## _wp0489

## Source details

**Canonical URL:** [_wp0489](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0489.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0489.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0489.pdf.json)

---

### Introduction and purpose
- Constructs two indices and quantifies their relationship:
  - Financial System Soundness Index (FSSI) — proxy for financial system stability (limited to banking system).
  - Regulatory Governance Index (RGI) — quality of regulatory governance (banking-sector supervisors).
- Threefold contribution:
  - Constructs an index for financial system soundness.
  - Builds an index for the quality of regulatory governance.
  - Quantifies the impact of the quality of regulatory governance on banking system soundness.
- Empirical focus limited to the banking system because of wider availability and comparability of cross-country banking data.
- Multivariate cross-sectional analysis indicates the quality of regulatory governance matters for banking system soundness.

### Financial System Soundness Index (FSSI): construction and rationale
- Purpose:
  - Capture resilience and soundness (capacity to withstand events) as an ex ante, continuous measure allowing assessment of degrees of (un)soundness.
- Composition:
  - Two quantitative variables: capital adequacy ratio (CAR) and ratio of nonperforming loans (NPLs).
    - CAR: indicator of resilience to balance-sheet shocks.
    - NPL ratio: signals portfolio quality and solvency risk (NPLs inverted so higher values indicate greater stability).
- Weighting:
  - Index weighted by share of bank credit in GDP to reflect cross-country comparability and the higher potential costs of instability where credit markets are larger.
- Specification (as presented):
  - FSSI = credit/GDP * [1/2 (CAR + NPLs)]
- Limitations:
  - Variables are somewhat backward looking; broader CAMEL-type aggregation desirable but not available with cross-country consistency.
- Sample grouping:
  - Country groups in sample: 11 advanced economies, 17 transition economies, and 27 developing countries.
- Empirical pattern:
  - Financial system soundness is significantly higher in advanced countries than in transition or developing countries (Figure 1 referenced).

### Regulatory Governance Index (RGI): framework and construction
- Conceptual framework:
  - Good regulatory governance comprises four institutional components: Independence, Accountability, Transparency, Integrity.
  - Governance nexus: governance at public sector, regulatory agency, and financial institution layers interact and amplify effects on financial-sector outcomes.
- RGI construction:
  - Based on FSAP assessments of observance of governance-related practices (MFP Transparency Code, Basel Core Principles, IAIS, IOSCO as available).
  - Scoring per practice: 1 = noncompliance; 2 = partial compliance; 3 = broad compliance; 4 = full compliance; 9 = not applicable/not answered.
  - Country SCORE_ij computed as:
    - SCORE_ij = [0*noncompliance_ij + 0,33*partial compliance_ij + 0,66*broad compliance_ij + full compliance_ij]*100
  - RGI_i = 1/n (Σ SCORE_ij) over j = 1..n, excluding not answered/not applicable items.
  - Weights derived from Sundararajan, Das, and Yossifov (2003) methodology.
- Empirical observation:
  - Mean RGI values are higher for advanced countries than for transition and developing countries (Figure 2 referenced).
  - Broader RGI that includes insurance supervision (where data permit) declines relative to banking-only RGI, indicating governance practices are less widespread in insurance.

### Empirical model, estimation approach, and key controls
- Reduced-form model (as analyzed):
  - FSSI_i = ƒ (MACRO_i, STRUCT_i, RGI_i, PSGOV_i) + μ_i
    - MACRO: macroeconomic environment (fiscal position, inflation, short-term real interest rates).
    - STRUCT: structure of the banking system (e.g., bank concentration, government-owned banks).
    - RGI: regulatory governance index.
    - PSGOV: public sector governance measures.
- Sample and data:
  - Sample coverage approximately 50 countries (countries participating in FSAP with complete 2001 data); N varies by specification.
  - Public sector governance variables drawn from ICRG, Freedom House (2001), Kaufman, Kraay and Zoido-Lobaton (2002).
- Estimation choice:
  - White test revealed heteroskedasticity; Weighted Least Squares (WLS) used instead of OLS.
- Bivariate correlation:
  - Correlation between RGI and FSSI = 0,645 (significant at the 1 percent level).

