## FSAP REVIEW—FURTHER ADAPTATION TO THE POST-CRISIS ERA—BACKGROUND STUDIES (excerpts from _081814b)

## Source details

**Canonical URL:** [FSAP REVIEW—FURTHER ADAPTATION TO THE POST-CRISIS ERA—BACKGROUND STUDIES (excerpts from _081814b)](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_081814b.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_081814b.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_081814b.pdf.json)

---

### Motivation and context
- Financial systems and institutions have become increasingly sophisticated, integrated within and across borders, larger, and complex (Otker-Robe et al., 2011).
- Systemically important institutions continue to be perceived as “too important to fail” (IMF, 2014).
- Key objectives for FSAPs since 2009: refine stress-testing methodologies to incorporate macro-financial and cross-border linkages; address regulatory perimeter, procyclicality, and supervision of cross-border entities; enhance national authorities’ capacity to coordinate domestically and internationally; strengthen crisis preparedness and management strategies.
- A central element is authorities’ capacity to implement macroprudential policies (MaPP) targeted at identifying and managing systemic risk.
- Review scope: 60 FSAPs with scoping missions during 2010–13; sample includes 23 advanced economies (AEs), 37 emerging market and developing economies (EMDEs), and 8 low income countries (LICs).
- Framework: Uses the “Key Aspects” paper (IMF, 2013) as frame of reference covering technical capacity, tool calibration/communication, regulatory/information gaps, institutional frameworks, policy coordination, and multilateral/cross-border challenges.

### MaPP coverage in FSAPs — coverage, evolution, and operational aspects
- Coverage and evolution
  - Many FSSAs completed in 2010–14 include explicit discussions of MaPP; depth and breadth vary across countries.
  - Advanced economy FSSAs contain slightly more extensive coverage.
  - Selected country examples with explicit MaPP technical notes: Albania, Armenia, Australia, Bahamas, Brazil, Canada, China, Chile, the Czech Republic, France, Ghana, Indonesia, Italy, Kosovo, Malaysia, Spain, Sweden, Switzerland, and the United Kingdom (U.K.).
- Systemic risk assessment and analysis (operational aspects)
  - Systemic risk assessments have become standard, replacing sector-focused risk assessments.
  - Methods used: network analysis, stress-testing, contingent-claims, distress-dependence tools for contagion and structural systemic-risk analysis.
  - FSAPs analyze bank–nonbank linkages (examples: Brazil, Chad, China, the Czech Republic, Kosovo, Malaysia, Uruguay) and bank–sovereign linkages (examples: Colombia, EU, France, Italy, Spain, Switzerland).
- Findings and common recommendations on systemic risk assessment
  - Strengthen monitoring and close information/regulatory gaps.
  - Revamp stress-testing exercises and intensify supervisory monitoring of financial linkages and contagion channels.
  - Collect/disseminate high-quality data on indicators such as real estate prices.
  - Grant supervisory authorities explicit legal authority to oversee all entities within a conglomerate, subject like-activities to like-regulation, and coordinate regulation (and taxation) of close substitutes.

### MaPP toolkit, calibration, and effectiveness
- Toolkit and calibration
  - FSAPs recommend expanding or tightening MaPP tools depending on country experience; less frequently examine tool effectiveness and leakages.
  - Calibration challenges: calibrating MaPP instruments to changes in systemic risk is challenging and receives limited coverage.
  - Authorities sometimes calibrate by simulation and cross-country experience (example: Singapore); other authorities rely on judgment due to insufficient evidence (example: Australia).
- Tool selection guided by risk source
  - Time-dimension risks (rapid credit growth or real estate overheating): tighten LTV or DTI limits; countercyclical capital buffers (examples: Austria, Brazil, Netherlands, Poland, Switzerland).
  - Structural/connectivity risks (growing connectivity, large conglomerates): SIFI capital surcharges, higher capital/liquidity buffers, intensive supervision, enhanced coordination, widened regulatory perimeter (examples: India, Malaysia).
- Selected quantitative illustrative findings (Box 1)
  - Czech Republic: simulated partial loss assuming 40 percent of gross exposure to parent lost; capital shortfall as much as two percent of GDP under severe double dip recession.
  - Poland: top-down stress tests suggested up to 30 percent of Polish banks may fail to meet minimum capital adequacy ratios under severe macroeconomic scenarios.
  - Country-specific methodological highlights: Germany, Hong Kong SAR, India, Italy, Korea, Malaysia, Singapore, Sweden, Switzerland, United Kingdom — various combinations of bank-by-bank stress testing, network analysis, contagion tests, and market-based indicators (JPoD, CoPoD).

### Institutional framework for MaPP — findings and representative recommendations
- General findings
  - Strong institutional frameworks are essential to ensure MaPP effectiveness, willingness to act, limit inaction biases, and reduce incentives to use MaPP for non-stability purposes.
  - No consensus on single institution vs. inter-agency committee; FSAPs provide country-specific recommendations.
- Representative country recommendations (selected)
  - Make central bank main MaPP authority, expand mandate to include systemic risk monitoring and financial stability, grant clear role and accountability (Belgium, Czech Republic, Germany, Netherlands).
  - Assign MaPP to inter-agency body or committee with “act or explain” powers and a leading central bank role (Austria, Brazil, Chile, China, El Salvador, EU, Israel, Korea, Malaysia, Mexico, Paraguay, Poland, Sweden, Turkey).
  - Less prescriptive models in some FSAPs (Canada, Italy, Switzerland).
- Core institutional features emphasized
  - Provide MaPP authority with appropriate powers, clear goals, mandate, and accountability.
  - Assign major role to central bank given expertise while protecting independence.
  - Ensure adequate government involvement to garner political support for MaPP actions.
- Typical omission: FSAPs typically do not detail MaPP implementation modalities (nature of powers, act-or-explain vs. direct control, rules vs. discretion).

### Interaction of MaPP with other policies and cross-border aspects
- Interaction with monetary, fiscal, microprudential, and other policies
  - MaPP cannot alone safeguard financial stability; needs support from microprudential supervision, monetary, fiscal, structural policies.
  - Clear delineation of objectives across policies is recommended to limit MaPP use for non-macroprudential objectives.
  - FSAP examples: Poland (separate MaPP objectives), Germany (need for free exchange of information between micro- and macro-prudential supervisors), Netherlands (synergies in twin-peaks approach).
- Crisis prevention and resolution
  - Emphasized importance of robust safety nets (emergency liquidity assistance, bank resolution, deposit insurance) and coordination with fiscal authorities and peers for cross-border resolution.
- Multilateral and cross-border MaPP aspects
  - Multilateral aspects received relatively less attention in FSAPs.
  - Need for coordination to limit cross-border arbitrage and reciprocity arrangements (examples cited: Poland, Australia, EU, Malaysia, Mexico, U.K.).
  - Home-host supervisory coordination emphasized: MOUs, supervisory colleges, crisis management groups, information and burden-sharing, group-wide supervision, joint assessments, monitoring intra-group exposures.

### Box 2 and Box 3 highlights — institutional recommendations and foreign-bank treatment
- Box 2: Detailed, country-level MaPP authority recommendations (examples summarized)
  - Austria: joint committee chaired by OeNB; OeNB for systemic risk monitoring, Financial Market Authority for prudential supervision.
  - Brazil: tri-partite Financial Stability Committee (FSC), central bank retaining key role.
  - China: permanent FSC with PBC secretariat and access to confidential supervisory data.
  - Poland: Systemic Risk Board (SRB) with NBP leading role, independence and “act or explain” powers, publication of minutes.
  - Switzerland: consider placing responsibility/powers for MaPP with one institution or committee.
- Box 3: Foreign-bank branches and subsidiaries (selected treatments and country recommendations)
  - Australia: subsidiaries of major Australian banks account for 90 percent of New Zealand banking assets; recommended enhancing cross-border coordination and ring-fencing.
  - Hong Kong SAR: foreign-bank branches account for 35 percent of total local banking system assets and 27 percent of domestic lending; encourage subsidiarization and robust bilateral supervisory arrangements.
  - Singapore: 116 of 122 banks are foreign-bank branches owning 65 percent of banking system assets; recommend asset maintenance ratios, strict licensing, monitoring parent-bank risks, international coordination.
  - Netherlands, Poland, Spain, U.K.: recommendations to increase supervisory resources, reciprocity use, MOUs, supervisory colleges, and collaboration to limit contagion and regulatory arbitrage.

### Trade-offs, interactions, and multilateral considerations
- Trade-offs between MaPP and crisis management
  - Example: India — precautionary buffers of government securities provide liquidity protection but expose banks to unmanaged interest rate risk.
  - When crisis management and MaPP roles coincide in same institution (e.g., Armenia), formal arrangements recommended to reduce conflicts.
- Interactions with monetary policy
  - Examples: Armenia (central bank establishes dedicated financial stability committee); India (MaPP instruments used in conjunction with monetary policy tightening/loosening).
  - Recommendation: limit use of central-bank tools without macroprudential intention; use conventional monetary policy for macro management.
- Interactions with fiscal policy and market infrastructure
  - Use of tax instruments to address systemic risk: stamp duty in Hong Kong SAR.
  - Tax features can contribute to systemic risk: tax deductibility of interest payments in Netherlands and Switzerland; special bank tax in Slovenia.
  - Market infrastructure policies (CCPs, OTC standardization) noted for systemic liquidity risk management.
- Multilateral cross-border aspects
  - Limited coverage of spillovers, leakage, and boundary effects; FSAPs call for coordination when unilateral national actions have cross-border implications.

### Stress testing — principles, perimeter, data, and practices
- Stress testing role and principles
  - Core part of FSAPs for macroprudential assessment; IMF (2012) “Macrofinancial Stress Testing—Principles and Practices” provides seven best-practice principles; review focuses on four relevant to risk propagation.
- Principle No. 1 — institutional perimeter
  - Institutional perimeter differs substantially across countries; analysis of interconnectedness concentrated on SIs, AEs, and few others; LICs are missing.
  - Stress tests of non-bank financial institutions concentrated on SIs, AEs, OFCs; EMs and LICs largely missing.
  - Many shadow banking entities and activities often left aside; substantive non-bank stress testing found only in U.S. FSSA.
- Principle No. 2 — channels of risk propagation
  - Cross-border spillover and contagion analysis conducted only for a small number of countries, concentrated on SIs.
  - Single-factor risk analysis (applying a single risk factor to a static bank model) used in some OFCs and LICs.
  - Policy feedbacks (authorities’ reactions) often not incorporated; omission can exaggerate perceived shocks.
- Principle No. 3 — include all material risks and buffers
  - Data gaps are common across AEs, EMs, and LICs; Figure evidence shows FSAPs explicitly mention data gaps.
  - Use of public data only by some teams (example: China FSAP 2010 executive summary noted FSAP team used public data only).
  - Consequence: data gaps hinder full assessment and contribute to selectivity in stress-testing coverage.
  - Capital buffer measurement issues: teams sometimes make judgmental adjustments for regulatory forbearance and low provisioning (examples: Russian Federation FSAP 2011; Rwanda FSAP 2011).
- Principle No. 5 — focus on tail risks and rule-of-thumb for GDP shock
  - Common rule of thumb: two standard deviation (2SD) shock on real GDP growth over two years applied to baseline real GDP projection.
  - Concerns with 2SD rule: sample period sensitivity, mean shifts, serial correlation, penalizing countries in recession at FSAP time.
  - Proposed alternative Methods:
    - Method (i): Decompose GDP into trend and cyclical component with HP filter; subtract 2SD of cyclical component from trend GDP expected in one to two years.
    - Method (ii): Estimate AR model of cyclical component; construct stress by subtracting 2SDs of AR error term from one- and two-period forecasts.
  - Comparison findings:
    - Alternative methods generally yield smaller shocks than current 2SD rule.
    - Actual scenarios used by FSAP teams are generally milder than the 2SD rule and closer to alternative methods, but deviations are ad hoc.
- Standardization and forward-looking guidance
  - Authors propose standardizing the meaning of two standard deviations using stochastic forecasting models and reducing influence of FSAP timing.
  - Recommend presenting a uniform-rule scenario in the FSSA and explaining any judgmental deviations.

