for surveillance and use of Fund resources, the ### Governance assessment: scope and rationale - The Fund’s assessment of governance vulnerabilities should not be limited only to those arising from corruption; the 1997 policy considered corruption within the broader rubric of governance issues (IMF, 1997a). - Reasons for addressing broader governance problems: - Other governance weaknesses, including those arising from lack of capacity or from inefficiency, can be macro-critical even when there is no evidence of corruption.10 - Effective anti-corruption strategies usually require broader governance reforms to curb incentives and opportunities for corruption.11 - Assessing broader governance weaknesses can bolster the assessment of the severity of corruption, which is often hidden (Section III discussion). - To retain focus, the Framework will select governance issues taking into account governance weaknesses that are typically associated with corruption; addressing such weaknesses may be macro-critical even if they do not, in fact, evidence corruption.13 ### Reliance on other institutions and cooperation - The Fund will need to rely on the work and comparative advantage of other institutions (especially the World Bank) where governance areas fall outside Fund expertise (IMF, 1997a, para. 6). - A comprehensive, systematic approach to assessing and advising on governance issues will require closer co-operation with partner institutions. ### Definition of corruption (Box 1) - The 1997 Guidance Policy definition: “abuse of public office for private gain” (aligned with World Bank usage). The policy does not intend to change this definition. - UNCAC-related mandatory criminal acts generally recognized as constituting corruption (UNCAC, 183 state parties have ratified): - (i) the offer of a bribe to a public official (Article 15(a)); - (ii) the acceptance of a bribe by a public official (Article 15(b)); - (iii) the offering of a bribe to a foreign public official or an official of a public international organization (Article 16(1)); - (iv) the embezzlement, misappropriation or other diversion of funds by a public official (Article 17); - (v) the laundering of the proceeds of crime, including the proceeds of corruption (Article 23); - (vi) obstruction of justice in relation to the prosecution of crimes covered by UNCAC (Article 25). - Acts arising from governance vulnerabilities that are not criminal may still have significant detrimental economic impact and fall within the policy’s scope; actions exclusively involving the private sector (e.g., fraud) are not covered unless they involve transactions with a public official abusing office. - UNCAC 2003 also identifies “non-mandatory” acts that signatories should “consider” criminalizing (e.g., bribery in the private sector and illicit enrichment). ### Addressing transnational facilitation by private actors - An effective strategy requires action to curb facilitation of corrupt practices by private actors, particularly in the transnational context. - Engagement with other countries is important to ensure they have effective systems to prosecute bribery of foreign officials and prevent concealment of proceeds of corruption. ### Non-interference in politics and in individual enforcement cases - The Fund should continue to avoid interfering in domestic or foreign politics of a member or expressing views on the design of particular political systems; the Fund must nevertheless take a view on whether a member can formulate and implement appropriate policies.15,16 - The application of non-interference requires judgment; conduct such as “state capture,” kleptocracy (embezzlement, extortion, intimidation by senior officials), or regulatory capture can be subject to Fund scrutiny when they create macro-critical governance vulnerabilities. - The Fund’s role is not to act as an investigative agency and it should avoid interference in individual enforcement cases (IMF, 1997a, para. 8). Where corruption-related allegations create program uncertainty, the Fund may condition further access on an independent audit and making results available to relevant national authorities in accordance with domestic law. ### Proposed Framework for Enhanced Fund Engagement (Section III) - The Framework consists of four elements: - Element 1: Enable systematic assessment of the nature and severity of governance weaknesses—including corruption—across all members. - Element 2: Guide assessment of the economic impact of identified governance weaknesses. - Element 3: Guide Fund policy advice where engagement is justified by Elements 1 and 2. - Element 4: Assess governmental measures to prevent private actors from offering bribes to foreign officials or enabling concealment of proceeds. - The Framework will be developed further in light of implementation and broader country coverage. ### Assessing the nature and severity of governance vulnerabilities (A) - Relevant categories of state functions to be included in the proposed framework (taking into account Executive Board inputs from July 2017 and the 2016 Staff Discussion Note “Corruption: Costs and Mitigating Strategies”18): - (a) fiscal governance; - (b) financial sector oversight; - (c) central bank governance and operations; - (d) quality of market regulation; - (e) rule of law; - (f) AML-CFT. ### Identifying the relevant issues to be assessed - The assessment will examine legal, policy and institutional frameworks and the effectiveness of implementation, with selection of issues taking into account vulnerabilities typically associated with corruption. ### Fiscal governance (detailed considerations) - Weaknesses in fiscal governance can have harmful effects in three areas: - Revenue side: weak tax administration practices (including those arising from corruption) can weaken revenue collection, affecting deficits, debt accumulation, and service delivery. - Expenditure side: weaknesses in public financial management (including procurement) can generate inefficiencies and inflate costs; corruption can distort expenditure choices toward wasteful items that generate kickbacks. - Fiscal transparency: insufficient transparency undermines accountability and provides opportunities for misappropriation. - Systematic assessment of these areas can inform dialogue on policy measures to reduce vulnerabilities to corruption (see Box 2). ### Financial sector oversight - Weak regulation and supervision can lead to financial instability with significant macro-economic implications. - Assessment elements include autonomy of the supervisory agency and risks of regulatory capture. - Specific scrutiny would include whether the regulatory framework prevents connected lending involving public officials. ### Central bank governance and operations - Even central banks can exhibit severe governance weaknesses that, in Fund-supported programs, can result in inappropriate use of Fund resources. - Assessment issues include adequacy of: - (a) mandate, decision-making structure and autonomy of the central bank; - (b) accountability and transparency framework; - (c) internal control environment. ### Fiscal governance evidence and empirical associations (Box 2) - The Fund’s fiscal governance work (transparency, public spending efficiency, revenue administration) reduces vulnerabilities to corruption by curbing opportunities for leakage. - Cross-country empirical analysis: several indicators of good fiscal governance are significantly associated with improved control of corruption (controlling for GDP per capita). - Reported correlations: - Coefficient of correlation is 0.44 (significant at the 1 percent level) for Fiscal Transparencey. - Coefficient of correlation is 0.25 for Public Procurement. - These associations corroborate previous studies linking practices in public financial management, procurement, audit agencies, and tax administration to reduced scope for corruption (Johnson and others, 2012; World Bank, 2012; DFID, 2015; OECD, 2016). *Source: pp030918govpaper (excerpt).