### Main WLS regression findings
- Core finding:
  - Regulatory governance (RGI) is positive and statistically significant in all model specifications tested — better regulatory governance tends to strengthen financial system soundness.
- Associations with financial soundness (as reported):
  - Positive associations: low fiscal deficit (or high surplus), low real interest rate, low inflation.
  - Negative and significant associations: presence of government-owned banks; higher bank concentration.
  - Public sector governance has a direct positive impact on financial system soundness.
  - Interaction effect (RGI * public sector governance) positive and significant — regulatory governance effects amplified when public sector governance quality is higher.
- Estimation details:
  - WLS chosen to address heteroskedastic residuals; sample size example: N = 48 reported in table.
  - R^2 and adjusted R^2 reported across models include values: 0,484***; 0,628***; 0,631***; 0,675***; 0,722***; 0,768***; 0,718 (R^2 examples) and adjusted R^2 examples: 0,473; 0,600; 0,583; 0,622; 0,677; 0,739; 0,672.

### Selected exact WLS coefficient estimates (from Table 1)
- Regulatory governance RGI:
  - (1) 0,696*** (t = 6,570)
  - (2) 0,331** (t = 2,263)
  - (3) 0,291* (t = 1,761)
  - (4) 0,282* (t = 1,796)
  - (5) 0,304** (t = 2,269)
  - (6) 0,157* (t = 1,710)
  - (7) 0,294** (t = 2,151)
- Government-owned banks (selected reported coefficients and t-statistics):
  - -0,283** (-2,446); -0,284** (-2,382); -0,271*** (-2,197); -0,323*** (-3,302); -0,339*** (-3,992); -0,388*** (-3,607)
- Fiscal balance (examples):
  - 0,317** (1,997); 0,290* (1,723); 0,315* (6%) (1,906)
- Concentration (examples):
  - -0,232** (-2,234); -0,258*** (-2,778); -0,214** (-2,668); -0,305*** (-3,228)
- Public sector governance:
  - 0,299*** (3,081)
  - Corruption (component): 0,536*** (4,401)
- Interaction term:
  - RGI * public sector governance: 1,068* (1,558)
- Significance notation:
  - *p<.1. **p<.05. ***p<.01.

### Endogeneity concerns and 2SLS robustness checks (summary of Table 2)
- Concern:
  - RGI may be endogenous (correlated with the error term) because public sector governance and/or financial soundness could influence regulatory governance.
- Instrumental strategy:
  - Two-stage least squares (2SLS) employed, using public sector governance as an instrument in the first stage.
- 2SLS results:
  - Impact of regulatory governance on financial system soundness remains positive and significant; coefficients are slightly larger and more significant than corresponding WLS estimates.
  - Sign and significance of control variables virtually unchanged relative to WLS.
- Example paired WLS versus 2SLS RGI coefficients (from Table 2):
  - WLS 0,696*** (6,570) versus 2SLS 0,699*** (6,703)
  - WLS 0,331** (2,263) versus 2SLS 0,436*** (3,151)
  - WLS 0,291* (1,761) versus 2SLS 0,495*** (4,287)
  - WLS 0,282* (1,796) versus 2SLS 0,396*** (3,466)
- 2SLS fit examples:
  - R^2 up to 0,755***; adjusted R^2 up to 0,715.

### Extension to the broader financial system and insurance supervision
- Motivation:
  - Rapid structural change and increasing mainstreaming of insurance sector motivate extending indicators beyond banking.
  - Risks in corporate and nonbank financial sectors can pose systemic risks; structural linkages between insurance and banking can transmit risk.
- Data and methodological caveats:
  - Nonbank financial soundness work is at an early stage; technical, accounting, and definitional issues affect cross-country consistency.
  - Construction of a broader index covering banking and nonbank sectors (insurance) is premature and carries methodological pitfalls.
- Preliminary broader-RGI finding:
  - Where data permit, a broader RGI that includes insurance supervision declines relative to the banking-only RGI, suggesting governance practices are less widespread in the insurance sector.
- Implications:
  - Regulatory governance practices in insurance need strengthening to limit systemic risk and contagion.
  - National and international efforts required to collect and disclose uniform, consistent nonbank sector data.
  - Further analyses needed to assess the impact of insurance company soundness on whole-system soundness.

### Policy implications and research agenda
- Policy implications (as presented):
  - Strengthen regulatory governance institutions emphasizing Independence, Accountability, Transparency, and Integrity.
  - Design supervisory mandates and organizational structures to monitor system-wide risks in addition to firm-level prudential oversight.
  - Improve overall public sector governance to enhance regulatory governance effectiveness.
  - Strengthen regulatory governance in nonbank sectors (insurance) as systemic importance grows.
- Research agenda and next steps suggested:
  - (i) Develop FSSI for the broader financial system by collecting consistent indicators for nonbank segments (insurance).
  - (ii) Fine-tune the FSSI by including other indicators, such as measures of liquidity and indicators for quality of financial market infrastructure.
  - (iii) Develop time series to assess whether adoption of international standards and institutional changes improve regulatory governance and financial system soundness.
  - Policy experimentation: explore defining thresholds for soundness indices for individual countries as part of an early warning system.