### Evenhandedness, BCP assessments, assessor effects, and FSAP distribution
- FSAP usage and distribution
  - Board made FSAPs mandatory for 25 systemically important jurisdictions (SIs) in September 2010; list expanded to 29 in 2013.
  - SIs’ FSAP share increased from 44.0 percent in FY2005–09 to 100 percent in FY2010–14.
  - Non-SI EMs’ FSAP share declined to 41.3 percent in FY2010–14; LICs’ FSAP share declined to 27.6 percent in FY2010–14.
  - Implied average FSAP intervals (years):
    - Total: based on FY2005–09 = 10.2; based on FY2010–14 = 11.9
    - SIs: based on FY2005–09 = 11.4; based on FY2010–14 = 5.0
    - Non-SI AEs: based on FY2005–09 = 12.0; based on FY2010–14 = 10.0
    - Non-SI EMs: based on FY2005–09 = 7.2; based on FY2010–14 = 12.1
    - LICs: based on FY2005–09 = 11.9; based on FY2010–14 = 18.1
  - Regional distribution shifts: AFR and MCD shares dropped; APD share increased; EUR consistently well-represented; WHD had highest share in FY2010–14 (close to 60 percent).
  - Policy issue for Board: trade-offs among expanding FSAP resources, reducing SIs’ FSAP frequency, or simplifying SI FSAPs to release resources.
- BCP assessments and comparability
  - BCP qualitative grades converted to numeric scores for analysis: compliant = 4, largely compliant = 3, materially non-compliant = 2, non-compliant = 1.
  - Sample: 69 BCP assessments (FY2010–14), excluding certain new BCP assessments in principle-level exercises.
  - Benchmarks correlated with average BCP scores: per capita income, Corruption Perception Index, Ease of Doing Business Index.
  - Regression results (overall sample; Yi = α + βXi + εi):
    - Per capita income (log): α = 1.298***, β = 0.207***
    - Corruption Perception Index value: α = 2.382***, β = 0.0159***
    - Ease of Doing Business Index ranking: α = 3.614***, β = -0.00630***
    - Note: *, **, *** imply ten percent, five percent, and one percent significance, respectively.
  - Observation: SI/EMs appear rated better than benchmarks suggest; published assessments bias toward more favorable results (unfavorable assessments tend to remain unpublished).
- Assessor identity effects
  - Assessments typically conducted by two experts; wide distribution of assessors complicates comparability.
  - Distribution of number of assessments per assessor (sample):
    - 7 → 1 assessor; 6 → 2; 5 → 1; 4 → 6; 3 → 6; 2 → 12; 1 → 40
  - Assessor dummy regressions for four frequent assessors showed significant assessor effects (example: Assessor A (7): PCI 0.373**, CPI 0.467***, EoDB 0.574***; Assessor C (6): PCI -0.399**).
  - Conclusion: assessor identity can influence average assessment scores by up to about half a notch in the four-notch BCP scale, suggesting need for greater standardization.

### Stress-testing perimeters, data gaps, and suggested practice improvements
- Institutional perimeter differences may reflect financial system structure or data availability; scope gaps noted for non-bank financial sector and shadow banking.
- Data constraints (Box 4 examples)
  - China (FSAP 2010): FSAP team used public data only; top-down tests by authorities used supervisory data; data gaps hindered full assessment.
  - Common data gaps: sectoral exposures, off-balance sheet positions, short time series with structural breaks, constraints on confidential data access.
- Capital buffer measurement and team adjustments (Box 5)
  - Example adjustments: Russian Federation (adjustments could wipe out as much as one-third of total bank capital); Rwanda (core capital-to-asset ratio adjustments dropping to 8.3 percent in a conservative scenario).
  - Recommendation: staff should document actual and hypothetical cases of under-reporting and explain adjustments to foster common understanding and enhance evenhandedness.
- Stress-test design and standard deviation sample-period issues
  - Teams apply varying sample periods and ad hoc treatments when calculating standard deviations for GDP shocks; examples show inclusion/exclusion of past crisis episodes (Asian Crisis) materially affects SD estimates.
  - Proposed uniform rules: standardize meaning of two standard deviations using stochastic forecasting models and require presentation of uniform-rule scenario in the FSSA with justification for deviations.

### FSAP costs, resources, and composition
- Cost estimation methodology
  - Two series constructed: cash-basis and accrual-basis (merge ACES-based series with pre-ACES; accrual allocates FSAP costs to fiscal year of Executive Board discussion).
- Aggregate cost trends and magnitudes
  - Statement: "The cost of running the FSAP in the last 4 years has been around US$14 million on a cash basis, and has fluctuated between US$8–16 million on an accrual basis."
  - Both series show rising costs, particularly after FY2011 and FY2012.
- Cash-basis key statistics (FY 2006 through FY 2014; figures presented exactly as in source)
  - Total Cost: 9,067,877; 6,598,027; 10,298,747; 6,932,133; 8,737,849; 13,560,272; 14,406,613; 13,009,222; 12,699,900
  - Total Personel: 6,452,193; 4,942,343; 7,619,195; 5,109,016; 7,069,627; 10,321,394; 11,394,243; 10,127,994; 10,394,687
  - Staff: 5,521,857; 4,528,463; 6,940,347; 4,487,330; 6,251,007; 8,855,712; 9,189,998; 8,483,672; 8,784,771
  - Experts: 880,519; 362,396; 653,306; 594,835; 750,256; 961,628; 1,390,345; 1,130,686; 1,171,429
  - Contractuals: 49,817; 51,484; 25,541; 26,851; 68,364; 504,054; 813,900; 513,636; 438,487
  - Travel: 2,615,684; 1,655,683; 2,679,552; 1,823,117; 1,668,222; 3,238,877; 3,012,370; 2,881,228; 2,305,212
  - Personnel in % of Total Cost (rhs): 71; 75; 74; 74; 81; 76; 79; 78; 82
- Accrual-basis key statistics (FY 2010–FY 2014; Table 11 as in source)
  - Total Cost: 5,677,532; 8,156,517; 16,989,765; 14,743,054; 14,300,957
  - Average Cost: 283,877; 543,768; 738,685; 867,238; 1,021,497
  - Advanced Economy (AE) Total Cost: 917,911; 3,299,904; 6,308,941; 8,322,697; 6,511,305
  - AE - S25 Total Cost: ...; 2,615,542; 5,639,385; 6,975,182; 6,511,305
  - Number of FSAPs (FY 2010; FY 2011; FY 2012; FY 2013; FY 2014): 20 15; 23 17 14
  - Number of FSAPs by category (as presented in source):
    - Advanced Economy (AE): 4 4 6 6 5
    - AE - S25: 0 1 5 4 5
    - Middle-Income Countries (MIC): 8 8 12 9 8
    - MIC-S25: 0 0 5 1 2
    - Low-Income (LIC): 8 3 5 2 1
    - LIC-S25: 0 0 0 0 0
- Accrual-basis sample composition by fiscal year (Table 12 — country lists preserved as in source for FY2007–FY2014).

### Survey findings, perceptions, and implementation
- Survey response and composition
  - Invitations sent to national authorities of 168 countries; responses received from 55 (33 percent).
  - Executive Directors surveyed: 24 invited, 8 responses received (33 percent).
  - Response rates by income group: Advanced countries 56 percent (24/43), Emerging 32 percent (28/88), Low income 8 percent (3/37).
- Perceptions of FSAPs’ analytical work
  - Over 90 percent of country authorities and all Executive Directors see FSAPs as adequately structured around three main pillars.
  - Over 75 percent of respondents agreed analyses were adequately framed in prevailing macro-financial conditions and calibrated to available data granularity.
  - Agreement on adequacy of models/methodologies in stress tests: 71 percent for country authorities; 51 percent for Executive Directors.
  - Lower approval for transmission, feedback, contagion analyses (examples: real-financial transmission channels approved by 59 percent country authorities; 38 percent Executive Directors).
- Recommendations, implementation, and impact
  - Implementation status: about one third of respondents implemented recommendations in full; slightly over 60 percent implemented partially.
  - Reasons for non/partial implementation: nearly half cited disagreement with recommendations; Executive Directors cited recent FSAP completion (60 percent) and limited capacity (40 percent).
  - Perceived contributions: assessment of financial system stability, regulatory and supervisory framework assessments, macrofinancial linkages, crisis management capabilities.
  - Overall usefulness: 79 percent for country authorities; 100 percent for Executive Directors.
- Publication effects
  - In about 60 percent of countries that published FSSAs, publication strengthened supervisory authorities’ credibility and was seen as positive by domestic financial entities.
  - For a third of published FSSAs, publication fostered broader national policy dialogue.

### Conclusions, recommended improvements, and areas for further study
- Overall review conclusions
  - Systemic risk assessment, MaPP frameworks, and role in financial stability have become important FSAP features; explicit MaPP discussions and technical notes are more common.
  - Depth, breadth, and integration of MaPP assessments vary across countries and over time; AEs and advanced EMs had more comprehensive coverage; more recent FSAPs show better integration.
- Recommended improvements for future FSAPs
  - Better integrate MaPP analysis into overall stability analysis and executive summaries; cross-reference key findings throughout FSSA reports.
  - Systematic treatment of MaPP frameworks and operationalization of the Key Aspects paper (and forthcoming Staff Guidance Note on Macroprudential Policy).
  - Use systemic risk as the organizing concept across three pillars: main threats; monitoring/regulation/supervision; authorities’ ability to mitigate/response.
  - Assess authorities’ capacity to operate MaPP frameworks effectively, including multilateral aspects, calibration, costs/benefits, and cross-border spillover effects as AEs normalize unconventional monetary policies.
- Areas for further study
  - Substance review of BCP assessments to corroborate or contradict apparent biases in aggregate scores, and to ensure even coverage of revised BCP with macroprudential elements.
  - Develop standardized simple models for stress scenarios to minimize ad hoc adjustments; clarify the types of stress events the Board wants tested.

*Source: FSAP REVIEW—FURTHER ADAPTATION TO THE POST-CRISIS ERA—BACKGROUND STUDIES (excerpts from the supplied content unit _081814b).*

### 1. Structural Dimension of Systemic Risk: Selected Cases of Contagion and Spillover Risk

### 1. Structural Dimension of Systemic Risk: Selected Cases of Contagion and Spillover Risk — Analysis in FSAPs

### Motivation
- Financial systems and institutions have become increasingly sophisticated, integrated within and across borders, larger, and complex (Otker-Robe et al., 2011).
- Systemically important institutions continue to be perceived as “too important to fail” (IMF, 2014).
- Increased economic and financial connectivity raises exposure to potentially destabilizing macro-financial feedback loops, heightening the importance of FSAPs providing a critical review of countries’ ability to manage and respond to systemic risk.
- A key element for managing systemic risk is the capacity to implement macroprudential policies (MaPP) targeted at identifying and managing systemic risk.
- The 2009 FSAP review envisaged:
  - refining stress-testing methodologies to incorporate macro-financial and cross-border linkages;
  - addressing the regulatory perimeter, procyclicality of prudential norms, and supervision of cross-border entities;
  - enhancing the capacity of national authorities to coordinate with domestic and foreign peers; and
  - strengthening crisis preparedness and management strategies.
- The coverage of MaPP frameworks in, and their integration to, FSAPs have evolved over time as MaPP policy lines were formulated in parallel with FSAP scope changes; coverage is expected to further improve with the forthcoming Staff Guidance Note on MaPP.

### Assessing the Macroprudential Policy Coverage of FSAPs
- Review scope and sample:
  - Based on a review of mostly published Financial System Stability Assessments (FSSAs).
  - Covers a set of 60 FSAPs for which the scoping missions were undertaken during 2010–13.
  - FSAPs initiated during the calendar years 2009 and 2014 were not included to limit analysis to those carried out after the last review and to those with completed FSSA reports.
  - The review covers countries and jurisdictions from all income groups, including:
    - 23 advanced economies (AEs)
    - 37 emerging market and developing economies (EMDEs)
    - 8 low income countries (LICs)
- Framework for assessment:
  - Uses the “Key Aspects” paper (IMF, 2013) as the frame of reference. These “key aspects” include:
    - having the technical capacity to assess systemic risk and assemble the toolkit to mitigate it;
    - calibrating and communicating the MaPP tools;
    - identifying/addressing gaps in regulation and information;
    - establishing institutional frameworks to implement MaPP;
    - ensuring coordination with monetary, fiscal, microprudential, and other policies; and
    - addressing challenges arising in a multilateral context (e.g., through cross-border spillover of MaPP actions (or lack of action), cross-country differences in the phase of financial cycles, and conflicts between home-host authorities of cross-border banking groups).
- Purpose of the review:
  - Documents the extent to which MaPP coverage has become an important part of FSAPs since the last review.
  - Discusses the areas covered more extensively.
  - Identifies ways the coverage could be strengthened and better integrated into financial stability assessments.

### Source material and scope specifics
- The paper analyzes MaPP coverage evolution in FSAPs over the post-2009 period and focuses on integrating MaPP analyses into overall financial stability assessments.
- Table 1 (FSAP Coverage, 2010–13) enumerates jurisdictions and FSSA publication status for the 60 FSAPs reviewed (sample spans AEs and EMDEs across regions).