* ### 21. Market Regulation—A key function of the state is to regulate the overall market in ways ### 21. Market Regulation—A key function of the state is to regulate the overall market in ways that promote private investment opportunities and minimize opportunities for fraud, abuse and rent extraction. ### Market regulation: vulnerabilities and effects - Regulatory and institutional frameworks should balance encouraging private sector involvement with promoting the public good through targeted interventions that address market failures and misaligned incentives, and foster competition. - Complex, inaccessible, or opaque regulatory frameworks can create opportunities for exploitation of discretionary authority in return for inducement (e.g., in granting licenses and business permits, or in exemption from existing regulations), resulting in uncertainty and discriminatory treatment. - Undue discretion in the regulation of international trade can be a source for rent-seeking behavior. - Identifying and addressing vulnerabilities in market regulation is crucial to achieving business facilitation while avoiding broad-based corruption that undermines strong and sustainable economic growth. - The proposed market regulation category would examine how regulatory complexity and opacity, including in trade, create significant incentives for corruption of public officials and hinder private business. ### Rule of law: protection and enforcement of economic rights - Focus: aspects of the rule of law relating to protection of property and contractual rights, including predictability and timeliness of enforcement. - Weaknesses can undermine availability of credit, financial inclusion, investment, and the authorities’ ability to restructure debt and regain market access during financial crises. - The most important determinant of enforcement of economic rights is the quality of the judiciary—technical capacity and independence from private influence and public interference. ### Other legal enforcement areas and AML/CFT - High levels of crime can engender corruption and devastate governance and the economy; enforcement of the anti-money laundering framework is particularly relevant. - AML/CFT evaluation: - Weaknesses can: (i) undermine stability of the financial system or broader economy; and (ii) facilitate corruption, including at the transnational level, by allowing concealment of proceeds of corrupt acts. - Assessment focuses on adequacy of the legal framework, overall institutional capacity, and effective implementation. ### Identifying relevant sources of information for governance assessments - Principle: rely, where possible and relevant, on information already obtained by the Fund from member authorities in existing activities. - Fiscal area: systematic assessments of quality of revenue administration, public financial management, and fiscal transparency. - Financial sector oversight and AML-CFT: Financial Sector Assessment Program (FSAP) and Report on Standards and Codes (ROSC). - Central bank governance and operations: Safeguards Assessments. - Market regulation: staff experience and ongoing macro-structural focus in surveillance. - These Fund instruments should be evaluated for relevance to governance and corruption concerns and periodically reviewed. - Where issues are not typically covered by the Fund, staff would rely on information from other institutions, especially the World Bank. - World Bank assessments: procurement systems, overall regulatory frameworks in Country Partnership Frameworks and Systematic Country Diagnostics. - Rule of law: rely on World Bank and other institutions. - Assessments would be holistic, combining quantitative and qualitative analysis, including indicators. - Some indicators are being developed by Fund staff (e.g., tax and expenditure efficiency). - Third-party indicators will complement—not displace—Fund staff analysis and other international organizations’ analysis, and any use will be consistent with the recently-approved Board policy in this area. - Acknowledge inevitable information gaps due to lack of universal coverage; these gaps should reduce over time as engagement increases. In the interim, regular engagement in surveillance and use of Fund resources will help bridge gaps through informed judgment. ### Assessing the severity of corruption: principles and indicators - Corruption should be addressed specifically and forthrightly when judged sufficiently severe; challenge arises because corruption is mostly hidden. - Widely accepted definition focuses on criminal activity (bribery, embezzlement, abuse of function or money laundering). - In systemic cases, conclusive evidence of criminality may not exist because institutions charged with investigation may themselves be corrupt or under-resourced. - Approach to assess severity: - Do not rely on a single third-party indicator; use a range of indicators. - Use qualitative assessments by other international organizations, especially the World Bank. - Assess the quality of the legal and institutional framework charged with combatting corruption. - Use information provided by authorities (including national surveys and other stakeholders) and staff judgment. - Use Fund assessments of governance weaknesses to test reliability of preliminary findings on corruption. - The Fund will not publish country rankings of its assessments of corruption or other governance vulnerabilities. ### Corruption indicators (Box 3): classes, strengths, and recommended approach - Three broad classes of indirect indicators: 1. Perception-based indicators: - Based on perceptions of private sector executives, economic experts, or ordinary citizens. - Available for a very large number of countries. - Defenders argue perceptions correlate with more “objective” measures and matter for investment decisions, tax compliance, capital flight, and brain drain. - Composite indices help minimize biases of single indicators. 2. Experience-based indicators: - Surveys asking respondents if they were solicited for or paid a bribe, typically in the last 12 months. - More “objective” than perceptions-based indicators but limited to bribery and subject to candidness concerns. 