*Source: _wp0489 - 1.     Weighted Least Squares Regression Results (IMF PDF).*

### 1.     Weighted Least Squares Regression Results ...................................................................21

### Weighted Least Squares Regression Results

### Introduction and Purpose
- Paper constructs two indices and quantifies their relationship:
  - Financial System Soundness Index (FSSI) — proxy for financial system stability, limited to banking system.
  - Regulatory Governance Index (RGI) — quality of regulatory governance.
- Threefold contribution:
  - Constructs an index for financial system soundness.
  - Builds an index for the quality of regulatory governance.
  - Quantifies the impact of the quality of regulatory governance on financial system soundness.
- Empirical focus limited to the banking system due to wider availability and comparability of cross-country banking data.
- Multivariate cross-sectional analysis indicates that the quality of regulatory governance matters for banking system soundness.

### Financial Stability and Operational Definition
- Adopted approach: positive and systemic view emphasizing resilience of the financial system to shocks; uses the term "financial system soundness" rather than the broader "financial stability."
- Two strands of definitions in literature:
  - Systemic resilience view (capacity to provide efficient allocation of savings to investment without major disruptions).
  - Absence-of-crises/asset-price-stability view (more directly observable but conceptually limited).
- Crockett (1997) bridge definition emphasizes:
  - (i) stability of key institutions to meet contractual obligations without interruption or outside assistance; and
  - (ii) stability of key markets so prices reflect fundamentals and do not vary substantially over short periods without fundamental changes.

### Role of Supervisors and Macroprudential Considerations
- Supervisors’ core mandate: maintain prudential and financial soundness of firms under supervision.
- Supervisors should focus on both individual institution health and system-wide implications to achieve complementarity with central banks pursuing monetary and potentially financial stability mandates.
- Growing advocacy for macroprudential approaches (Borio 2003) that incorporate system-wide risk management alongside traditional microprudential supervision.
- Financial Soundness Indicators (FSIs) and off-site analysis tools (Evans and others 2000; Sundararajan and others 2002) support macroprudential monitoring.

### Construction and Rationale of the Financial System Soundness Index (FSSI)
- FSSI purpose:
  - Capture resilience and soundness (capacity to withstand events).
  - Provide ex ante, continuous measure allowing assessment of degrees of (un)soundness rather than binary crisis/no-crisis indicators.
- FSSI composition:
  - Two quantitative variables: capital adequacy ratio (CAR) and the ratio of nonperforming loans (NPLs).
    - CAR: indicator of resilience to balance-sheet shocks.
    - NPL ratio: signals portfolio quality and solvency risk.
- Weighting:
  - The index is weighted by the share of bank credit in GDP to allow meaningful cross-country comparisons and reflect higher potential costs of instability in systems with larger credit markets.
- Justification for choices:
  - Index approach captures degrees of soundness and can detect financial distress episodes that do not culminate in full-blown crises.
  - Limitation acknowledged: variables are somewhat backward looking; broader CAMEL-type aggregation would be desirable but cross-country consistent data are not yet available.
- Sample grouping and FSSI patterns:
  - Country groups (pre-2004 WEO classification) in the sample: 11 advanced economies, 17 transition economies, and 27 developing countries.
  - Figure 1 (referenced) suggests financial system soundness is significantly higher in advanced countries than in transition or developing countries.

### Empirical Findings on Governance and Soundness
- Multivariate cross-sectional model:
  - Tests the impact of regulatory governance on banking system soundness using a range of control variables covering macroeconomic environment, financial system structure, and political/institutional setting.
- Key empirical insights:
  - Quality of regulatory governance has a direct positive effect on banking system soundness.
  - Political and institutional setting has both direct and indirect impacts on soundness.
    - Indirect links indicate effects of good regulatory governance are amplified when supported by good public sector governance.
- Preliminary evidence on nonbank sectors:
  - Including insurance sector in preliminary tests suggests regulatory governance practices are less prevalent outside the banking sector.
  - Growing systemic importance of nonbank sectors implies need for stronger regulatory governance in nonbank financial sectors to limit spillovers.