*Source: IMF — MacRoprudential Policy Coverage in FSAPs: Progress and Remaining Challenges (excerpt from _081814b)_*

### 6.      The review suggests that macroprudential issues have indeed become an important

### 6. The review suggests that macroprudential issues have indeed become an important feature of FSAPs in recent years

### Coverage and evolution
- Many FSSAs completed in 2010–14 include explicit discussions of MaPP, some with technical notes dedicated to various aspects of the MaPP framework and systemic risk, although depth and breadth of coverage differ across countries.
- Macroprudential issues are sometimes covered without explicit references to MaPP (for example, discussions of system-wide risks associated with growing indebtedness of the corporate or household sectors; interconnectedness and complexity of financial systems; risks and supervisory challenges in oversight of financial conglomerates; or the need for home-host coordination to address systemic risks from cross-border bank activities).
- Advanced economy FSSAs contain a slightly more extensive coverage, reflecting the prominence of macroprudential issues in a given economy.
- Depth of coverage has grown along with the evolution of policy lines on MaPP, with FSAP teams drawing on a wide range of conceptual, analytical and empirical MaPP work within and outside the Fund.

*Selected country examples noted in explicit MaPP technical notes:* Albania, Armenia, Australia, Bahamas, Brazil, Canada, China, Chile, the Czech Republic, France, Ghana, Indonesia, Italy, Kosovo, Malaysia, Spain, Sweden, Switzerland, and the United Kingdom (U.K.).

### Operational aspects of MaPP — systemic risk assessment and analysis
- Systemic risk assessments have become a standard feature of FSAPs, replacing assessments that focused primarily on risks to individual sectors rather than the financial system as a whole.
- FSAP teams carry out systemic risk analyses covering both time and structural dimensions of systemic risk associated with buildup of vulnerabilities or financial interlinkages.
- Methods used include network analysis, stress-testing, contingent-claims, and distress-dependence tools to carry out contagion analyses and assess structural dimensions of systemic risk.
- FSAPs discuss systemic risk arising from bank–nonbank linkages (for example, Brazil, Chad, China, the Czech Republic, Kosovo, Malaysia, Uruguay) and, in some cases, bank–sovereign linkages (for example, Colombia, the European Union (EU), France, Italy, Spain, Switzerland).

Findings and recommendations related to systemic risk assessment:
- FSAPs emphasize challenges in properly identifying and measuring systemic risk and recommend strengthening monitoring and closing information and regulatory gaps.
- Common recommendations include revamped stress-testing exercises, intensive supervisory monitoring of financial linkages and potential contagion channels, more granular supervisory data on exposures of financial institutions, and collection/dissemination of high-quality data on indicators such as real estate prices.
- FSAPs recommend that supervisory authorities have explicit legal authority to oversee all entities within a conglomerate, to subject like-activities to like-regulation, and to coordinate regulation (and taxation) of close substitutes.

### MaPP toolkit, calibration, and effectiveness
- FSAP assessments often discuss measures undertaken by authorities to address emerging risks and offer recommendations to strengthen the toolkit.
- Some FSAPs recommend expanding the MaPP toolkit to mitigate time and structural dimensions of systemic risk; others recommend tightening or eliminating tools based on country experience.
- Less frequently, FSAPs examine the effectiveness of tools in place; in a few cases they assess costs and potential leakages (examples: countercyclical tax on banks in Slovenia; direct lending limits in China; leakages noted in Brazil, China, India FSSAs).
- Brazil FSSA: financial transaction tax and differentiated capital requirements were generally effective, but effects were sometimes temporary owing to circumvention through other tools.

Calibration challenges:
- Calibrating MaPP instruments to changes in systemic risk is challenging and receives relatively limited coverage in FSAPs.
- Some FSAPs note authorities calibrate instruments to target specific risk factors using simulation of impacts and cross-country/past experience and accounting for uncertainty (example: Singapore).
- Other FSAPs report authorities relying on judgment due to insufficient evidence on efficacy of certain MaPP tools in comparable countries (example: Australia uses judgment about borrower repayment ability or lending standards instead of imposing LTV and DTI limits).

Tool selection guided by risk source:
- FSAP teams typically discuss specific MaPP tools guided by the prevalent or emerging source of systemic risk (consistent with risks identified in the Risk Assessment Matrix).
- Time-dimension risk responses: where rapid credit growth or real estate overheating is identified, recommended tools include tightening LTV or DTI limits or countercyclical capital buffers (examples: Austria, Brazil, Netherlands, Poland, Switzerland).
- Structural/connectivity risk responses: where growing connectivity and complexity or large conglomerates are identified, recommended tools include SIFI capital surcharges, higher capital and liquidity buffers, more intensive supervision, enhanced inter-agency coordination, greater supervisory powers and resources, or widening regulatory perimeter (examples: India, Malaysia).

Selected quantitative and illustrative findings from country case analyses (Box 1):
- Czech Republic: contagion tests simulated a partial loss on exposures to parent banks under a severe double dip scenario that assumed 40 percent of the gross exposure of banks to their parent would be lost; tests revealed a capital shortfall of as much as two percent of GDP under a severe double dip recession.
- Poland: top-down stress tests suggested that up to 30 percent of Polish banks may fail to meet minimum capital adequacy ratios under severe macroeconomic scenarios.

Other country-specific methodological applications highlighted in Box 1:
- Germany: bank-by-bank balance sheet and market-based systemic stress testing; incorporation of Basel III requirements, endogenous bank behavior, and macrofinancial linkages via satellite models.
- Hong Kong SAR: network analysis to illustrate inward and outward spillovers; system judged resilient to significant credit shocks, including those from mainland China.
- India: network analysis showed failure of the two most connected net borrowing banks would give rise to financial contagion in five distinct stages.
- Italy: analysis of domestic and foreign bank exposures suggested limited exposure-to-foreign-bank risk domestically but Italian subsidiaries were systemically important to many Central and Eastern European countries; additional buffers and ECB liquidity support would mitigate transmission risks.
- Korea: stress tests accounting for simultaneous interbank and macroeconomic shocks found banks would remain solvent despite second-round spillover effects from liquidity shortfalls or asset fire-sales.
- Malaysia: interbank contagion stress test using a gross interbank exposures model found simultaneous default of multiple large domestic commercial banks would give rise to severe financial contagion.
- Singapore: network analysis showed greater vulnerability to cross-border interbank exposures (most at risk from U.K. credit shocks) than to domestic exposures.
- Sweden: contagion tests indicated the sector could withstand the default of a major bank but not the joint default of the three largest bank exposures.
- Switzerland: combined stress tests and contagion tests concluded domestic systemic risk was increasing while other sources were stable or decreasing; contagion risk from Swiss banks to other sovereigns had fallen significantly in recent years.
- United Kingdom: nonparametric empirical copula used to construct market perception indicators (Joint Probability of Distress (JPoD) and Conditional Probability of Distress (CoPoD)); indicators suggested implied spillover threat posed to foreign LCFIs by U.K. banks had returned to pre-crisis levels after a sharp rise during the crisis.

### Institutional framework for MaPP
- FSAP assessments increasingly recognize that a strong institutional framework guiding MaPP implementation is essential to ensure MaPP effectiveness, willingness to act, limit biases for inaction, and reduce incentives to use MaPP for non-stability purposes.
- Most FSAPs provide assessments of institutional frameworks, reflecting efforts to establish mechanisms that assign explicit mandates for monitoring, assessing, and addressing systemic risk to safeguard financial stability.
- Consensus has not emerged on whether a single institution (such as the central bank) or a separate financial stability committee/systemic risk council should carry out systemic risk monitoring, regulation, and crisis management.

Representative FSAP recommendations on institutional architecture:
- Several FSAPs (Belgium, Czech Republic, Germany, Netherlands) recommend making the central bank the main MaPP authority, expanding its price stability mandate to include systemic risk monitoring and financial stability, and granting the central bank a clear role to implement tools with appropriate accountability frameworks to enhance willingness to act.
- A number of FSAPs (Austria, Brazil, Chile, China, El Salvador, the EU, Israel, Korea, Malaysia, Mexico, Paraguay, Poland, Sweden, Turkey) recommend assigning the MaPP mandate to an inter-agency body, board, or committee, in some cases with “act or explain” powers for the committee and a leading role for the central bank.
- Other FSAPs are less prescriptive regarding institutional models (examples: Canada, Italy, Switzerland).

Core institutional features emphasized across FSAPs:
- Provide the MaPP authority with appropriate powers to achieve its goals and establish clear goals, mandate, and accountability mechanisms.
- Assign a major role to the central bank, given its expertise in analyzing systemic risk, and protect its independence when it assumes MaPP responsibility.
- Ensure an adequate level of government involvement to help garner political support for MaPP actions.

- FSAPs typically do not provide detailed discussion on modalities of MaPP implementation (for example, the nature of powers—direct control, act-or-explain, opinion/recommendation—or the choice between rules vs. discretion in decision making).

*Italicized source attribution:* _FSAP REVIEW—FURTHER ADAPTATION TO THE POST-CRISIS ERA—BACKGROUND STUDIES (excerpts from the supplied content)._

### Box 2. Selected FSAP Recommendations on the Institutional Framework for MaPP

### Box 2. Selected FSAP Recommendations on the Institutional Framework for MaPP

### Description of the Institutional Framework — Macroprudential authority
- Austria: Set up a new joint committee to be designated as the macroprudential authority, formulating macroprudential policies and setting-related rules—either on its own initiative, or to implement European Systemic Risk Board (ESRB) or ECB decisions. The Committee should be chaired by the Oesterreichische Nationalbank (OeNB), in line with the ESRB Recommendation to give a leading role to central banks, and have a clear legal mandate for policy formulation and rulemaking. In this structure, OeNB should be responsible for systemic risk monitoring and Financial Market Authority for prudential supervision.
  - A joint committee (with a leading role for the central bank).

- Belgium: The National Bank of Belgium is best placed to become the designated macroprudential authority, given its broad supervisory mandate, its statutory financial stability objective, and extensive regulatory powers, which would provide it with willingness to act in a timely manner while having sufficient powers to identify, assess, and mitigate systemic risk. Adequate accountability mechanisms would need to be designed while periodic consultations and regular dialog should take place with other relevant authorities, such as Financial Services and Market Authority and the Ministry of Finance (MOF), on macroprudential and financial stability matters to ensure a holistic view and effective coordination across policies to address systemic risk.
  - The National Bank of Belgium.

- Brazil: Consider establishing a tri-partite Financial Stability Committee (FSC), with the central bank retaining a key role reflecting its expertise in systemic risk analysis. Currently, the central bank assumes de facto responsibility for MaPP, and takes a lead role for financial stability but does not have explicit legal mandate. More explicit cooperation arrangements and greater transparency/accountability is warranted going forward as the system gets more complex and risks arise outside banking. The FSC should include all financial regulators, deposit agency and the MOF and conduct systemic risk monitoring, MaPP coordination, and systemic crisis preparedness.
  - A tri-partite FSC, with the central bank retaining a key role.

- Canada: Provide a clear mandate to a single body to monitor systemic risk to facilitate macroprudential oversight, and to carry out system-wide crisis preparedness. Assigning such a mandate could strengthen the ability of the authorities to respond to emerging problems in the most effective manner. Such a mandate would include the development of an overarching policy and operational guidance to respond to a system-wide crisis and in particular to bring together system wide (both federal and provincial) issues.
  - A single body (unspecified).

- Chile: Ensure that the FSC, established to monitor and identify system-wide risks, recommend policy measures in support of financial stability and coordinate actions in the event of systemic crises, can discharge its functions effectively. The Committee is chaired by MOF and includes the supervisory agencies but includes the central bank as observer. Because the central bank can bring important expertise in monitoring the financial system and discussing policy actions, in its role as observer it should be able to adequately influence the Committee’s decisions while preserving its independence. To preserve regulatory autonomy, the FSC should delegate institution-specific issues to a subcommittee of supervisors.
  - Supports the FSC, with the central bank given the ability to influence decisions in its observer status.

- China: A permanent FSC and systemic risk should be established to enhance macroprudential surveillance. The committee should have access to all relevant supervisory and other financial information, be chaired by a very senior official, and have a clear mandate and authority to identify/monitor the emergence of systemic risks and make recommendations to address them. Consistent with its financial stability mandate, the People’s Bank of China (PBC) should serve as its secretariat and be empowered to provide and receive necessary information, including confidential institution-specific supervisory data. Ensuring the operational autonomy of the PBC and financial supervisors is crucial. FSC would include PBC, three supervisory commissions, the MOF, and other relevant ministries/agencies.
  - FSC (with a secretariat role for the central bank).