3. Indicators of strength of anti-corruption institutions: - Expert analyses of legal, regulatory, and institutional frameworks; peer country compliance assessments with anti-corruption treaties. - Measure observable phenomena but may include judgmental components and may capture formal features rather than implementation. - Recommended measurement strategy: - Use a holistic approach with multiple indicators to smooth biases. - Be transparent and clear about what each indicator measures. - Understand source and methodology of indicators. - Exercise judgement in coming to an overall view. ### Balancing caution and attention in corruption assessment - Caution needed because definitive evidence is often elusive; early anti-corruption efforts may raise perceptions of corruption even as progress is being made. - A pervasive perception of corruption alone can materially affect economic performance (e.g., investment levels). - Ignoring perceptions of systemic corruption because of lack of definitive criminal evidence poses reputational risk for the Fund—especially where perceptions extend to prosecution and conviction institutions. ### Assessing economic impact of governance vulnerabilities - Under the ISD, the Fund focuses on member policies that can significantly influence balance of payments and domestic stability; time horizon traditionally up to five years but includes longer-term structural risks. - Empirical evidence linking governance and growth: - A meta-analysis of 149 estimates suggests higher corruption is typically correlated with lower growth: - A worsening in the corruption indicator from the 50th percentile to the 25th percentile is associated with a decline in GDP per capita growth between 0.4 and 0.5 percentage point. - A similar analysis on 125 estimates shows a negative relation between quality of governance and growth, stronger for regulatory quality or rule of law. - Staff panel regression estimates (Annex 4): - A deterioration in the control of corruption index from the 50th percentile to the 25th percentile is associated with a contemporaneous fall in GDP per capita growth in the range of 0.8–1.4 percentage points. - Worsening corruption (50th to 25th percentile) is associated with declines in investment, FDI, and tax to GDP ratios of 1.6, 1.7 and 0.3 percentage points, respectively. - Results are robust to different measures of corruption. - Cross-section analysis (Annex 4): - A deterioration in the corruption (governance) indicator from the 50th percentile to the 25th percentile is associated with a drop in long-run per capita growth between 0.1 (0.2) and 0.5 (0.7) percentage point. - Deterioration in corruption and governance are also associated with higher inequality (measured by the Gini index) and lower inclusive growth (proxied by the percent change in the Equally Distributed Equivalent Income). - Extended IMF Debt-Investment-Growth model (DIG) simulations (Annex 5): - Captures channels through which weak governance and corruption affect macroeconomic outcomes. - Illustrate how improving governance and reducing corruption could raise the private investment ratio, increase efficiency in taxation and/or public investment, boost real per capita output, create fiscal space and improve debt sustainability. *Source: REVIEW OF THE 1997 GUIDANCE NOTE ON GOVERNANCE—ENHANCED ENGAGEMENT (pp. 16–20).* ### 37. In light of the above, it is proposed that, a determination as to whether a member’s ### pp030918govpaper - 37–48 (Excerpts) ### Determination and timing for addressing governance and corruption vulnerabilities - Determinations of whether a member’s governance and corruption vulnerabilities significantly affect its prospective or present stability within the meaning of the ISD will be based on an assessment of the severity of these vulnerabilities, taking into account the application of the assessment framework described in Section III(A) and the overall circumstances of the member. - Flexibility on timing of inclusion in the Article IV process: - Where vulnerabilities raise immediate concerns—because of their immediate impact on, for example, fiscal sustainability or donor financing—they would need to be discussed at the next Article IV consultation. - At a minimum, governance and corruption vulnerabilities that are sufficiently severe would be substantively discussed within the context of a medium-term surveillance cycle (normally within three years). - Executive Board guidance: discussions need to be clearer and more candid than has previously been the case. ### Use of Fund resources and conditionality - Measures to address governance and corruption weaknesses will be established as conditions under a Fund-supported program if they are “of critical importance for achieving the goals of the member’s program” as provided by the Guidelines on Conditionality. - Where governance and corruption weaknesses are assessed to be severe using the framework in Section III(A), they will be judged “of critical importance for achieving the goals of the member’s program” when addressing those weaknesses is required to reverse the member’s underlying balance-of-payments problem within the time frame needed to repay the Fund for the use of its resources. - Principles and context: - This approach is consistent with applying program conditionality narrowly given the requirement that conditionality be parsimonious. - Programs should be directed to “achieving medium term external viability while fostering sustainable economic growth.” This is generally understood to encompass the repayment period, which is 3¼–5 years for the credit tranches. - Extended facilities recognize longer-term periods to address entrenched structural problems: up to 4½ -10 years for the EFF, and 5½-10 years for the ECF. - Even where the repurchase period is shorter, in balance of payments crises with significant governance problems, recommended measures—e.g., domestic revenue mobilization, PFM and central bank governance—are often already present in existing programs as conditionality. - Recent program experience has shown the importance of including anti-corruption measures in program conditionality in some cases, and of national ownership developed through joint assessments at an early stage of program design. - 1997 Governance Policy precedent: - Financial assistance in the context of completion of a review under a program or approval of a new Fund arrangement could be suspended or delayed on account of poor governance if it could have significant macroeconomic implications or put in doubt the purpose of Fund resources. - In cases of delayed approval or review, measures that at least begin to address these issues should be in place before resumption, including specific conditionality—such as the conduct of an independent audit or passage of specific laws. - Expected impacts of the Enhanced Framework on conditionality: - More evenhanded integration of governance-related reforms into Fund-supported programs. - More specific and candid identification of the nature and severity of problems and proposed reforms. ### Provision of policy advice - Policy advice guided by diagnosis: - Where governance problems are judged sufficiently severe under surveillance or UFR standards, advice is guided by diagnosis of the nature of weaknesses. - Systematic application of assessment elements (Section III(A)) provides a basis for the design of reform recommendations and is expected to increase coverage of governance and corruption issues in surveillance and UFR. - Staff report language should be more direct, clear, and transparent. - Scope of Fund expertise and collaboration: - Many relevant areas fall within Fund expertise: most areas of fiscal governance, financial sector oversight, central bank governance and