### Policy Implications and Recommendations (as presented)
- Institutional enhancements across regulatory and supervisory machinery are vital to maintain financial system soundness.
- Adoption of international practices and standards is important for financial system stability.
- Supervisory mandates and organizational structures should be designed to encourage monitoring of system-wide risks in addition to firm-level prudential oversight.
- Strengthened regulatory governance in nonbank sectors is needed as their systemic importance grows.

### Paper Structure (overview of sections and appendices cited)
- Section II: Financial stability and FSSI construction.
- Section III: Linkages between regulatory governance and FSSI; construction of RGI.
- Section IV: Model and empirical results.
- Section V: Preliminary findings on regulatory governance beyond banking.
- Section VI: Conclusions and further research agenda.
- Appendices cover: construction of the FSSI index; FSSI and Moody’s Bank Financial Strength Index comparison; construction of the Regulatory Governance Index; financial stability and its three pillars; definitions and data sources for regression variables.

*Source: _wp0489 - 1.     Weighted Least Squares Regression Results (IMF PDF).*

### Appendix II compares the FSSI with the Moody’s index. Our findings are broadly

### _wp0489 - Appendix II compares the FSSI with the Moody’s index. Our findings are broadly

### Comparison with Moody’s index and FSSI
- Appendix II compares the FSSI with the Moody’s index and states: "Our findings are broadly consistent with those of the Moody’s index."
- The detailed list of the variables in Moody’s index and the aggregation method used are not publicly available.
- Figure 1 (Financial System Soundness Index for Country Groupings 1/) presents FSSI values across country groupings with an axis spanning -0.4 to 0.6 and country groupings labeled Advanced, Transition, Developing. 1/ Country groupings according to WEO classification.

### Regulatory governance and its measurement — Governance nexus
- Borio (2003) notes the life of supervisors has changed dramatically over the past 20 years to 30 years.
- Two factors driving the shift: financial liberalization and advances in risk management. These imply:
  - rise in competitive pressures,
  - structural increase in liquidity and potential for leverage,
  - rise in the option value implicit in safety nets,
  - heightened significance of common factors that could lead to cross-market and cross-sector financial distress. 21
- Financial system strength depends on governing practices of market participants, regulators and supervisors, financial soundness of institutions, and market infrastructure efficiency.
- Good governance is a shared responsibility; it deters unsound market practices and moral hazard and enhances system-wide stress management.
- Good regulatory governance reinforces credibility and moral authority of regulatory agencies and central banks; ill-defined or dysfunctional governance undermines credibility and can spread unsound practices. 22, 23
- Regulatory and financial institutions are influenced by broader public sector governance; public sector governance impacts regulatory governance directly or indirectly through supervisory authorities. 24
- The "governance nexus" models the impact of governance practices at each layer on the practices and outputs of the next layer (Box 1). The nexus aligns with New Institutional Economics and empirical findings on institutional primacy (Williamson (2000); Rodrik and others (2002)).

### Box 1 — The Governance Nexus (components and responsibilities)
- Financial institutions:
  - Ultimate responsibility for establishing good governance to gain and keep confidence of clients and markets.
  - As lenders and stakeholders in the corporate sector, they have special interest in effective corporate governance by clients (Caprio and Levine, 2002).
  - Good bank governance helps efficient allocation of resources and financial system soundness.
- Regulatory agencies:
  - Play key role in promoting and overseeing sound practices in financial intermediaries.
  - Must establish and operate sound governance practices themselves.
  - Regulatory governance leads to good financial sector governance via agency credibility; failure to apply good governance undermines credibility and can create moral hazard.
- Public sector governance:
  - Good regulatory governance cannot be sustained without good public sector governance.
  - Includes absence of corruption, sound competition policies, effective legal and judicial system, and arm’s length approach to government ownership.
  - Political interference that is not costly prevents effective regulatory governance.

### Framework for good regulatory governance
- Good regulatory governance defined (drawing on Kaufman and others (2000)) as:
  - (a) capacity to manage resources efficiently and to formulate, implement, and enforce sound prudential policies and regulations—the duty to meet delegated objectives; and
  - (b) respect of the agency for broader goals and policies of the (elected) legislature. 25
- Supervisory uniqueness:
  - Financial supervision—especially banking—is more crucial than most other sectors because of financial intermediation’s public good aspect. 26
  - Supervisory tools (sanctioning, enforcement, revoking licenses) can affect property rights; safeguarding integrity of supervisory function requires high quality governance.
- Vulnerability of supervision:
  - Supervisory function is often invisible, making it vulnerable to subtle interference from politicians and supervised entities.
  - Government interference often appears as forbearance; consequences range from prolonged insolvency of institutions to systemic threats—justifying high quality governance.
- Four institutional components underpinning good regulatory governance (Das and Quintyn (2002)):
  - Independence,
  - Accountability,
  - Transparency,
  - Integrity.
- Interaction of components:
  - Independence and accountability are two sides of the same coin.
  - Transparency supports independence, accountability, and integrity.
  - Legal protection of agency staff and clear appointment/removal rules support independence and integrity.
  - Accountability reinforces integrity.