- Czech Republic: Elevate financial stability (from being a supporting element of achieving price stability) to a policy objective in the Czech National Bank (CNB) law, to strengthen the CNB’s financial stability mandate. Elevating the financial stability to an explicit objective—without prejudice to the primary objective of price stability—would strengthens the mandate of the CNB on financial stability (including microprudential and macroprudential supervision, as well as crisis management and bank resolution) and establish a stronger accountability framework. Improve the decision-making mechanisms within the CNB to take timely action to address systemic risk by formalizing the mechanism with (a) a pre-set schedule for CNB Board meetings on MaPPs to discuss systemic risks and decide if action is needed; (b) recommendation to be made by the Financial Stability Department on policy actions; and (c) a press statement explaining the nature of the risk and the policy action/inaction to address this risk.
  - The CNB.

- El Salvador: Require that the Central Bank of El Salvador (BCR) and Financial System Superintendency (SSF) coordinate on financial stability issues and that specific policies and procedures for information sharing, systemic risk monitoring, and crisis management be mutually agreed upon, formalized, and implemented by all the safety net providers. The BCR should focus on macroprudential policies and monitoring, the SSF on microprudential issues, and the Deposit Insurance Fund (IGD) on least-cost resolution. A decree should be given that formally establishes an Inter-Institutional FSC and specifies its objective.
  - A joint committee (with an ambiguous role for the central bank).

- European Union: Strengthen macroprudential oversight of the EU financial system by enhancing capacity of the ESRB (which currently lacks binding legal authority and relies on “soft” power), the Single Supervisory Mechanism (SSM), and national authorities, and ensuring their effective coordination. Within the countries of the SSM, ECB will have a role in MaPP, as well as national authorities, as it takes on its microprudential responsibilities. A key challenge for macro-prudential supervision will be to design and calibrate MaPP instruments and implement them against political interference. The independence of the ECB would help in this regard; the national macroprudential authorities also need adequate independence. The ECB should cooperate closely with national authorities to benefit from their local knowledge and with the ESRB in the oversight of non-EMU countries and the nonbank financial sector. It should be responsible for a wide range of instruments beyond those included in the Capital Requirements Directive IV (CRD IV) and the Capital Requirements Regulation (CRR).
  - Systemic Risk Board (with a role for the ECB and MaPP instruments).

- Germany: Define the role of the Bundesbank as macroprudential supervisor, and institute free exchange of information between macro and microprudential supervisors. The Bundesbank as macroprudential supervisor will need to assess domestic, foreign, and regional risk factors, identify SIFIs and define policy toward them. In most of these areas, the Bundesbank will not have decision-making power over most relevant instruments. Considerations could be given to introducing at least an “act or explain” requirement.
  - Bundesbank.

- Israel: Strengthen the institutional framework for macroprudential oversight and policy setting by more formally establishing a FSC and initiating its operations, with the Bank of Israel (BoI) playing a leading role. Stress testing activities should be designed to guide micro- and macroprudential policy to enhance financial stability. Integrated stress testing analysis should be expanded to analyze macroprudential risks and policies (e.g., through risk transfer among corporate groups, insurance companies, savers, and government).
  - FSC (with BoI playing a leading role).

- Italy: Specify ultimate objective of MaPP as per ESRB and designate a MaPP authority, entrust it with tools sufficient to pursue the financial stability mandate. The MaPP authority could be either Bank of Italy (BI) with cooperation arrangements with other supervisors, or a new body with a “comply or explain” mechanism.
  - Either BI or a MaPP body (with a leading role for BI).

- Korea: Establish a dedicated and formal macroprudential council, with a stronger role for the Bank of Korea (BOK), the power to recommend regulatory action from other bodies, and transparency over policy deliberations. Enhanced ability to detect emerging systemic risks and initiate timely responses argues for establishing more formal and dedicated arrangements for macroprudential policy, enhanced transparency and accountability, a greater degree of independence from the political process, clarification and strengthening of the role of the BOK in financial stability, and separation of the macroprudential policy function from crisis management.
  - A Macroprudential Council (with a stronger role for the BOK).

- Malaysia: Formalize a high-level committee involving Bank Negara Malaysia (BNM), Securities Commission, Perbadanan Insurans Deposit Malaysia (PIDM), and the fiscal authority with the responsibility for systemic risk monitoring, information sharing, and crisis action. Formalizing a monitoring/ coordination committee could enhance interagency communications and risk identification. Including financial sector supervisory agencies with significant oversight roles as members would strengthen its effectiveness.
  - A high-level committee (with multiple agencies).

- Mexico: Build on the recently created Financial Sector Stability Council (FSSC) to strengthen the existing macroprudential framework. The Council should enhance coordination among agencies, analyze emerging risks on a regular basis, and put forward recommendations to address risks using a “comply or explain” mechanism.
  - An interagency council; central bank serving as a member.

- Netherlands: De Nederlandsche Bank (DNB) should have the ability to vary macroprudential instruments within a given range in line with emerging economic or financial conditions. The Authority for Financial Markets (AFM) should have clear legal authority to exercise oversight for product development. The ability of DNB and AFM to apply prudential or conduct-of-business rules at a system-wide level is constrained. The DNB and AFM should have greater discretion to put in place enforceable rules. Insufficient rule-making authority leads to ad hoc approaches that risk becoming arbitrary and subject to legal challenge, limiting supervisors’ willingness to act forcefully, on concern of challenging of the request. For implementation of enhanced supervisory regime, supervisory authorities need adequate resources and power.
  - The Netherlands Bank.

- Paraguay: Create an interagency forum to discuss financial stability issues and harmonize prudential norms. A committee of all regulators, including the Instituto Nacional de Cooperativismo (INCOOP), could be created with participation at the highest level. The committee should be chaired by the CBP governor and have a clear mandate, an agenda with concrete deliverables, sufficient technical support teams and the responsibility to report on progress regularly.
  - A committee of regulators lead by the central bank.

- Poland: Systemic Risk Board (SRB), the macroprudential decision-maker to be, responsible for systemic risk identification and analysis, empowered to request that other agencies take action to limit systemic risk, and empowered with tools to control or limit systemic risk) should be provided by law to have independence, accountability to Parliament, and power to make recommendations with an “act or explain” mechanism. It will be important for the National Bank of Poland (NBP) to take a leading role in the SRB, to shield the SRB from political pressure and ensure that MaPP draws on central bank expertise in financial/macroeconomic analysis. It is essential to safe-guard the SRB’s independence and endow it with powers to recommend using tools. The law should require publication of the minutes of SRB meetings and establish accountability.
  - SRB (in which the NBP to take a leading role).

- Slovenia: Expand each supervisory agency’s (Bank of Slovenia (BoS) and insurance and securities regulators) mandate to include macroprudential oversight, with BoS assigned to play a leading role in systemic risk assessment. Currently, the coordinating group, comprising MOF, BoS Governor, directors of the other regulatory agencies, and the group’s secretary, is tasked to identify and monitor systemic risk and consider ways to address them, but while each financial sector supervisory agency contributes de facto to the stability of the financial system, no agency’s statutory framework includes a de jure responsibility to do so. The MaPP authority should be ultimately accountable to national parliament. Macroprudential oversight decisions and their motivations should be made public in a timely manner, taking into consideration the need for confidentiality on specific institutions.
  - Supervisory agencies for banking, securities and insurance sectors (with BoS having a leading role for systemic risk assessment).

- Sweden: Establish a high-level SFSC to focus solely on financial stability and related MAPPs. The SFSC would become the focal point for coordinating financial sector policies and actions across agencies. In setting up the proposed council and to ensure its effectiveness, decisions would be needed on the scope of analysis, range of risks to be addressed, the set of instruments, and institutional/governance frameworks. Consideration could be given to include independent members in addition to the MOF, Riksbank , Finansinspektionen, and the agency for the stability fund and deposit insurance. Member authorities must be represented at the highest level to ensure effective discussions. Given the legal independence of the involved authorities in carrying out their separate mandates, SFSC should be consultative in nature, requiring, to ensure its effectiveness, clear lines of public accountability (possibly reporting to Parliament) for discussions, positions, and commitments expressed by members.
  - A high level SFSC.

- Switzerland: Currently, the Swiss National Bank (SNB) has a financial stability mandate in the context of monetary policy, and the Swiss Financial Market Supervisor Authority (FINMA) has responsibility to protect the functioning of financial markets. There is MOU between FINMA and the SNB and between the federal government, with significant regulatory powers, FINMA, and the SNB. While the present structure seems to have worked well, no clear macroprudential mandate is assigned to any institution. Powers over specific policy instruments are clear, but not where the overarching responsibility for the financial stability outcome lies. Consideration should be given to placing responsibility and powers for MaPPs with one institution or committee. Transparency and accountability could be strengthened by better highlighting the cross agency work to the public.
  - A single body (an institution or a committee).

### Interaction with Other Policies
- 14. FSAPs, acknowledge that MaPP cannot alone be effective in safeguarding financial stability. The interaction of MaPP with other policies is important when assessing the stability of a financial system, because achieving and maintaining financial stability is a shared responsibility. To achieve its goals, MaPP must hence be supported by strong microprudential supervision and enforcement, and complemented by appropriate monetary, fiscal, structural, and other financial sector policies. Effective MaPPs can, in turn, help these policies attain their goals (IMF, 2013). Clear definition and delineation of roles across policies would help limit the risk of MaPP being used (or abused) to achieve non-macroprudential objectives, which other policies would be better placed to address (e.g., fiscal, monetary, or structural policies). In this context, many FSAP teams discuss the need to specify clearly the objectives of MaPPs, as well as bringing out the complementarities and trade-offs their interaction with other policies may present in attaining their respective goals.

- 15. FSAP assessments have increasingly emphasized the importance of a clear delineation of micro- and macroprudential responsibilities. Missions have stressed (e.g., in Poland) that the authorities need to set clear MaPP objectives that are separate from monetary and microprudential policy objectives, emphasizing that the regulatory agency should not use tools to manage the credit cycle but focus on the soundness and safety of financial institutions, while the MaPP authority should focus on reducing systemic risk. FSAPs have also highlighted (e.g., Germany) the need for free exchange of information between micro- and macro-prudential supervisors, since the macroprudential supervisor, in its role, needs to assess domestic, foreign, and regional risk factors and make recommendations that may affect microprudential supervision.

- 16. The importance of micro-macroprudential linkages has been reflected in assessments of supervisory architectures. For example, analyses of the Dutch twin-peaks supervisory approach in the Netherlands FSAP underscored the need to take advantage of the synergy between micro- and macroprudential oversight. Where these responsibilities lie in the central bank, teams have noted (e.g., Armenia) that assigning these responsibilities to separate departments may help address potential conflicts arising from the central bank’s multiple mandates (including price stability, financial stability, regulation, supervision, and crisis management).

- 17. FSAP assessments have also underscored the importance of effective crisis prevention and crisis resolution frameworks to help limit and mitigate systemic risk. Teams had rich discussions on the scope to improve the existing financial safety nets (including emergency liquidity assistance, bank resolution, and deposit insurance) to ensure that they can cope with future shocks and limit moral hazard in increasingly larger and more complex financial systems (e.g., in Brazil). FSAPs have also highlighted (e.g., in the Czech Republic) the importance of close coordination between the MaPP authority and fiscal authorities on crisis management issues, as well as of coordination with peers through participation in cross-border crisis management groups and better access to information on recovery and resolution plans for the banking groups active in the domestic financial system.

*Source: Box 2. Selected FSAP Recommendations on the Institutional Framework for MaPP*

### 18.      At the same time, FSAP assessments have also highlighted the potential trade-offs

### 18.      At the same time, FSAP assessments have also highlighted the potential trade-offs

### Trade-offs between Macroprudential Policy (MaPP) and Crisis Management
- FSAP assessments highlight potential trade-offs between MaPP and crisis management that could undermine financial stability.
- Example: substantial precautionary buffers of government securities that banks are required to hold in India may provide liquidity protection during crisis times but leave banks potentially exposed to unmanaged interest rate risk.
- Where crisis prevention and management responsibilities and MaPP roles rest in the same institution (e.g., Armenia), FSAP teams underlined potential conflicts and emphasized the role of formal arrangements to reduce such conflicts.

### Interactions and Complementarities between MaPP and Monetary Policy
- FSAPs documented possible interactions and both complementarities and tradeoffs between MaPP and monetary policy.
- Example: when interest rate policy may have adverse financial stability implications, central banks may establish a dedicated committee for financial stability issues (e.g., as in Armenia) to delineate MaPP from monetary policy.
- India FSSA example: authorities tightened (or loosened) several macroprudential instruments in conjunction with, and as a complement to, tighter (looser) monetary policy, to reinforce its effect in dampening the credit cycle.
- FSSAs noted that some tools used by central banks do not always have a macroprudential intention and recommended limiting their use to reducing systemic risk while using conventional policies, such as monetary policy, for macro management.