operations, and AML-CFT. - Market regulation aspects (e.g., complexity and opacity of regulatory frameworks) are typically already addressed in surveillance. - Areas outside Fund expertise will rely on other institutions (e.g., World Bank work on procurement). Some activities are shared (e.g., FSAP module on creditors rights). - The Fund has provided extensive advice in areas judged macro-critical, most recently in Europe. - Anti-corruption strategy elements when corruption is macro-critical: - Strengthening the rule of law—including an effective threat of prosecution—is critical but insufficient alone. - A broader reform agenda is needed: institutional and regulatory reform tailored to diagnosed governance weaknesses; elimination of excessive regulation; promotion of transparency and accountability; capacity-building of institutions. - Strategy should take account of political economy considerations and be supported by effective communication and close engagement with authorities and stakeholders. - The Fund has experience in applying such multipronged approaches in both surveillance and UFR (see Box 4 examples). - Authority interactions and program consequences: - Differences between national authorities’ views and staff views may arise on diagnosis or policy recommendations. - In surveillance, authorities’ views should be solicited and reflected in documentation, including the “authorities’ views” section of staff reports. - In Fund-supported programs, unresolved differences may delay or preclude Fund support where relevant reforms are macro-critical. - Capacity development and technical assistance: - Governance assessments will help identify which governance areas would most benefit from Fund capacity development interventions. - Technical assistance will be prioritized for identified areas as part of a medium-term strategy and designed to support UFR and surveillance priorities. - Example: where corruption is severe and inadequate fiscal transparency exacerbates the problem, priority would be placed on capacity development addressing fiscal transparency inadequacy. - Training and outreach activities will be better tailored to facilitate exchanges of experience between member countries. ### Illustrative Fund engagement examples (Box 4): Ukraine and Mexico - Ukraine (engagement spans Fund-supported programs since 2010 and Article IV consultations): - Anti-corruption specific engagements: - Legislative reforms to strengthen anti-corruption legal framework, asset disclosure by public officials, and anti-money laundering. - Establishment—with active engagement of civil society—of a new, independent anticorruption investigative agency (National Anti-Corruption Bureau of Ukraine) and appointment of an anti-corruption prosecutor; ongoing efforts to establish an independent anti-corruption court. - Broader governance reforms supporting anti-corruption objectives: - Streamlining and simplification of the regulatory framework, including permits and administrative procedures, and inspections of businesses. - Strengthening public procurement law to reduce exemptions, introduce e-procurement, and increase transparency. - Reforming tax administration: stepping up revenue collection by breaking tax evasion schemes, strengthening the Large Tax-Payer Office and the value-added tax (VAT) refund process, integrating tax and customs compliance, enhancing dispute resolution systems and accountability frameworks for leadership, and introducing automated electronic features. - Reforming the state-owned enterprise (SOE) sector, especially in the gas industry, to remove low and differential tariffs that create opportunities for arbitrage and corruption—enhancing transparency and fairness in privatization/restructuring of SOEs. - Reforming central bank legislation to strengthen governance arrangements and decision-making structures, and strengthening financial sector oversight through asset quality reviews, tightening related party regulations, and recapitalization or resolution of insolvent banks. - Judicial reform efforts: new law on the judiciary adopted in 2016, initiating judicial reform starting with the Supreme Court; actions to strengthen management of the judiciary (including disciplinary oversight) and to speed up debt-claim enforcement. - Mexico (engagement spans Flexible Credit Line Arrangements since 2009 and annual Article IV consultations): - Anti-corruption specific engagements: - Creating and empowering the new National Anticorruption System with support and involvement of civil society to improve coordination among relevant authorities in prevention, detection, investigation, prosecution, and penalization of acts of corruption. - Introducing financial disclosure requirements, identifying conflicts of interest, creating a specialized federal court for corruption cases, and stricter penalties for corruption offenses. - Strengthening the anti-money laundering framework to provide further tools for anti-corruption efforts (e.g., enhancing transparency of beneficial ownership and updating due diligence on politically-exposed persons). - Broader governance reforms: - Strengthening the tax compliance gap, further strengthening the fiscal framework, and improving the efficiency of public spending. - Structural reforms in the energy and telecommunications sectors, including improved SOE governance and efficiency. ### Combating factors that facilitate corruption - Two facilitating factors the Fund must address: - The “supply” side: extent to which systemic corruption involves extensive bribery of public officials by private actors exploiting lax regulatory environments; mitigation requires reducing incentives for private actors to engage in bribery. - Weaknesses in the AML/CFT framework that enable hiding proceeds of corrupt practices (e.g., availability of non-transparent corporate vehicles and entities provided by lawyers, accountants, and company service providers). - Bribery-specific considerations: - Domestic level: organized crime can be a major source of bribery and is empirically linked with systemic corruption, particularly at local government level; organized crime can adversely affect inclusive economic growth. The fight against organized crime requires a broad strategy, many elements outside the Fund’s expertise; the Fund plays an active role in strengthening AML/CFT frameworks to deny criminals the ability to enjoy proceeds of crimes. - International level: frameworks required to deter private actors in one jurisdiction from bribing public officials in another jurisdiction. - The OECD Convention on Combating Bribery of Foreign Public Officials provides an effective basis for signatories; it criminalizes bribery of foreign public officials in international business transactions and identifies related measures to make it effective. - Forty-three Fund members (all OECD members plus eight others) are parties to the Convention and undergo periodic assessments of implementation of their obligations. - The UNCAC also has a framework for assessments in this area. - Extractive industries focus: - A 2014 OECD study found that the extractive industry sector contributes the most to the bribery of foreign government officials—19 percent of all cases—and that bribes paid in the sector constitute 21 percent of the total value of a transaction—the highest of any sector. - The Fund will continue to promote transparency and build capacity for country authorities to better manage wealth from the sector. - The Fund is a partner organization of the Extractive Industry Transparency Initiative (EITI), a global standard promoting open and accountable management of natural resources. - Countries fully implementing the EITI should disclose information on tax payments, licenses, contracts, and production. - The Fund provides extensive technical assistance on managing wealth from extractive industries to improve tax policy, revenue administration, public financial management, statistics, transparency, and accountability. - Several countries enacted legislation to enhance transparency of extractive industries’ companies by requiring disclosure of payments they make to foreign countries. *Source: Excerpts from the IMF document pp030918govpaper - 37–48.