### Box 2 — Institutional basis for good governance (summarized)
- Independence:
  - Agency should be insulated from political and supervised-entity influence; delegation to independent agencies can rely on expertise and potentially shield interventions from political interference, increasing credibility.
- Accountability:
  - Essential to justify agency actions against mandate; independent agents should be accountable to government/legislature, stakeholders, and public.
- Transparency:
  - Agency objectives, frameworks, decisions and rationale, data, and terms of accountability should be provided comprehensively, accessibly, and timely (IMF 2000).
  - Transparency counters poor practices and contains market uncertainty.
- Integrity:
  - Mechanisms ensuring staff pursue institutional goals without compromising behavior or self-interest.
  - Includes procedures for appointment/terms/removal of heads, internal audit arrangements, standards for conduct to prevent conflicts of interest, and legal protection for staff.

### Index of Regulatory Governance (RGI)
- RGI construction:
  - Based on the framework above; a country's RGI is the weighted combination of the country’s compliance with the four components, derived from FSAP assessments (Appendix III). 27
  - Weights derived from methodology by Sundararajan, Das, and Yossifov (2003).
  - Index pertains only to banking sector supervisors, aligned with FSSI definition. 28
- Empirical observation:
  - Figure 2 shows mean RGI values for three country groups; RGI values for advanced countries are higher than for transition and developing countries.

### Empirical model and findings — Selection of variables
- Objective:
  - Verify impact of regulatory governance on financial stability by estimating correlation between RGI and FSSI using multivariate cross-sectional analysis; RGI acts as explanatory variable with control variables. 29
- Three sets of control variables used to account for:
  - (a) macroeconomic impact,
  - (b) structure of the financial system,
  - (c) broader institutional and governance environment. 30
- Macroeconomic environment captured by three indicators:
  - a measure of the government’s fiscal position (expected sign: positive—better fiscal position positively impacts soundness),
  - the rate of inflation (expected sign: negative—may negatively affect quality of bank assets),
  - short-term real interest rates (expected sign: negative—may negatively affect quality of bank assets).
- Note: Detailed data definitions and sources provided in Appendix IV.

*https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0489.pdf*

### Section V discusses the broader index, which includes insurance sector supervisors.

### _wp0489 - Section V discusses the broader index, which includes insurance sector supervisors.

### Sample, coverage, and data
- The sample includes those countries that have participated in the FSAP and for which a complete set of data for 2001 was available. Coverage is approximately 50 countries.
- The number of observations available for the regression analysis changes depending upon the variables included in the regressions.
- Public sector governance variables were drawn from the International Country Risk Guide (ICRG), Freedom House (2001), and Kaufman, Kraay and Zoido-Lobaton (2002).

### Regression framework and key empirical relationships
- Reduced form analyzed:
  - FSSI_i = ƒ (MACRO_i, STRUCT_i, RGI_i, PSGOV_i) + μ_i
  - Where FSSI is the financial system soundness index; MACRO, STRUCT, RGI, and PSGOV measure macroeconomic environment, banking system structure, regulatory governance, and public sector governance respectively.
- Bivariate correlation between RGI and FSSI:
  - 0.645, significant at the 1 percent level.
- Weighted least squares (WLS) used because the White test revealed heteroskedasticity; ordinary least squares (OLS) was considered inappropriate given heteroskedastic residuals.

### Main WLS regression findings (summary of Table 1)
- Regulatory governance (RGI) is positive and statistically significant in all model specifications tested, indicating better regulatory governance tends to strengthen financial system soundness.
- Associations with financial soundness:
  - Positive associations: low fiscal deficit (or high surplus), low real interest rate, low inflation (as expected).
  - Negative and significant associations: presence of government-owned banks and higher bank concentration.
- Public sector governance:
  - The quality of public sector governance has a direct positive impact on financial system soundness.
- Interaction effect (Column 7):
  - RGI * public sector governance has a positive and significant coefficient, indicating the impact of regulatory governance on financial stability is stronger when public sector governance quality is higher.