### Interactions with Fiscal Policy and Market Infrastructure
- FSAPs captured potential complementarities between MaPP and fiscal policy.
- Examples:
  - Use of tax instruments to address systemic risk: stamp duty in Hong Kong SAR to cool the housing market when overheating pressures reflect demand factors.
  - Tax features contributing to systemic risk: tax deductibility of interest payments in the Netherlands and Switzerland; the special tax on banks in Slovenia.
- Market infrastructure policies discussed less frequently but noted for improving systemic liquidity risk management and limiting counterparty risk (examples: CCPs, standardization of OTC derivatives in Hong Kong SAR, India, Spain, Singapore).
- Structural policies such as competition policy featured less, though some FSAPs highlighted their role in facilitating access to alternative funding to reduce systemic risk from large complex conglomerates.

### Multilateral and Cross-Border Aspects of MaPP
- The multilateral aspects of MaPPs have received relatively less attention in FSAPs despite growing importance in interconnected financial systems.
- Limited coverage of spillover effects, leakage, and boundary effects of unilateral national regulatory actions reflects ongoing international deliberations.
- Several FSAPs (e.g., Poland, Australia, the EU, Malaysia, Mexico, and the U.K.) discuss the need for coordination to limit cross-border arbitrage, including reciprocity arrangements across countries.5
- FSAPs highlight importance of close supervisory coordination to address implications of policies imposed by one country on others (examples cited: Austria, Australia, the EU, France, Poland, Singapore, Switzerland, and the U.K.).
- Some FSAPs comment on (insufficient) evidence of adverse spillover effects of policies aimed at external sources of systemic risk (example: capital flow management measures targeted at financial stability in Brazil).

### Home-Host Supervisory Coordination and Cross-Border Banking Groups
- Home-host supervisory coordination is a principal area of FSAP focus, reflecting increased cross-border issues since the 2007 global financial crisis.
- FSSAs emphasize formalizing cooperation via Memoranda of Understanding (MOUs), active participation in supervisory colleges and crisis management groups (CMGs), and information—and burden—sharing arrangements.
- Active coordination is underscored for host countries with systemically important G-SIFI operations (e.g., Albania, the Czech Republic, Ghana, Poland, Singapore) and for home countries through sharing supervisory information and joint on-site inspections (e.g., India, the Netherlands, Sweden, the U.K.).
- FSAPs stress group-wide supervision, regular information exchanges, joint assessments, monitoring of intra-group exposures, and cross-border resolution agreements.

### Conclusions and Implications for Future FSAPs
- Review of 60 FSSAs completed during 2010–14 suggests systemic risk assessment, MaPP frameworks, and their role in financial stability have become important FSAP features.
- Recent FSAPs include explicit MaPP discussions, some with dedicated technical notes; teams used extensive systemic risk monitoring and advanced techniques to capture time and structural dimensions of systemic risk.
- The depth, breadth, and integration of MaPP assessments vary across countries and over time:
  - FSSAs for advanced economies and more advanced emerging markets had more comprehensive coverage.
  - More recent FSAPs had more systematic treatments and better integration of MaPP into overall stability analysis.
- Recommended improvements for future FSAPs:
  - Better integration of MaPP analysis into the stability analysis, including reflecting key messages in executive summaries and FSAP recommendations and cross-referencing key findings throughout FSSA reports.
  - More systematic treatment of MaPP frameworks, disseminating best practices and operationalizing the framework for MaPP advice developed in the Key Aspects paper (and forthcoming Staff Guidance Note on Macroprudential Policy).
  - Use systemic risk as the organizing concept by treating MaPP issues across the three pillars of stability assessments: main threats to systemic stability; monitoring, regulation, and supervision of these risks; and authorities’ ability to mitigate or respond to risks.
  - Assess authorities’ capacity to operate MaPP frameworks effectively, including multilateral and calibration aspects, costs and benefits, and cross-border spillover effects—particularly important as AEs normalize unconventional monetary policies.

### Box 3 — Selected Treatments of Foreign-Bank Branches and Subsidiaries (highlights)
- Albania: central bank converted foreign bank branches into subsidiaries as a MaPP measure; recommended support letters from mother banks and formalized cooperation with foreign supervisors.
- Australia: subsidiaries of major Australian banks account for 90 percent of New Zealand banking assets; subsidiaries are ring-fenced; recommended enhancing cross-border coordination arrangements.
- Czech Republic: prescribed intra-group exposure limits and enhanced cross-border cooperation to address contagion channels from foreign parents to Czech subsidiaries.
- El Salvador: regulatory limits on foreign asset exposure to a certain percentage of capital; noted liquid assets provide self-insurance at the cost of profitability and forgone credit.
- Finland: conversion of subsidiaries into branches would limit supervisory effectiveness; recommended enhanced home-host cooperation, information exchange, burden sharing, harmonization, and common reporting.
- Ghana: high degree of foreign-bank subsidiary ownership; highlighted failures in due diligence when issuing licenses and the need to correct supervision weaknesses.
- Hong Kong SAR: foreign-bank branches account for 35 percent of total local banking system assets and 27 percent of domestic lending; authorities encourage subsidiarization and have robust bilateral arrangements for confidential supervisory information exchange.
- India: potential change to allow foreign banks to incorporate as subsidiaries; Reserve Bank of India defers to foreign supervisors of overseas offices and subsidiaries for hands-on supervision; absence of formal arrangements for information exchange with host supervisors noted as a serious gap.
- Luxembourg: crisis exposed vulnerabilities from large cross-border exposures to foreign-parent banks; recommended monitoring parent-bank exposures, readiness to limit unsound exposures, and strengthening supervisory collaboration.
- Mexico: seven large banks dominate, five are foreign-owned subsidiaries; recommended MOUs, functioning supervisory/crisis management colleges, crisis-simulation exercises, and reciprocal legal/regulatory knowledge.

*Source: _081814b - 18.      At the same time, FSAP assessments have also highlighted the potential trade-offs*

### Box 3. Selected Treatments of Foreign-Bank Branches and Subsidiaries in FSAPs (concluded)

### Box 3. Selected Treatments of Foreign-Bank Branches and Subsidiaries in FSAPs (concluded)

### National findings and supervisory recommendations
- Netherlands
  - Finding: The Dutch central bank frequently relied on supervision performed by host supervisors, although it had the necessary legal and regulatory powers to engage in consolidated supervision of cross-border banking groups.
  - Finding: The activities of Dutch banks’ foreign subsidiaries were significant enough to threaten the viability of their entire banking groups should they fail.
  - Recommendations:
    - Dedicate additional resources for more intrusive monitoring of the activities of Dutch bank subsidiaries located abroad.
    - Revamp supervision and information sharing with foreign supervisors to obtain timely information about the contagion risks that host subsidiaries’ operations pose to their home banks.
    - Participate in supervisory colleges to avoid risk of regulatory arbitrage and a fragmented view of risk.

- Poland
  - Finding: High degree of interconnection between the country’s banking system and the rest of the European financial system.
  - Recommendations:
    - Fully utilize the “reciprocity” principle agreed upon by EU institutions (e.g., the ESRB), thereby ensuring that measures taken by Polish authorities would be applied to EU-bank branches operating in Poland.
    - Intensify supervision of foreign-bank branches’ risk management practices as a general precaution.

- Singapore
  - Findings:
    - 116 of the country’s 122 banks are foreign-bank branches that collectively own 65 percent of banking system assets.
    - Operational constraints on branches: limits to the number of branches with retail access; wholesale branches allowed to take wholesale domestic funding but not retail; purely offshore branches generally not allowed to accept deposits in Singaporean dollars from residents.
    - Branches are subject to the same liquidity requirements as domestic banks, and the Monetary Authority of Singapore (MAS) can impose corrective and remedial action on branches.
    - Authorities moved to convert large retail branches into domestically incorporated subsidiaries to address potential spillover risks from parents.
    - Authorities highlighted efforts in international fora for greater sharing of supervisory information on G-SIFIs.
  - Risks discussed: parent instability, challenges of supervision/resolution, and whether branches pose material risk to financial stability.
  - Recommendations:
    - Asset maintenance ratios imposed by the MAS to limit the risk arising from branches should be complemented with a strict licensing policy, monitoring parent-bank risks, and international supervisory coordination.

- Spain
  - Analysis: Cross-border analysis examined the impact of ring-fencing by host authorities of major Spanish banks, where host supervisors’ unilateral ring-fencing of bank affiliates may prevent transfer of excess subsidiary resources within a banking group.
  - Finding: Countries such as Spain, whose two largest banks have sizeable foreign operations outside of the EU, are particularly susceptible to spillover effects from ring-fencing of their domestic banking groups’ foreign subsidiaries.
  - Result: The analysis found that the banks could be affected somewhat by such actions under adverse conditions and their capital ratios would be lower but the impact would be limited.

- U.K.
  - Finding: Domestic authorities’ powers varied greatly depending on whether a financial institution was classified as a branch of an EU-based parent bank or a U.K.-based subsidiary.
  - Recommendations:
    - Enhance collaboration between home and host authorities, including joint inspections and decision-making processes.
    - Increase the amount of information shared via supervisory colleges and bilateral contacts.
    - Pursue financial reforms related to ring-fencing and standalone subsidiarization of EU-bank branches should such reforms fail to pass at the EU level.

### Evenhandedness of FSAPs — use and distribution of FSAP resources (high-level analysis and statistics)
- Context and objective:
  - The paper offers benchmarks for assessing FSAP practices and analyzes recent FSAPs (published and unpublished) over the past five years to examine evenhandedness.
  - Sample: 82 FSAPs for FY2010–14 (list in Appendix Table 1 in source).

- Change to mandatory FSAPs:
  - In September 2010, the Board made financial stability assessments under the FSAP mandatory for 25 systemically important jurisdictions (SIs).
  - The list expanded to 29 jurisdictions in 2013.

- Access and resource implications:
  - When a member formally requests an FSAP, that request is generally met, though sometimes with some delay; staff reported no case of “denied access to FSAP.”
  - Allocation concern: More FSAP resources allocated toward SIs implies fewer resources for other member countries; the Board was informed of this implication when deciding to make FSAPs mandatory for a subset of members.

- Distribution across country groups (comparison FY2005–09 vs. FY2010–14)
  - SIs’ FSAP share: increased from 44.0 percent in FY2005–09 to 100 percent in FY2010–14.
  - Non-SI EMs’ FSAP share: declined to 41.3 percent in FY2010–14.
  - LICs’ FSAP share: declined to 27.6 percent in FY2010–14.
  - Implied average FSAP intervals (Table 2; in years):
    - Total: based on FY2005–09 = 10.2; based on FY2010–14 = 11.9
    - SIs: based on FY2005–09 = 11.4; based on FY2010–14 = 5.0
    - Non-SI AEs: based on FY2005–09 = 12.0; based on FY2010–14 = 10.0
    - Non-SI EMs: based on FY2005–09 = 7.2; based on FY2010–14 = 12.1
    - LICs: based on FY2005–09 = 11.9; based on FY2010–14 = 18.1
  - Example calculation note from source: The average FSAP interval for non-SI EMs is calculated as follows: 12.1 = 100/41.3*5. 100/41.3 is the number of five-year cycles it takes to conduct an FSAP for all EMs. Multiplying it by 5 gives the average FSAP interval years.

- Regional distribution (FY2005–09 vs. FY2010–14)
  - Observed shifts:
    - AFR and MCD shares dropped very substantially.
    - APD share jumped up from a low level.
    - EUR consistently well-represented.
    - WHD had the highest share in FY2010–14, close to 60 percent.
  - Hypothetical FSAP shares (weighting countries by FSAP shares of country groups shown in Figure 4) implications:
    - Expected FSAP incidence over a five-year period: AFR and MCD about 30 percent; EUR more than 60 percent.
    - Actual use in FY2010–14 was more lopsided than hypothetical: AFR and MCD actual use was lower than hypothetical; WHD’s over-representation was not removed by weighting.
  - Translated into interval years per source note:
    - AFR and MCD countries receive an FSAP every 17.3 years and 15.5 years on average, respectively.
    - EUR countries receive an FSAP every 8.1 years on average.

- Issue identified for Board consideration:
  - Given the substantial fall in FSAP frequency for non-SIs, possible responses include expanding FSAP resources, reducing SIs’ FSAP frequency, or simplifying SI FSAPs to release resources. The paper presents this as an issue for further Board consideration.

### BCP Assessments and comparability caveats
- Role: BCP assessments are widely included in FSAPs and provide grades such as “compliant” and “largely compliant,” allowing cross-country analysis.
- Caveat: BCP assessments are not exactly comparable across countries due to the Basel Committee’s “proportionality principle” and other factors; the Basel Committee cautions that “Seeking to compare countries by a simple reference to the number of ‘Compliant’ versus ‘Non-Compliant’ grades they receive is unlikely to be informative.” The principle requires taking into account jurisdiction-specific features and context, such as complexity and systemic importance, in assessments.