* ### Box 5. Nexus Between Organized Crime and Corruption ### Box 5. Nexus Between Organized Crime and Corruption ### Mutual dynamics and mechanisms - Organized crime and corruption typically feed on each other.1 - To further their criminal goals, organized criminal groups (OCGs)2 use bribery and intimidation to co-opt public officials in order to: - ensure continuity of their illegal operations; - maintain their monopoly control over local areas (including local governments); - launder the proceeds of these operations; and - ensure their immunity from law enforcement.3 - Bribery by OCGs often takes a continuous form since some of them enroll public officials on their payroll. - The combination of criminal networks and political patronage can create “a vertical system of corruption that functions from top to bottom in all public institutions.”4 - This self-interested complicity between political actors and organized criminal networks can institutionalize criminal enterprises and activities and undermine the compacts of governance between citizens and the state.5 ### International scope and scale - Organized crime is a major threat at the international level. - OCGs are capitalizing on globalization and the expansion of international trade to broaden their range of criminal activities, including in corrupting foreign officials and money laundering.6 - Their global revenues are estimated at above a trillion dollars a year, of which illicit drugs are a major component.7 - Illegal proceeds including “proceeds of grand corruption end up in international financial centers.”8 - As recognized by the United Nations, “the growing threat of organized crime, with its highly destabilizing and corrupting influence on fundamental social, economic and political institutions, represents a challenge demanding increased and more effective international cooperation.”9 *Source: Box 5. Nexus Between Organized Crime and Corruption.* ### 54. It is proposed that the Executive Board adopt the text set forth in Annex 1, entitled ### Addressing Governance Vulnerabilities—A Framework for Enhanced Fund Engagement ### Overview and Purpose - The Framework distills key elements discussed in the paper and seeks to build on earlier discussions with Executive Directors. - Objective: promote more systematic, effective, candid and evenhanded engagement with member countries regarding governance vulnerabilities, including corruption, judged to be macro-critical. - The Framework is designed to ensure issues are addressed consistently with the 1997 Guidance Policy and the Fund’s overall mandate. ### Implementation and Institutional Arrangements - Management intends to initiate a centralized, institutional process to ensure similarly-situated countries (in terms of their governance vulnerabilities) are treated similarly in surveillance and use of Fund resources. - A standing Working Group on Governance will implement the centralized assessment process, comprising representatives of all Area Departments and the relevant Functional Departments. - Assessments will be member-by-member with key input from the country desk and relevant Functional Department; the Working Group will make final judgements in the context of a review of a group of countries to ensure uniformity. - In surveillance, groups of countries would be determined by the calendar (e.g., all members whose Article IV missions are scheduled in a particular quarter). - Input will be sought from the World Bank where identified vulnerabilities fall within the Bank’s expertise. - Specific timing of engagement with each member will be determined by: - (a) the timing of the relevant Article IV consultation and the program approval/review; and - (b) the availability of the information needed to apply the Framework effectively. ### Timing of Discussion in Surveillance - Once weaknesses are judged sufficiently severe to merit inclusion in surveillance, there is limited flexibility as to timing of discussion. - A determination must be made as to when and how frequently a comprehensive discussion of the issue should take place within a medium–term surveillance strategy for the country (normally within three years). - The timing determination will be based primarily on the urgency of the problem in terms of its impact on balance of payments or domestic stability, plus country-specific circumstances and competing policy issues. ### Information Gaps and Sources - Information gaps are recognized due to lack of universal coverage of several existing exercises. - Over time, these gaps are expected to be reduced as the Fund engages more systemically on these issues in surveillance and the use of Fund resources. - Previous Article IV consultation reports will constitute a key source of information over time. - The Fund will, to the extent possible, rely on information already obtained in the context of existing Fund activities, including discussions with authorities. - In areas outside the Fund’s remit, the Fund should rely on information provided by other institutions, especially the World Bank. - The assessments should be holistic, relying on both quantitative and qualitative inputs; third-party indicators may be used to complement—but not displace—Fund staff analysis and other international organizations’ analysis. - The Fund will not publish country rankings of its assessment of corruption or other general governance vulnerabilities. ### Resources and Review - Management intends to assess resource implications of applying the Framework in the context of the consideration of Administrative Budget for FY 2020. - Until that assessment, departments will implement the Framework within their existing budgets. - The Executive Board will review implementation of the Framework within three years of its adoption (an early review relative to the normal five-year period) to assess whether greater effectiveness and uniformity have been achieved. ### Elements of the Assessment Framework - The Fund will assess severity of governance vulnerabilities systematically, including the extent to which state functions relevant to economic activity are undermined. - Subject to availability of information, state functions and specific issues normally include: - (a) Fiscal Governance — quality of institutional framework and practices supporting revenue administration, public financial management, and fiscal transparency for the public sector. - (b) Financial