### Selected exact WLS coefficient estimates and model statistics (as reported in Table 1)
- Regulatory governance RGI:
  - (1) 0,696*** (t = 6,570)
  - (2) 0,331** (t = 2,263)
  - (3) 0,291* (t = 1,761)
  - (4) 0,282* (t = 1,796)
  - (5) 0,304** (t = 2,269)
  - (6) 0,157* (t = 1,710)
  - (7) 0,294** (t = 2,151)
- Government-owned banks:
  - (models reported) -0,283** (-2,446); -0,284** (-2,382); -0,271*** (-2,197); -0,323*** (-3,302); -0,339*** (-3,992); -0,388*** (-3,607)
- Fiscal balance (examples):
  - 0,317** (1,997); 0,290* (1,723); 0,315* (6%) (1,906)
- Concentration:
  - -0,232** (-2,234); -0,258*** (-2,778); -0,214** (-2,668); -0,305*** (-3,228)
- Public sector governance:
  - 0,299*** (3,081)
  - Corruption (component): 0,536*** (4,401)
- Interaction term:
  - RGI * public sector governance: 1,068* (1,558)
- Sample size and fit (selected):
  - N = 48 (reported in table)
  - R^2 values reported across models include: 0,484***; 0,628***; 0,631***; 0,675***; 0,722***; 0,768***; 0,718
  - Adjusted R^2 reported examples: 0,473; 0,600; 0,583; 0,622; 0,677; 0,739; 0,672
- Significance notation:
  - *p<.1. **p<.05. ***p<.01.

### Endogeneity concerns and 2SLS robustness checks (summary of Table 2)
- Potential endogeneity: RGI may be correlated with the error term because public sector governance and/or financial soundness indicators could influence regulatory governance.
- Two-stage least squares (2SLS) employed, using public sector governance as an instrument in the first stage (consistent with the paper’s theoretical framework).
- 2SLS results:
  - The impact of regulatory governance on financial system soundness remains positive and significant; coefficients are slightly larger and more significant than the corresponding WLS estimates.
  - The sign and significance of control variables are virtually unchanged relative to WLS.
  - Example paired WLS vs 2SLS RGI coefficients (from Table 2):
    - WLS 0,696*** (6,570) versus 2SLS 0,699*** (6,703)
    - WLS 0,331** (2,263) versus 2SLS 0,436*** (3,151)
    - WLS 0,291* (1,761) versus 2SLS 0,495*** (4,287)
    - WLS 0,282* (1,796) versus 2SLS 0,396*** (3,466)
  - 2SLS R^2 and adjusted R^2 examples reported: R^2 up to 0,755***; adjusted R^2 up to 0,715.

### Extension to the broader financial system and inclusion of insurance supervision
- Motivation:
  - Rapid structural changes in the global financial industry and increasing mainstreaming of the insurance sector make extending indicators beyond banking necessary.
  - Risks in corporate and nonbank financial sectors can pose risks to financial stability; structural linkages between insurance and banking sectors can transmit risk.
- Data and methodological caveats:
  - Work on nonbank financial soundness is at an early stage; technical, accounting, and definitional issues affect consistency and uniformity of nonbank sector data across countries.
  - Construction of a broader index covering banking and nonbank sectors (insurance) is premature and carries methodological pitfalls.
- Broader Regulatory Governance Index (including insurance):
  - Using the same methodology and data sources as the banking-sector RGI, a broader RGI including insurance supervision was derived where data permitted.
  - Finding: the value of the RGI declines when governance in insurance regulatory and supervisory processes is incorporated, suggesting that good governance practices are less widespread in the insurance sector than in the banking sector.
  - Implications:
    - Regulatory governance practices in insurance need to catch up to limit systemic risk and contagion across sectors and jurisdictions.
    - Need for national and international efforts to collect and disclose uniform, consistent data for nonbank sectors.
    - Further analyses desirable to assess the real impact of insurance companies’ soundness on the soundness of the whole financial system.