*Source: Box 3 and accompanying sections in the provided FSAP Review background study PDF.*

### 43.      Yet, it seems wrong to completely foreclose the possibility that these grades can be

### _081814b - 43.      Yet, it seems wrong to completely foreclose the possibility that these grades can be 

### Comparability and methodology of BCP assessments
- Assessments are treated as fully comparable for the purposes of the analysis.
- Conversion of qualitative grades to numeric scores: compliant = 4, largely compliant = 3, materially non-compliant = 2, non-compliant = 1.
- For each country, a simple average of scores on all principles is computed.
- Sample: BCP assessments conducted during FY2010-14, altogether 69 assessments (published and unpublished).  
  - OFCs in sample: 6 (but many analyses exclude OFCs when benchmark data are unavailable).  
  - For the principle-level dispersion exercise: exclude five assessments based on the new BCP and focus on the remaining 64 assessments.
- Country groupings used: systemically-important AEs (SI/AEs); systemically-important EMs (SI/EMs); non-SI AEs; non-SI EMs; LICs; OFCs.

### Distributional patterns and publication bias
- Visual overview (Figure 7) findings:
  - SI/AEs and SI/EMs have received high average scores.
  - LICs are dispersed in both average and standard deviation of scores.
  - Published vs. unpublished: relatively unfavorable assessments tend to remain unpublished, giving rise to a bias in the public message from BCP assessments.

### Benchmarks and correlation analysis
- Benchmarks tested: per capita income, Corruption Perception Index (Transparency International), Ease of Doing Business Index (World Bank).
- Visual inspection (Figure 8) findings:
  - Average BCP scores are correlated with all three benchmarks.
  - Country clustering: AEs, EMs, LICs form clusters; dispersion smallest for AEs, largest for LICs.
  - SI/EMs appear to score better than others conditioned on benchmarks.

### OLS regressions (Table 3): Yi = α + βXi + εi — Overall sample and sub-samples
- Regression specification: Yi = α + βXi + εi.
- Overall sample coefficients:
  - Per capita income (log): α = 1.298***, β = 0.207***
  - Corruption Perception Index value: α = 2.382***, β = 0.0159***
  - Ease of Doing Business Index ranking: α = 3.614***, β = -0.00630***
- Sub-sample: SI/AEs + AEs:
  - Per capita income (log): α = 0.258, β = 0.307**
  - Corruption Perception Index value: α = 2.902***, β = 0.0087**
  - Ease of Doing Business Index ranking: α = 3.566***, β = -0.00242
- Sub-sample: SI/EMs + EMs:
  - Per capita income (log): α = 2.077*, β = 0.129
  - Corruption Perception Index value: α = 2.803***, β = 0.0108
  - Ease of Doing Business Index ranking: α = 3.407***, β = -0.00227
- Sub-sample: LICs:
  - Per capita income (log): α = 2.143***, β = 0.084***
  - Corruption Perception Index value: α = 2.384***, β = 0.0104
  - Ease of Doing Business Index ranking: α = 3.253***, β = -0.00444*
- Notes from the source:
  - *, **, and *** imply ten percent, five percent, and one percent significance, respectively.
  - Since a smaller number represents a better status in the EoDB ranking, the β coefficient in the EoDB regression is estimated to be negative.
- Illustrative interpretation from PCI benchmark:
  - A country with $20,000 PCI (AE) is expected to have an average BCP score higher than a country with $1,000 PCI (LIC) by 0.6 (i.e., slightly more than half a notch in the four-notch grading scale). This difference may reflect the “proportionality principle” embedded in BCP.

### Z-score analysis and group-specific observations
- Z-score definition used: (y – y*)/σ, where y is actual BCP score, y* predicted BCP score from regression, σ estimated standard deviation of regression error.
- Observations from z-score plots (Figure 9):
  - SI/EMs are rated better as a group and individually; one SI/EM has z-scores for Corruption Perception Index and Ease of Doing Business Index exceeding two (a possible threshold for abnormal observations).
  - AEs and non-SI EMs’ z-scores are balanced around zero, with no abnormal observations.
  - LICs’ z-scores are slightly biased negative; one conspicuously positive observation (per capita income and Corruption Perception Index); four LICs recorded abnormal observations for some or all benchmarks.
- Structural-break checks:
  - A Chow test found a structural break between the SI/EM sub-group and the rest with one to five percent significance, and between the LIC sub-group and the rest with one to ten percent significance.
  - For AEs and non-SI EMs, the null hypothesis of no structural break cannot be rejected.

### Contribution of individual principles to cross-country dispersion (Figure 10)
- Method: compute average and standard deviation of BCP scores for each principle; highlight principles with high standard deviations (yellow circle) as main contributors to cross-country differences.
- Principles with highest standard deviations across countries (descending order):
  - Principle 12 Country and transfer risks
  - Principle 16 Interest rate risk in the banking book
  - Principle 1.5 Legal protection
  - Principle 13 Market risks
  - Principle 15 Operational risk
  - Principle 18 Abuse of financial services
  - Principle 24 Consolidated supervision
- Interpretation:
  - Some items (e.g., Principles 1.5 and 18) may link to institutional strength (as proxied by benchmark indicators).
  - Technical skills of supervisory authorities and banks in monitoring, assessing, and managing risks (Principles 12, 16, 13) may also be correlated with benchmarks.
  - Question posed: are countries with particularly high (or low) z-scores notably stronger (weaker) on these principles than predicted by benchmarks?

### Assessor identity, distribution of assessors, and assessor effects
- Assessments conducted by two experts in principle; sometimes three.
- Sample counts: 63 assessments in one part of the sample (68 experts involved across those 63 assessments); Table 4 counts number of assessments per assessor for FY2009–14:
  - Number of assessments per person → Number of assessors:
    - 7 → 1
    - 6 → 2
    - 5 → 1
    - 4 → 6
    - 3 → 6
    - 2 → 12
    - 1 → 40
  - Four assessments were done by pairs of one-time assessors (zero continuity).
  - Wide distribution of assessors complicates cross-country and over-time comparisons.
- Assessor dummy regressions (Table 5): regression Yi = a + bXi + cADj + εi where ADj is dummy for particular assessor j.
  - Tested only for four assessors who did five or more assessments over past five years.
  - Table 5 reported estimated assessor dummy effects (significance indicated by ** and *** for 5 percent and 1 percent):
    - Assessor A (7): PCI 0.373**, CPI 0.467***, EoDB 0.574***
    - Assessor B (6): PCI 0.235, CPI 0.169, EoDB -0.020
    - Assessor C (6): PCI -0.399**, CPI -0.235, EoDB -0.275
    - Assessor D (5): PCI 0.026, CPI -0.114, EoDB 0.121
  - Conclusion: average assessment scores can be influenced by assessor identity, sometimes as much as by half a notch in the four-notch BCP rating scale; suggests need for greater standardization of assessors’ judgments.

### Stress testing: overview and institutional perimeter
- Stress testing is a core part of FSAP for assessing financial system risks from a macroprudential viewpoint.
- The IMF (2012) “Macrofinancial Stress Testing—Principles and Practices” offers seven “best practice” principles; the review selects four principles most pertinent to risk propagation and examines published FSSAs completed in 2009–14 for consistency.
- Principle No. 1: “Define Appropriately the Institutional Perimeter for the Tests”
  - Institutional perimeter definition should consider interconnectedness, size, substitutability, and complexity.
  - Network analysis can help identify relevant financial institutions.
  - The Global Financial Crisis showed the significant role of non-banking financial system (shadow banking institutions and activities).
- Review findings on perimeters (Figures 11 and 12):
  - Institutional perimeter differs substantially across countries.
  - Analysis of interconnectedness is limited to SIs, AEs and a few others; LICs are entirely missing.
  - Stress tests of non-bank financial institutions are concentrated on SIs, AEs, and OFCs; EMs and LICs are almost missing.
  - Most non-bank stress tests focus on insurance companies and some long-term non-bank financial institutions; many shadow banking entities (investment funds, finance companies) and shadow banking activities (securities lending, repos of securitized products) are left aside.

*Source: Authors’ estimates; material excerpted from the IMF FSAP Review—Background Studies (FY2010–14 sample described in the source).*

### 60.      The difference across countries in the institutional perimeter may well be explained by

### _081814b - 60.      The difference across countries in the institutional perimeter may well be explained by

### Institutional perimeter, non-bank coverage, and shadow banking
- Finding: Differences across countries in the institutional perimeter may reflect the structure of the recipient country’s financial system or lack of data.
- Practice: Where non-bank activities are at an infancy stage, conducting stress tests on them may be unnecessary.
- Observation: There may be scope to expand stress test perimeter in some AEs and EMs.
- Evidence: The FSB’s monitoring report shows large asset shares of shadow banking entities in both AEs and EMs; substantive stress testing covering shadow banking is found only in the United States (U.S.) FSSA.
- Figure notes: Radar-chart-based shares are shown for FSAPs that have analysis of interconnectedness and for FSAPs that covered non-bank financial institutions (percent scales annotated 0.0, 20.0, 40.0, 60.0, 80.0 for Figure 11; 0.0, 10.0, 20.0, 30.0, 40.0, 50.0 for Figure 12).

### Principle No. 2 — Identify All Relevant Channels of Risk Propagation (cross-border spillovers and single-factor limitations)
- Finding: Cross-border spillover and contagion analysis has been conducted only for a small number of countries, heavily concentrated on SIs and a few others.
- Illustration: Recent global events (e.g., tapering of the U.S. Federal Reserve’s unconventional monetary policy) demonstrate that spillover risks are not limited to SIs.
- Practice gaps:
  - In OFCs and LICs there are some cases where only single-factor risk analysis has been conducted.
  - “Single-factor risk analysis” is defined as applying a single risk factor to a static bank model, not capturing synergy of multiple risks or time variation.
- Concern: Highly selective application of spillover and macro scenario analysis across countries may violate evenhandedness.
- Figure notes: Radar chart shares for FSAPs with spillover stress tests (percent scales annotated 0.0, 10.0, 20.0, 30.0, 40.0, 50.0 for Figure 13) and for countries with only single-factor tests or no stress tests (percent scales annotated 0.0, 10.0, 20.0, 30.0, 40.0, 50.0, 60.0 for Figure 14).

### Policy feedbacks and implications for stress-testing practices
- Concept: “Policy feedbacks” — authorities’ policy reaction (or lack thereof) affects the nature and extent of risk propagation.
- Finding: Numerous survey responses indicate stability assessments often do not incorporate effects of mitigating policy actions; perceptions of shocks may be exaggerated as a result.
- Example: One FSSA noted that ongoing restructuring meant one third of banks in the stress test sample no longer exist as stand-alone entities, and merged entities’ financial strength may differ from the sum of individual parts — not captured by stress tests.
- Implication: Practices that ignore policy feedbacks run counter to the spirit of capturing policy-induced propagation and mitigation.

### Principle No. 3 — Include All Material Risks and Buffers (data gaps and use of public data)
- Finding: Capturing a variety of risks requires necessary financial data; data gaps are a potential obstacle.
- Evidence:
  - Figure 15 shows FSAPs in which data gaps are explicitly mentioned; insufficient data are common for AEs, EMs and LICs.
  - Breakdown (Figure 16) shows cases where only public data were provided to FSAP teams.
  - Where teams could not conduct their “own” stress tests, this accounts for 12 percent of SIs and 4 percent of AEs among countries citing data gap issues.
- Consequence: Data gaps hinder full assessment of the extent of risks and their permeation through economic and financial systems (see Box 4).
- Evenhandedness risk: Authorities’ willingness or ability to offer confidential source data can undermine evenhandedness — some countries may be able to hide behind data insufficiency.

### Box 4 — Data constraints (selected illustrative cases)
- China (FSAP conducted 2010; executive summary published November 2011): Top-down stress tests by authorities used supervisory data; FSAP team used public data only. The FSSA noted: “A full assessment of the extent of these risks [multiple shocks occurring at the same time] and how they could permeate through the economic and financial system, however, was hindered by data gaps, the lack of sufficiently long and consistent time series of key financial data, weakness in the informational infrastructure, and constraints on the FSAP team’s access to confidential data.”
- Typical data gaps cited: sectoral exposures and entities banks lend to; off-balance sheet positions and operations of banks; short time series with structural breaks; constraints on access to confidential data.
- Other countries referencing data gaps include many LICs (Mozambique, Tanzania, El Salvador, Paraguay, Rwanda) and some non-SI EMs (Romania, Serbia).