Sector Oversight — capacity and effective autonomy of supervisory agency; aspects of regulatory and supervisory design relevant to safeguarding integrity and minimizing corruption opportunities. - (c) Central Bank Governance and Operations — adequacy of mandate, decision-making structure and autonomy; adequacy of accountability and transparency framework; effectiveness of internal control environment. - (d) Market Regulation — extent to which complexity and opacity of regulatory environment creates rent-seeking opportunities or hinders private business operation. - (e) Rule of Law — aspects that support protection of property and contractual rights, including predictability and timeliness of enforcement. - (f) Anti-Money Laundering and Countering the Financing of Terrorism — adequacy of legal framework, institutional capacity and effective implementation. - The assessment will explicitly examine the severity of corruption given its particularly pernicious impact on sustainable, inclusive growth. ### Use of Assessments in Surveillance and Conditionality - Bilateral surveillance: - A determination that governance vulnerabilities significantly affect prospective or present balance of payments or domestic stability within the meaning of the Integrated Surveillance Decision will be based on assessment of severity using the Framework. - If sufficiently severe, there will be flexibility on timing of inclusion in Article IV; at a minimum, these vulnerabilities would be substantively discussed within the context of a medium-term surveillance cycle (normally within three years). - Use of Fund resources: - To determine whether reforms to address governance vulnerabilities should be conditions for use of Fund resources, the Fund will assess: - (i) the severity of governance vulnerabilities; and - (ii) whether addressing the identified vulnerabilities is of critical importance for achieving the goals of the member’s program. ### Policy Advice, Technical Assistance, and Collaboration - Policy advice will be guided by diagnosis of the nature of weaknesses in each country. - Engagement on corruption requires a multi-pronged strategy: anti-corruption measures plus broad-based regulatory and institutional reforms. - The Fund should collaborate closely with other international organizations when weaknesses fall outside the Fund’s expertise. - The Fund’s technical assistance will be prioritized to focus on areas of governance reform deemed critically important for surveillance and use of Fund resources. - The Fund should continue to: - avoid interfering in individual enforcement cases; - avoid interference in domestic or foreign politics of a member; and - avoid expressing views on the design of a particular political system. - An effective strategy requires action to curb facilitation of corrupt practices by private actors, particularly in the transnational context. *SM/18/55 Annex 1 — Addressing Governance Vulnerabilities—A Framework for Enhanced Fund Engagement* ### 8. In order to address corruption, it is imperative that members implement measures to ### 8. In order to address corruption, it is imperative that members implement measures to ### Policy recommendations and institutional guidance - Members should implement measures to prevent private actors from offering bribes or providing services that enable the proceeds of corrupt acts to be concealed, with particular relevance in the transnational context. - The Fund urges all members to volunteer to have their own legal and institutional frameworks assessed in the context of bilateral surveillance for purposes of determining whether: - (a) they criminalize and prosecute the bribery of foreign public officials; and - (b) they have an effective AML/CFT system that is designed to prevent foreign officials from concealing the proceeds of corruption. - Experience with the application of this Framework will be reviewed within three years. ### Meta-analysis: scope and selection - Meta-analysis covers: - 26 studies focusing on the growth impact of corruption (providing 149 estimates). - 17 studies on the growth impact from weak governance (125 estimates). - Inclusion criteria: - Studies that investigate effects of corruption or governance on growth. - Use of widely-recognized indicators of corruption and governance. - Address issues of endogeneity. - Published in high-quality journals. ### Main empirical findings on corruption and growth - Corruption and weak governance are associated with slower growth. - Distribution of per-capita growth impact estimates from a one-quartile increase along the corruption index: - Bulk distributed between -0.3 and -1.1 percentage points. - Roughly 90 percent of the estimates are negative. - 80 percent of those negative estimates are statistically significant. - Distribution of growth impact estimates from a one-quartile deterioration in the governance index: - Mainly distributed between -0.2 and -1.5 percentage points. - Overall synthesized effect for corruption (one-quartile deterioration): - On average associated with a ½-percentage-point loss in per-capita growth, with somewhat larger loss in the case of low income countries (LICs). - Overall mean effect is statistically significant at -0.438, with the confidence interval of -0.388 to -0.487. - Between-study heterogeneity: - Most observed variance in estimates is due to differences in studies, with little owing to sampling error (I2 close to 100 percent). - Subsample patterns: - Studies based only on LICs data or with regional focus tend to have larger effects than the overall mean. - Studies using mixed-country samples, including only AEs and EMs data, tend to report estimates lower than the overall mean effect. - Direct versus indirect effects: - The direct effects of corruption tend to be larger than the indirect effects coming from either total investment or FDI. - Sensitivity and robustness: - Estimates do not appear to depend significantly on the measure of corruption used (between-study variation due to different indicators: Q-value = 3.5, with 2 df, and p-value = 0.169). - Estimates do not appear to depend on the estimation technique used (Q-value = 1.633, with 3 df, and p-value = 0.651). - No evidence that the choice of instruments used in the estimation technique mattered. - Estimates do not vary significantly when other measures of institutional quality are included. - Robustness checks indicate possible publication bias, but staff’s findings remain robust to this. ### Variation by estimation type - Mean effects by estimation approach for corruption-related studies: - Annual panel data estimation: mean effect = -0.85 - Panels with 5-year averaging: mean effect = -0.23 - Cross-section estimation: mean effect = -0.51 - Between-study variation across these estimation types is statistically significant (Q-statistic of 81.3 with p-value of 0.00). ### Findings on governance and growth - Weaker governance also has a statistically significant effect on growth. - The mean effect across all governance studies is negative. - Estimates vary depending on the governance indicator used: - Freedom House Index tends to yield larger estimates. - Followed by the International Country Risk Guide (ICRG) rule-of-law indicator. - Then Worldwide Governance Indicators. ### Key numeric results and statistics (selected) - Meta-analysis