### Construction of indices and key methodological points
- Financial System Soundness Index (FSSI) construction (Appendix I):
  - FSSI composed of two Financial System Indicators (FSIs): CAR and NPLs.
  - NPLs ratio values inverted so higher values refer to greater financial stability.
  - Subindicators standardized (subtract mean, divide by standard deviation).
  - Index weighted by country’s degree of financial intermediation: share of bank credit to the private sector in GDP.
  - Index specification:
    - FSSI = credit/GDP * [1/2 (CAR + NPLs)]   (equation (2) as presented)
  - Profitability was available but excluded due to ambiguous interpretation.
- Regulatory Governance Index (RGI) construction (Appendix III):
  - Based on FSAP assessments of observance of governance-related practices of: MFP Transparency Code, Basel Core Principles (BCP), IAIS, and IOSCO (as available).
  - Scoring per practice:
    - 1 = noncompliance; 2 = partial compliance; 3 = broad compliance; 4 = full compliance; 9 = not applicable/not answered.
  - In MFP Transparency Code, practices relevant to multiple dimensions received multiplicative weighting (double, triple, quadruple) according to relevance across dimensions.
  - Country SCORE_ij computed as:
    - SCORE_ij = [0*noncompliance_ij + 0,33*partial compliance_ij + 0,66*broad compliance_ij + full compliance_ij]*100   (equation (3) as presented)
  - RGI_i = 1/n (Σ SCORE_ij)  over j = 1..n  (equation (4) as presented)
  - Only valid scores were used (not answered or not applicable items excluded).

### Conclusions, policy implications, and research agenda
- Core conclusion:
  - Regulatory governance matters for financial system soundness; better regulatory governance is associated with higher FSSI scores, robust to specifications and endogeneity correction.
- Policy implications:
  - Emphasize strengthening regulatory governance institutions — independence, accountability, transparency, and integrity — to improve financial system soundness.
  - Improvements in overall public sector governance will enhance the effectiveness of regulatory governance and contribute to financial system soundness.
- Research agenda and next steps suggested by the paper:
  - (i) Develop the FSSI for the broader financial system by collecting consistent indicators for nonbank segments (insurance) closely linked to banking—urgent given preliminary findings of lower RGI in insurance.
  - (ii) Fine-tune the FSSI by including other indicators, such as measures of liquidity and indicators for quality of financial market infrastructure.
  - (iii) Develop time series to assess whether adoption of international standards and institutional changes (e.g., unification of supervisory functions) improve regulatory governance quality and, consequently, financial system soundness.
  - Policy experimentation:
    - Explore defining thresholds for soundness indices for individual countries as part of an early warning system (in the spirit of Potter (1995)).

*Source: _wp0489 - Section V discusses the broader index, which includes insurance sector supervisors.*

### REFERENCES

### _wp0489 - REFERENCES

### Major themes and topical coverage
- Financial stability and central banking
  - Brealey et al., 2001, Financial Stability and Central Banks. A Global Perspective.
  - Crockett, 1997, “Why Is Financial Stability a Goal of Public Policy?”
  - McDonough, William, 2002, “Promoting Financial Stability,” International Finance, Vol. 5, No. 1, pp. 115–127.
  - Schinasi, Garry, 2003, “Responsibility of Central Banks for Stability of Financial Markets,” IMF Working Paper 03/121.
  - Padoa-Schioppa, Tommaso, 2002 and 2003, “Central Banks and Financial Stability.”

- Bank regulation, supervision, and governance
  - Barth, Caprio, and Levine, 2000, “Banking Systems Around the Globe: Do Regulation and Ownership Affect Performance and Stability?” World Bank Policy Research Working Paper No. 2325.
  - Barth, Caprio, and Levine, 2001, “Bank Regulation and Supervision: What Works Best?” NBER Working Paper No. 9323.
  - Goodhart, 2001 and 2002, on regulating regulators and the organizational structure of banking supervision.
  - Quintyn and Taylor, 2003, “Regulatory and Supervisory Independence and Financial Stability,” CESifo Economic Studies, Vol. 49, No. 2, pp. 259–294.
  - Hupkes, Quintyn, and Taylor, 2004, “The Accountability of Financial Sector Supervisors. Principles and Practice,” IMF Working Paper (forthcoming).

- Financial soundness indicators (FSIs), macroprudential indicators, and stress testing
  - Evans et al., 2000, Macroprudential Indicators of Financial System Soundness, IMF Occasional Paper No. 192.
  - IMF, 2001, ”Financial Soundness Indicators: Policy Paper.”
  - IMF, 2003, “Financial Soundness Indicators.”
  - Das, Davies, and Podpiera, 2003, “Insurance and Issues in Financial Soundness,” IMF Working Paper 03/138.
  - International Association of Insurance Supervisors, 2003a, Insurance Core Principles and Methodology; 2003b, Stress Testing by Insurers, Guidance Paper, Guidance Report No. 8.
  - Craig and Sundararajan, 2003, “Using Financial Soundness Indicators to Assess Risks to Financial Stability.”