### Capital buffer measurement and case-by-case adjustments (Box 5)
- Finding: Stress test quality depends on accurately capturing financial system buffers; some FSAP teams suspected over-reporting of capital adequacy and attempted judgmental adjustments.
- Recommendation: Rather than leaving adjustments entirely to team discretion, staff should produce a note presenting actual and hypothetical cases of under-reporting and explain possible adjustments to foster a common understanding, increase transparency, and enhance evenhandedness.
- Box 5 examples:
  - Russian Federation (FSAP 2011): Adjustments for regulatory forbearance and low provisioning (raising required provisions to midpoint of regulatory range; assuming poor-quality collateral has no value) — team estimated adjustments could wipe out as much as one-third of total bank capital; team admitted adjustments were somewhat ad hoc and arguably extreme.
  - Rwanda (FSAP 2011): As of end-September 2010, the banking industry had 12.8 percent of core capital-to-asset ratio on average, and 20.3 percent of regulatory capital-to-asset ratio. By correcting under-provisioning and imposing general/watch provisions, the core capital ratio dropped to 11.6 percent and the regulatory capital ratio to 18.5 percent. With conservative assumption that all collateral held against classified loans becomes worthless, core capitalization declined to 8.3 percent and regulatory capitalization to 13.3 percent; as a result, two banks fell below 6 percent core capitalization with one of them falling below three percent.

### Principle No. 5 — Focus on Tail Risks (rule of thumb, questions, and alternative methods)
- Current rule of thumb: A two standard deviation (2SD) shock on real GDP growth over a period of two years applied to baseline real GDP projection by IMF (WEO or Article IV adjusted by FSAP teams and the authorities).
- Practice: The standard deviation of real GDP is usually calculated using 20–30 years of past data. No strict guideline imposes adherence to this rule; practice differs across countries.
- Concerns about the current use of 2SD:
  a) Appropriateness of using historical mean GDP growth over two years as central value if mean shifts over time or if serial correlation exists.
  b) Appropriateness of applying 2SD shocks to the economy’s current position (e.g., immediately after a crisis), potentially penalizing countries in recession at FSAP time.
  c) Choice of sample period for calculating standard deviation: inclusion/exclusion of past tail episodes can substantially influence SD and thus the shock magnitude.
- Proposed alternative methods (rules of thumb) — aim to standardize macro stress scenarios across countries while being simple and reflecting a uniform low-probability event and recent economic conditions:
  - Common steps: Decompose real GDP into trend and cyclical component C_t using HP filter.
  - Method (i): Calculate standard deviation of C_t (SD_C) from past data, and subtract 2SD_Cs from trend GDP expected in one to two years from the time of FSAP.
    - Note: Influence of recent developments is not zero because they affect HP-filtered trend.
  - Method (ii): Estimate an autoregressive (AR) model of C_t; produce one-year and two-year forecasts for C_t (C_{t+1}^e and C_{t+2}^e). Construct stress scenarios by subtracting 2SDs of the error term of the AR model from C_{t+1}^e and C_{t+2}^e.
    - Under Method (ii) recent economic conditions have greater influence as they set the starting point of the projection.
- Comparison findings (Table 6 summary insights):
  a) Under the alternative methods, shocks are smaller than under the current rule of thumb because trend fluctuations are removed when estimating volatility and mean-reversion of real GDP’s cyclical factor is taken into account.
  b) Discrepancies between alternative methods and current rule tend to be larger when an economic downturn is projected than when a recovery is projected; current rule may amplify shocks during downturns.
  c) Shocks actually used by FSAP teams are generally milder than the current rule of thumb and closer to the alternative methods, but deviations are often ad hoc and case-by-case.
- Notes on comparisons and samples:
  - “Rule of thumb” in Table 6 is 2SD shocks where SD is calculated from 30 years of historical data, applied to the FSAP team’s forecast for real GDP growth for 2 years.
  - Sample periods for calculating SD_C (Method (i)) and estimating an AR model (Method (ii)) are from 1980 to the year when the FSAP was conducted.

*Source: Excerpt from FSAP Review Background Studies (text and figures as provided).*

### 74.      Finally, regarding the period over which standard deviations are to be calculated, we

### 74.      Finally, regarding the period over which standard deviations are to be calculated, we

### Issues with historical sample period for SD calculation
- Some FSAP teams apply ad hoc treatments to the period over which standard deviations (SDs) are calculated, producing inconsistent SD estimates across teams.
- Example: For APD EM F, including the Asian Crisis episode in the sample makes the estimated SDs unstable depending on the length of sample periods (Figure 17).
- Team practices cited:
  - One team did not calculate SDs from historical data for a given case.
  - Another APD EM team calculated SD over 12 years that do not include the Asian Crisis episode but used a 2.6SD shock, severer than the current rule of thumb.
  - Some teams for European Debt Crisis countries (e.g., Europe SI B) used a very long sample period (30 years) to dilute the impact of the recent major shock.
- Concern: Different teams use different compromise solutions, raising issues of consistency and evenhandedness.

### Guidance on forward-looking stress-test design and SD standardization
- Stress test design should be intrinsically forward-looking, but historical data use is inevitable due to forecasting difficulty.
- Objective: use historical data in a way less affected by past crisis experience while utilizing recent information.
- Proposed elements of any uniform rules:
  - (a) standardize the meaning of two standard deviations across all countries using some stochastic forecasting model; and
  - (b) reduce the influence of when an FSAP happens to take place.
- Position on alternative “rules of thumb”:
  - The authors believe their alternative "rules of thumb" offer a reasonable uniform benchmark.
  - Case-by-case adjustments based on team’s and authorities’ forward-looking judgments are not excluded, but:
    - A scenario based on uniform rules should be constructed and presented in the FSSA.
    - Any judgmental deviation from it must be clearly explained and justified.

### Findings from analysis of FSAPs over the last five years (evenhandedness assessment)
- Scope: analyzed FSAPs conducted over the last five years across three aspects: members’ use of FSAPs; BCP assessments; and stress testing.
- Key patterns and findings:
  - FSAP usage:
    - Board decision to designate systemically important (SI) jurisdictions and make FSAP mandatory every five years led to evolving FSAP use.
    - Average interval between FSAPs is expanding for LICs and non-SI EMs.
    - Geographic use patterns: AFR and MCD under-represented; WHD over-represented.
  - BCP assessments:
    - Correlated with standard indicators of countries’ technical or institutional strength (benchmarks), though correlation does not imply explanation.
    - SI EMs appear to be assessed better than might be expected from benchmark indicators.
    - Evidence that assessments are influenced by assessor identity; differences across assessors can be fairly large.
  - Stress testing:
    - No common framework exists for determining key macro assumptions that underlie stress testing for each country.
    - There is a "rule of thumb", but teams deviate significantly depending on circumstances.
    - Authors proposed two alternative methods aimed at standardizing the likelihood of stress events across countries and compared them with the "rule of thumb" and actual FSAP scenarios.
    - Finding: actual scenarios being used are more moderate than the "rule of thumb" and closer to the authors' alternative methods.
    - Concern: frequent ad hoc, case-by-case adjustments required to agree with authorities raise questions about whether all members face stress events whose likelihoods are broadly the same.
    - Scope of stress testing differs substantially across countries; highly selective application of spillover and other macro scenario analysis appears inappropriate given the variety of risks and propagation channels.
    - Data insufficiency contributes to selectivity; efforts needed to improve coverage and quality of data or authorities’ cooperation in data provision within legal constraints.

### Conclusions and suggested areas for further study
- Overall assessment: Findings do not necessarily imply lack of evenhandedness but suggest room for improvement and areas for further study.
- Two areas recommended for exploration:
  - BCP assessment substance review:
    - Assign experts to examine the substance of BCP assessments (beyond average scores) to corroborate or contradict apparent biases from aggregate numbers.
    - Examine how the recently revised BCP, which introduced macroprudential elements based on the GFC experience, is being assessed to ensure even coverage in all FSAPs.
  - Standardized simple models for stress scenarios:
    - Systematic search for simple models usable by all FSAP teams to construct stress scenarios to minimize ad hoc adjustments.
    - Model choice requires clarity on what kinds of stress events the Board wants to test (both degree and nature of risks) to select appropriate models.

### Appendix: Two alternative methods for rule-of-thumb stress testing (technical notes)
- Alternative Method (i):
  - Assumes GDP moves back to a stochastic trend.
  - During the stress period, a 2SD shock of the cyclical component of the historical GDP path hits the trend GDP expected in one to two years.
  - Does not utilize information about recent GDP developments.
  - Procedure summary:
    - Divide real GDP (Yt) into trend GDP (Gt) and cyclical component (Ct) using HP filter; use FSAP teams’ GDP forecasts at the time of the FSAP to avoid HP’s end-point problem.
    - Divide Ct by trend GDP (Ct/Gt) and calculate its standard deviation from thirty years of past data.
    - Shock = 2 × that standard deviation; subtract from Gt in two years.
- Alternative Method (ii):
  - Assumes GDP moves around a stochastic trend and forecasts future GDP by estimating an AR model of the cyclical component.
  - Utilizes recent GDP developments and the team’s baseline scenario in the near future (team’s baseline used to avoid HP end-point problem; the method’s baseline is produced from the AR model).
  - Procedure summary:
    - Divide real GDP into trend and cyclical components as in Method (i).
    - Estimate AR(2) model for the cyclical component εt:
      - εt = α εt−1 + β εt−2 + ut
    - Calculate conditional forecasts for εt+1 and εt+2 using estimated parameters (α*, β*).
    - Stress scenario derived by subtracting from Yt+1 and Yt+2 two standard deviations of the AR(2) error term.
    - Standard deviation at t+1 = standard error of the regression (σ).
    - Standard deviation at t+2 computed from model dynamics; with Durbin-Watson statistics close to two, serial correlation is disregarded.
    - Shock to Y two periods ahead = two times the square root of α*(α* + 1) σ^2 / ??? (formulaic derivation preserved in source).

### Selected survey and coverage statistics from the review
- Surveys:
  - National authorities of the 168 countries, territories, and currency unions that had an FSAP by early 2014 were invited; responses received from 55, a response rate of 33 percent.
  - Response composition:
    - About half of respondents had a second or third FSAP (or stability module) after the 2009 FSAP review; about a fifth had a first-time FSAP during the same period.
    - Most responses came from central banks; the rest from ministries of finance.
  - IMF Executive Directors: 24 surveyed, 8 responses received (33 percent completion rate).
- Response rates by income group (from note):
  - Advanced countries: 56 percent (24 out of 43 eligible).
  - Emerging countries: 32 percent (28 out of 88 eligible).
  - Low income countries: 8 percent (3 out of 37 eligible).
- Perceptions:
  - Over 90 percent of country authorities and all Executive Directors see FSAPs as adequately structured around the three main pillars.
  - National authorities’ satisfaction with safety nets treatment: 75 percent.
  - Satisfaction with articulation of main threats to financial stability: 69 percent for country authorities; 88 percent for Executive Directors.
  - Large majority view FSAPs as covering most important issues with adequate analytical depth: country authorities, 76 percent or more; Executive Directors, 88 percent or more.
  - Country authorities’ satisfaction with FSAPs taking account of countries’ unique features: 57 percent.

*FSAP REVIEW—FURTHER ADAPTATION TO THE POST-CRISIS ERA—BACKGROUND STUDIES (excerpts from content unit _081814b - 74.)*

### 84.      FSAPs’ analytical work on vulnerabilities is seen favorably, but satisfaction with the

### 84.      FSAPs’ analytical work on vulnerabilities is seen favorably, but satisfaction with the

### Analytical work on vulnerabilities and stress testing
- Over 75 percent of respondents agreed that the analysis was adequately framed in countries’ prevailing macro-financial conditions.
- Over 75 percent agreed that the technical work was calibrated to the availability and granularity of the financial sector data.
- Agreement on adequacy of models and methodologies used in stress tests:
  - 71 percent for country authorities.
  - 51 percent for Executive Directors.
- Lower approval rates for transmission, feedback, and contagion analysis:
  - Real-financial transmission channels: 59 percent for country authorities; 38 percent for Executive Directors.
  - Feedback effects: 58 percent for country authorities; 25 percent for Executive Directors.
  - Potential contagion among domestic financial institutions: 51 percent for country authorities; 38 percent for Executive Directors.