coverage: - 26 studies on corruption → 149 estimates. - 17 studies on governance → 125 estimates. - Corruption one-quartile deterioration: - Overall mean effect: -0.438 - Confidence interval: -0.388 to -0.487 - Distribution bulk: between -0.3 and -1.1 percentage points - Roughly 90 percent negative; 80 percent of negative estimates statistically significant. - Subsample point estimates (Table A2.2 excerpt): - AM; EM: -0.336 point estimate; standard error 0.121; lower limit -0.573; upper limit -0.099; Z-Value -2.780; p-Value 0.005 - LA: -0.585 point estimate; standard error 0.130; lower limit -0.840; upper limit -0.331; Z-Value -4.502; p-Value 0.000 - LICs: -0.879 point estimate; standard error 0.100; lower limit -1.075; upper limit -0.682; Z-Value -8.770; p-Value 0.000 - mixed: -0.335 point estimate; standard error 0.025; lower limit -0.383; upper limit -0.287; Z-Value -13.615; p-Value 0.000 - SSA: -0.733 point estimate; standard error 0.070; lower limit -0.870; upper limit -0.596; Z-Value -10.469; p-Value 0.000 - Overall: -0.407 point estimate; standard error 0.022; lower limit -0.450; upper limit -0.364; Z-Value -18.581; p-Value 0.000 - Table A2.4 selected entries (impact of governance on growth; dependent variable rows): - GDP Growth: No. of Estimates 6; Point estimate 1.545; Standard error 0.262; Variance 0.069; Lower limit 1.031; Upper limit 2.059; Z-value 5.889; P-value 0.000 - GDP per Capita Growth: No. of Estimates 31; Point estimate 0.992; Standard error 0.085; Variance 0.007; Lower limit 0.826; Upper limit 1.158; Z-value 11.697; P-value 0.000 - Real GDP Growth: No. of Estimates 36; Point estimate -0.034; Standard error 0.014; Variance 0.000; Lower limit -0.061; Upper limit -0.008; Z-value -2.515; P-value 0.012 - Real GDP per Capita Growth: No. of Estimates 51; Point estimate 0.115; Standard error 0.015; Variance 0.000; Lower limit 0.087; Upper limit 0.144; Z-value 7.855; P-value 0.000 - Overall (governance): No. of Estimates 124; Point estimate 0.050; Standard error 0.010; Variance 0.000; Lower limit 0.030; Upper limit 0.069; Z-value 5.029; P-value 0.000 ### Methodology summary - Random-effect (RE) meta-analysis model used to derive synthesized estimates. - Reported statistics include: - Weighted mean effects (M) using θi, νi, Τ2 and study-specific weights. - I2 statistic as a measure of heterogeneity (I2 = T2/(T2+ν̄θ) * 100). - Standard error of the combined effect, SEM = √VM, and its 95 percent confidence interval, M ± 19.6 SEM. - Z-statistic to test null hypothesis that mean true effect size is zero; p-values computed from the standard normal cumulative distribution Φ(Z). - Between-study variance Τ2 computed following Borenstein and others (2009) formulas using Q, df, and weight sums. *Source: IMF staff estimates and meta-analysis as presented in Annex 2 of the chapter.* ### Annex 3. Corruption/Governance and Inclusive Growth: Empirical ### Annex 3. Corruption/Governance and Inclusive Growth: Empirical Analysis ### Overview - Two econometric approaches are used to analyze the association between corruption (governance quality) and inclusive growth: - Panel regression framework examining short- to medium-term growth and corruption. - Model-averaging cross-sectional analysis assessing robustness of corruption’s association with long-run inclusive growth. ### I. Panel Data Analysis — Model Specification and Methodology - Estimated specification (B1): y_it = α_i + λ_t + β CI_it + ∑(θ_k X_it^k) + ε_it - y_it: annual GDP growth (GDP per capita, investment, FDI or tax revenues). - X_it^k: control variables robustly associated with short-term growth. - α_i and λ_t: country and time fixed effects. - CI_it: indicator of corruption (baseline: WGI-CCI; robustness: Transparency International GCB). - Estimation approach: - Generalized method of moments (GMM) panel estimation; results similar using fixed country and time effects. - Panel co-integration analysis (results not reported) suggests worsening corruption is negatively associated with output in short and long term. - Data coverage notes: - WGI-CCI available annually for 1996–2016, ~200 countries. - GCB available for 2003–2016, covers <100 countries in each year. ### I. Panel Data Analysis — Key Results - Macro effects of worsening corruption: - A worsening from the 50th to the 25th percentile in the corruption ranking is associated with GDP per capita growth losses ranging from -0.7 to -2 percentage points. - Associated declines in ratios: - Investment-to-GDP: 1.6 percentage points. - FDI-to-GDP: 1.7 percentage points. - Tax-to-GDP: 0.7 percentage points. - Robustness: - Results are robust to different corruption measures (WGI-CCI and GCB). - Panel estimates based on 5-year or 10-year averages show a smaller negative association that is no longer statistically significant. - Selected coefficient estimates from Table A3.1 / A3.2 (preserve reported magnitudes and significance markers): - WGI-CCI: 0.115**, 0.117**. - Lagged WGI-CCI: 0.0657**, 0.0584*. - WGI-CCI*weak: -0.0242, -0.0131. - Lagged WGI-CCI*weak: 0.0473*, 0.0472**. - GCB: 0.088*, 0.070***. - Lagged GCB: 0.035*, 0.047*. - GCB*weak: 0.065, 0.041. - lagged GCB*weak: -0.037*, -0.0327. - Sample sizes reported in estimation tables: - N: 3406; 3140; 3406; 3112; 485; 303; 485; 303. - Num. Countries: 174; 159; 174; 159; 101; 159; 101; 73. - Additional point estimates and t-statistics reported in tables; significance codes: * p < 0.1, ** p < 0.05, *** p < 0.01. ### II. Cross-Section Analysis — Model Specification and Methodology - Cross-sectional specification (B2): y = X1 β1 + X2 β2 + ε - y: vector of n observations (e.g., average growth rate of GDP per capita for each country over 1995–2015). - X1: focus regressors (always included). - X2: auxiliary regressors (to be tested). - Estimation method: Weighted Average Least Squares (WALS) model-averaging technique. - WALS advantages cited: neutral prior, minimizes reject criterion risks, and computationally practical compared with standard Bayesian model averaging (BMA). - Two model versions: - Model 1: focus variables = proximate determinants of Solow model (initial GDP per capita, investment-to-GDP, education, life expectancy, population growth) + constant. - Model 2: focus variables = constant only (all explanatory variables treated as auxiliary). - Dependent variables analyzed (averaged over 1995–2015, unless noted): - (i) GDP per capita growth. - (ii) Investment-to-GDP ratio. - (iii) Revenue-to-GDP ratio. - (iv) Inequality (Gini coefficient of disposable income). - (v) Growth in Equally Distributed Equivalent Income (EDEI) — proxy of inclusive growth (EDEI average 2003–2015 cited). - Sample size for cross-sectional WALS: 71 countries (determined by auxiliary variable data availability). ### II. Cross-Section Analysis — Key Results - Robust negative association: - Corruption (weak governance) is robustly and negatively associated with long-run per capita growth in WALS results, even when bilateral correlations are weak. - A worsening in the corruption (governance) indicator from the 50th percentile to the 25th percentile is associated with a fall in long-run per capita growth of 0.3 (0.5) percentage point. - Other associations: - Deterioration in corruption/governance quality is associated with: - Lower investment. - Lower fiscal revenue. - Higher inequality. - Worse