- Crisis analysis, early warning, contagion, and financial fragility
  - Corsetti, Resenti, and Roubini, 1998, “Paper Tigers? A Model of the Asian Crisis,” NBER Working Paper No. 6783.
  - Corsetti, Resenti, and Roubini, 1999, “The Asian Crisis: an Overview of the Empirical Evidence and Policy Debate,” pp. 127–163.
  - Goldstein, Kaminsky, and Reinhart, 2000, Assessing Financial Vulnerability: An Early Warning System for Emerging Markets.
  - Mishkin, 1991, NBER Working Paper No. 3934; 1997 and 1999 works on financial instability and consolidation.
  - Demirgüç-Kunt and Detragiache, 1998, “Financial Liberalization and Financial Fragility,” IMF Working Paper 98/831.
  - Rossi, Marco, 1999, IMF Working Paper 99/66, “Financial Fragility and Economic Performance in Developing Economies.”

- Macroprudential frameworks, systemic risk, and policy regimes
  - Borio, 2003, “Towards a Macro-Prudential Framework for Financial Supervision,” BIS Working Paper No. 128.
  - Borio and White, 2003, “Whither Monetary and Financial Stability? The Implications of Evolving Policy Regimes,” Jackson Hole presentation.
  - Tsatsaronis, 2002, “Systemic Financial Risk and Macroprudential Supervision.”

- Ownership, competition, and banking structure
  - La Porta, Lopez-de-Silanes, and Schleifer, 2002, “Government Ownership of Banks,” Journal of Finance, Vol. 57, No. 1, pp. 265–301.
  - Beck, Demirguc-Kunt, and Levine, 2003, “Bank Concentration and Crises,” NBER Working Paper No. 9921.
  - Boyd and de Nicoló, 2003, “Bank Risk Taking and Competition Revisited,” IMF Working Paper 03/114.
  - Barth, Caudill, Hall, and Yago, 2000, “Cross-Country Evidence on Banking Crises: Do Financial Structure and Bank Regulation Matter?” in Bank Fragility and Regulation, pp. 3–23.

- Transparency, codes of good practice, and institutional governance
  - IMF, 1999, Code of Good Practices on Transparency in Monetary and Financial Policies.
  - IMF, 2000, Supporting Document to the Code of Good Practices on Transparency in Monetary and Financial Policies, July.
  - Sundararajan, Das, and Yossifov, 2003, Cross-Country and Cross-Sector Analysis of Transparency of Monetary and Financial Policies, IMF Working Paper 03/94.
  - Kaufmann, 2002, and Kaufmann, Kraay, and Zoido-Lobaton, 2003, “Governance Matters II; Updated Indicators for 2000-01,” World Bank Policy Research Working Paper No. 2772.
  - Damania, Fredriksson, and Muthukumara, 2003, “The Persistence of Corruption and Regulatory Compliance Failures: Theory and Evidence,” IMF Working Paper 03/172.

- Regulatory standards, Basel, and international coordination
  - Large, Andrew, 2003, “Basel II and Systemic Stability,” Basel II-CAD 3 Conference paper, March.
  - Financial Stability Forum, 2002, “Review of Vulnerabilities and Efforts to Strengthen the International Financial System.”
  - Joint Forum, 2003, “Trends in Risk Integration and Aggregation.”

- Analytical methods and econometric approaches referenced
  - Greene, William, 1997, Econometric Analysis, 3rd Edition.
  - Potter, Simon, 1995, “A Non-Linear Approach to U.S. GNP,” Journal of Applied Econometrics, Vol. 10, pp.109–125.

### Representative documents and serial identifiers (selected)
- IMF Working Papers and Occasional Papers cited with identifiers:
  - IMF Working Paper 03/114; 03/138; 03/121; 03/172; 03/94; 00/76; 98/831; 99/66; 01/62.
  - IMF Occasional Paper No. 192; No. 212.
- NBER Working Papers cited:
  - No. 6783; No. 9323; No. 9921; No. 3934; No. 9305.
- Journal and volume citations:
  - Journal of Finance, Vol. 57, No. 1, February, pp. 265–301.
  - International Finance, Vol. 5, No. 1, pp. 115–127.
  - CESifo Economic Studies, Vol. 49, No. 2, pp. 259–294.
  - Journal of Applied Econometrics, Vol. 10, pp.109–125.
- Institutional reports and web resources:
  - Bank of International Settlements, “70th Annual Report,” June 5, 2000.
  - Freedom House, 2001, Annual Survey of Freedom.
  - World Bank, “World Development Indicators.”

*Referenced document list extracted from _wp0489 - REFERENCES (source PDF).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0489.pdf_