inclusive growth performance (EDEI). - Magnitudes are described as economically significant. - Robustness checks performed and reported: - (i) Alternative choices of auxiliary regressors. - (ii) Alternative time periods of the dependent variables (1980–2015; 1990–2015). - (iii) Including additional variables as in Sala-i-Martin and others (2004). - (iv) Alternative data sources for growth and investment (World Development Indicators and Penn World Table). - (v) Excluding corruption from the composite measure of governance. - (vi) Ordinary least squares (OLS) and WALS approaches based on a more restricted set of variables (which increases sample size to about 90 observations). ### Variables Used in Cross-Sectional Analysis (selected from Table A3.4) - Outcome and key variables: - EDEI (average 2003-2015) — Gaspar et al. 2017. - Average per capita GDP growth rate (1995-2015) — Penn World Tables v.9. - Average ratio of investment to GDP (1995-2015) — Penn World Tables v.9. - Gini coefficient of disposable income (average 1995-2015) — SWID. - Control of corruption (estimate) — WB-WGI. - Average ratio of government consumption to GDP — WB-WDI. - Focus and auxiliary regressors (sources include Barro-Lee; Sala-i-Martin et al. 2004; Penn World Tables v.9; WB-WDI; SWID), examples: - Log of average population attended/completed secondary school. - Log of initial per capita GDP (1999 for 2000-2004 period). - Log of average population growth. - Average CPI inflation. - Income group (WB-classification). - Regional dummies (East Asia; Sub-Saharan Africa; Latin American and Caribbean). - Measures including: executive constraints; risk of forced nationalization; life expectancy reciprocal; fertility rate; geographic and climate measures; trade measures; legal system dummies; resource indicators (hydrocarbon deposits; oil producing country dummy); political rights; real exchange rate distortions; revolutions and coups; outward orientation; public investment share; and others (see Table A3.4 for full list). _italic_ Source: IMF staff estimates and annex content. ### Annex 4. Improving Governance: General Equilibrium Analysis ### Annex 4. Improving Governance: General Equilibrium Analysis ### The Model - Framework: Extended Debt, Investment, and Growth (DIG) model (Buffie and others (2012)); a real, dynamic, open economy model with public capital as input to production. - Key features: - Firms optimally choose private capital and labor. - Government levies distortionary taxes to finance public investment. - Policymakers respond to changes in the debt/GDP ratio with a lag to return it gradually to steady state: increases (decreases) in the debt ratio are followed by tax rate hikes (cuts). - General equilibrium structure allows simultaneous analysis of output, private investment, consumption, and fiscal outcomes. - Governance channels incorporated: 1. Waste of output due to weak governance (parameter κt ∈ (0,1)): a share of economic output is lost (reduces incentives to invest and hire). Firm profit: Πt = (1−κt)Yt − wtLt − Rkt Kt. Weak governance creates a wedge (1−κt) < 1 between real wages/gross returns and marginal products, lowering employment, investment, and output. 2. Public investment efficiency (εt ∈ (0,1)): law of motion for public capital Kt+1^G = (1−δG)Kt^G + εt It^G. Weak governance implies fraction 1−εt of public investment is wasted, lowering the accumulated stock of public capital and output. 3. Revenue leakages (ϖt ∈ (0,1)): government debt evolution (only expenditure is public investment and only revenue is consumption tax) Bt+1 = Rt Bt + It^G − (1−ϖt) τt^C Ct. Higher ϖt implies loss of tax revenue, worsening debt dynamics or necessitating higher tax rates that curb private demand and output. ### Model Calibration - Representative country: an Emerging Market and Developing Economy (EMDE) with weak governance. - Calibration targets: - Tax collection efficiency and public investment efficiency set at the 1st quartile of EMDE distributions (based on staff-assembled indicators). - Share of output lost to weak governance calibrated to match the private investment-to-GDP ratio at the 1st quartile. - Remaining parameters as in Buffie and others (2012). - Reform scenario assumptions: - Country gradually improves governance from the 1st to the 2nd quartile over 10 years. - Specific parameter changes: - Private investment-to-GDP ratio rises from 12 percent to 16 percent (implied reduction in the share of output lost). - Public expenditure efficiency rises from 48 percent to 62 percent (εt: 48 percent → 62 percent). - Government efficiency in collecting taxes (VAT) rises from 35 percent to 55 percent (ϖt implication). ### Simulation Results — Single-Channel Reforms - Reducing the share of output wasted to weak governance (reducing κt): - Real per-capita output rises by 20 percent over 10 years. - Private consumption increases. - Public debt decreases by 4.5 percent of GDP (from an assumed initial level of 30 percent of GDP). - Improving public investment efficiency (εt: 48 percent → 62 percent): - Stock of public capital increases. - Real per-capita income rises by 3 percent relative to the pre-reform era. - Note: "The reform causes output to decline initially, owing to the large positive income effect on labor supply." - Crowding-in of private demand (consumption and investment). - Public debt-to-GDP ratio declines by 2 percentage points. - Improving revenue mobilization (VAT collection efficiency 35 percent → 55 percent): - Expands tax base, allowing the government to reduce the tax rate. - Private consumption, investment, employment, and per-capita output increase (per-capita output increases by more than 2 percent). - Public debt falls by 9 percent of GDP after 10 years despite lower tax rates (higher tax revenues and higher output). - If the tax rate were kept fixed at the pre-reform level, the decline in public debt would be stronger. ### Simulation Results — Comprehensive Reform (All three channels simultaneously) - Combined effect over 10 years: - Private investment is 70 percent higher in real terms. - Output and private consumption increase by about 30 percent. - Public debt falls by almost 15 percent of GDP after 10 years. - Lower public debt allows the tax rate to fall (per the assumed fiscal reaction function), further stimulating private demand. ### Key Mechanisms and Policy Implications - Reducing distortions that discourage firms (lower κt) directly raises private investment and employment, generating large output gains and improving public debt positions. - Raising public investment efficiency (higher εt) increases the productive stock of public capital, crowds in private demand, and modestly improves growth and debt dynamics. - Improving tax administration reduces revenue leakages, widens the tax base, permits lower tax rates, and significantly strengthens public debt dynamics while supporting growth. - A comprehensive governance reform package yields substantially larger gains than isolated reforms due to reinforcing general equilibrium effects across private demand, public capital accumulation, and fiscal space. *IMF staff calculations based on DIG simulations (Annex 4, "Improving Governance: General Equilibrium Analysis").* --- _Source: https://www.imf.org/-/media/files/publications/pp/2018/pp030918govpaper.pdf_