### Assessment of supervisory standards and RAMs (Risk Assessment Matrices)
- About 75 percent of country authorities viewed supervisory assessments as:
  - Sufficient to provide a robust evaluation of the regulatory and supervisory framework.
  - Adequately reflective of the main risks threatening the financial system.
  - Well aligned with other parts of the FSAP.
- 65 percent of country authorities considered the assessments sufficiently tailored to country-specific circumstances.
- Views on RAMs:
  - RAMs focus on the most relevant risks: 67 percent for country authorities; 75 percent for Executive Directors.
  - RAMs help frame discussion of risks with national authorities and anchor risks in FSSAs: 59 percent for country authorities; 63 percent for Executive Directors.
  - RAMs adequately explain the likelihood of shocks: 53 percent for country authorities; 25 percent for Executive Directors.
  - RAMs properly account for mitigating policies in assessment of impact: 45 percent for country authorities; 75 percent for Executive Directors.

### Modular approach and recommendations
- Stability modules: survey responses ambiguous due to small number of modules at survey launch (5) but 38 responses indicated possible misformulation or misinterpretation.
  - Executive Directors: 63 percent agreed modules were less demanding on country internal resources.
  - Executive Directors: 38 percent agreed stability modules allowed better selection of country-specific themes.
- FSAP recommendations:
  - Over 80 percent of country authorities and 75 to 100 percent of Executive Directors agreed main recommendations in the first table of Aide Mémoires (for country authorities only) and FSSAs were clear and candid.
  - At or above 50 percent in both groups agreed recommendations were actionable with reasonable effort and adequately prioritized and sequenced.

### Implementation rates and reasons for non/partial implementation
- Implementation status reported:
  - About a third of all respondents indicated having implemented FSAP recommendations in full.
  - Slightly over 60 percent reported having implemented FSAP recommendations partially.
- Reasons for non- or partial implementation:
  - Nearly half of country authorities cited disagreement with the recommendations.
  - 60 percent of Executive Directors cited the recent completion of the FSAP (country authorities’ second choice).
  - Executive Directors also cited limited capacity to implement recommendations: 40 percent (versus 14 percent of country authorities).
  - Executive Directors cited lack of financial resources as important; national authorities cited limited political support and recommendations not representing government priorities.

### Perceived contributions and impact
- Top FSAP contributions (both groups):
  - Adequate assessments of financial system stability (and identification of main risks).
  - Regulatory and supervisory framework assessments.
  - Macrofinancial linkages.
  - Crisis management capabilities.
- Divergence:
  - Integrated financial sector analysis ranked second by country authorities but eighth by Executive Directors.
- Lower-ranked contributions: outward spillovers, macroprudential frameworks and tools, and developmental assessments.
- Overall usefulness:
  - 79 percent for country authorities.
  - 100 percent for Executive Directors.
- High contribution areas: raising awareness on specific risks and need for policy action; improved coordination among supervisory agencies; improving dialogue with the Fund.
- Low contribution areas: reaching understandings with national legislatures; generating public debate and consensus on financial sector reforms.

### Engagement with stakeholders and follow-up
- FSAPs’ role in promoting dialogue:
  - Promoting discussions among government agencies: 66 percent for country authorities; 88 percent for Executive Directors.
  - Discussions between government and financial sector: 43 percent for country authorities; 75 percent for Executive Directors.
  - Between financial sector authorities and legislators: 42 percent for country authorities; 38 percent for Executive Directors.
  - Discussions with the public at large: 9 percent for country authorities; 13 percent for Executive Directors.
- Other impacts:
  - About half of country authorities and similar fraction of Executive Directors agreed FSAPs led to deeper discussions of financial sector issues with Article IV missions and to a stronger strategic view on financial sector development.
  - Only about a fifth of country authorities and a quarter of Executive Directors noted enhanced coordination among financial sector supervisors following FSAPs.
  - Small percentage believed FSAPs led to increased foreign investor interest.
- Integration with bilateral surveillance:
  - Executive Directors: 75 percent agreed FSAP recommendations were prioritized based on macroeconomic relevance and that FSAP and Article IV macroeconomic baselines were aligned.
  - Country authorities: 61 percent agreed recommendations were prioritized by macroeconomic relevance; 49 percent agreed baselines were aligned.
  - Follow-up requires specialization in legal, regulatory and supervisory issues: 61 percent for country authorities; 75 percent for Executive Directors.
  - General agreement there has been at least some follow-up of FSAP recommendations by Article IV teams.

### Publication of FSSAs
- Among countries that published FSSAs, national authorities reported:
  - In about 60 percent of these countries, publication strengthened credibility of supervisory authorities and was interpreted as positive by domestic financial entities (Executive Directors reported somewhat lower figures).
  - For a third of countries that published the FSSA, publication fostered a broader national policy dialogue on financial sector issues (both country authorities and Executive Directors).
  - Lower scores from country authorities on publication being welcomed by financial markets.

### Priorities for improving FSSAs and FSAP stability analyses
- FSSAs should:
  - Focus more systematically on systemic risk (top-ranked by both groups).
  - Be more concise/streamlined by dropping more technical parts.
  - Better tailor analysis of regulatory and supervisory issues to countries’ financial landscape.
  - Strengthen discussion of macroprudential tools.
  - Place relatively less weight on more cross-country comparisons and rely less on written material in favor of charts and tables.
- Improving stability analyses — differing views:
  - Country authorities’ top priority: strengthening analytical tools, particularly models for stress testing.
  - Executive Directors ranked strengthening stress-testing models 3 out of 5.
  - Both groups recommended deepening analysis of contagion across domestic financial institutions as a priority.
  - Country authorities ranked strengthening analysis of cross-border exposures third; Executive Directors ranked it as their main priority.
  - Both groups ranked lower: deepening analysis of spillovers and better aligning FSAP baseline scenarios with Article IV consultations.

*FSAP REVIEW—FURTHER ADAPTATION TO THE POST-CRISIS ERA—BACKGROUND STUDIES  
INTERNATIONAL MONETARY FUND*

### 98.      National authorities and Executive Directors ranked as the preferred option to assess

### _081814b - 98.      National authorities and Executive Directors ranked as the preferred option to assess

### Preferences for assessing supervisory standards
- Executive Directors saw as equally preferable (50 percent) to continue with the assessment of all principles and using a reduced set of principles to be applied to all countries.
- Country authorities ranked as preferred the option to assess a subset of international standards tailored to country-specific circumstances.
- Both groups considered as least desirable the option of assessing all principles of a smaller number of supervisory standards.

### Sample characterization (survey respondent makeup)
- Institutional affiliation (Country Authority Survey): central bank, superintendency of banks and/or other financial institutions, ministry of finance, financial stability agency, other.
- Financial Stability and FSAP Exposure (Country Authority Survey): 44%, 51%, 5% (categories shown in Figure 18).
- FSAP Experience (Country Authority Survey): Participated in an FSAP update (and/or stability module) in 2010–13: 47%; Participated in a first time FSAP in 2010–13: 64%; Participated in our only FSAP, which took place before 2010: 16%.

### Assessments of FSAP structure, scope, and analytical focus
- Structure and Coverage (Figure 19): Responses across Strongly Agree / Agree / Neutral / Disagree / Strongly Disagree indicate mixed but generally positive views by both Country Authorities and Executive Directors on whether "The FSAP was clearly structured around the three core themes" and related statements (detailed percent distributions shown in Figure 19).
- Scope (Figure 20): Both Country Authorities and Executive Directors evaluated whether "The FSAP covered the most relevant financial sector issues", "The FSAP provided adequate depth of analysis", and whether the analysis was tailored to country-specific circumstances (detailed percent distributions shown in Figure 20).
- Analytical focus (Figure 21): Evaluations addressed calibration to data, adequacy of stress-test methodologies, identification of transmission channels, feedback effects, and contagion channels (detailed percent distributions shown in Figure 21).

### Assessment integration, RAM, and stability module
- Integration of selected principles (Figure 22): Country Authorities reported on whether assessments were guided by main risks, sufficient for robust evaluation, integrated with other parts of the FSAP, and tailored to country-specific circumstances (detailed percent distributions shown in Figure 22).
- Risk Assessment Matrix (Figure 23): Surveyed perceptions covered whether the RAM focused on the most relevant risks, explained effects of materialization, anchored discussion in the Aide Memoire and FSSA, and considered mitigating policy responses (detailed percent distributions shown in Figure 23).
- Stability Module (Figure 24): Respondents evaluated whether the modular approach allowed more relevant selection of themes, aligned with country priorities, and was less demanding of internal resources (detailed percent distributions shown in Figure 24).

### Recommendations and implementation
- Recommendations (Figure 25): Country Authorities and Executive Directors provided percent distributions on whether recommendations were clear, candid, actionable with reasonable effort, prioritized well, and sequenced well (detailed percent distributions shown in Figure 25).
- Implementation (Figure 26): Implementation status of FSAP recommendations was reported (percent distributions shown in Figure 26).
- Motivations for not fully implementing recommendations (Figure 27): Reported motivations include disagreement with recommendations, limited political support, recommendations not representing government priorities, too many recommendations, limited technical capacity, lack of financial resources, difficulty prioritizing (percent distributions shown in Figure 27).

### Most valuable FSAP contributions (Table 7 and Figure 28)
- Country Authorities — Mean rankings (lowest mean = highest rank):
  - Assessment of financial system stability and identification of main risks: Mean 2.4, Rank 1
  - Integrated financial sector analysis: Mean 2.9, Rank 2
  - Assessment of regulatory frameworks and supervisory practices: Mean 3.5, Rank 3
  - Assessment of macro-financial linkages: Mean 4.2, Rank 4
- Executive Directors — Mean rankings:
  - Assessment of financial system stability and identification of main risks: Mean 1.8, Rank 1
  - Assessment of regulatory frameworks and supervisory practices: Mean 2.5, Rank 2
  - Assessment of crisis management capabilities: Mean 4.1, Rank 3
  - Assessment of macro-financial linkages: Mean 5.1, Rank 4
- Perceived contributions (Figure 28): Country Authorities and Executive Directors rated FSAP impacts on raising awareness of specific risks, raising awareness of need to take policy actions, overall usefulness, improving policy dialogue and IMF support, contributing to positive changes in the financial sector, expanding technical capabilities, prioritizing reform agenda, reaching understanding with legislature, generating public debate, and improving dialogue with the World Bank (detailed percent distributions shown in Figure 28).

### FSAPs’ contribution to policy debate and potential impacts
- FSAPs encouraged discussion among government agencies, promoted discussion between government and financial sector stakeholders, promoted discussion between financial sector authorities and legislators, and promoted discussion with the public at large (percent distributions shown in Figure 29).
- Potential impacts reported (Figure 30): depth of discussion on financial sector issues in Article IV missions increased after FSAP; strategic view on development of the financial sector strengthened after FSAP; coordination among financial sector supervisors in the region improved following FSAP; foreign investors’ interest increased following FSAP completion (detailed percent distributions shown in Figure 30).

### Integration with Article IV assessments (Figure 31)
- Country Authorities and Executive Directors reported on whether main FSAP recommendations were prioritized based on macroeconomic relevance, similarity of baseline macroeconomic scenarios used in the FSAP and Article IV, whether follow-up required high specialization, and whether there has been meaningful IMF follow-up (detailed percent distributions shown in Figure 31).

### Publication impact
- Publication of the FSSA was perceived to have strengthened the credibility of supervisory authorities, been seen as a positive development by domestic financial entities, fostered a broader national policy dialogue on financial sector issues, and been welcomed by financial markets; some respondents indicated the FSSA was not published (percent distributions shown in Figure 32).

### Improving the structure and presentation of FSSAs and stability analysis (Tables 8–10)
- Improving the structure of the FSSA (Table 8 — Mean, Rank):
  - Focus the analysis more systematically on systemic risk: Country Authorities Mean 1.7 Rank 1; Executive Directors Mean 1.8 Rank 1
  - Strengthen the discussion of macro-prudential tools: Country Authorities Mean 2.5 Rank 2; Executive Directors Mean 3.0 Rank 4
  - Better tailor the coverage of regulatory and supervisory issues to the country’s financial landscape: Country Authorities Mean 2.6 Rank 3; Executive Directors Mean 2.4 Rank 2
  - Provide more cross-country comparisons: Country Authorities Mean 3.3 Rank 4; Executive Directors Mean 2.9 Rank 3
- Improving presentation of results in FSSAs (Table 9 — Mean, Rank):
  - Structuring the FSSAs around main drivers of systemic risk, rather than around sectors: Country Authorities Mean 1.9 Rank 1; Executive Directors Mean 2.3 Rank 1
  - Shortening the FSSA’s length by streamlining its most technical parts: Country Authorities Mean 2.1 Rank 2; Executive Directors Mean 2.4 Rank 2
  - Providing more charts and tables with cross-country comparisons: Country Authorities Mean 2.7 Rank 3; Executive Directors Mean 2.5 Rank 3
  - Relying more on tables and charts and less on written material: Country Authorities Mean 3.2 Rank 4; Executive Directors Mean 2.9 Rank 4
- Improving FSAP’s stability analysis (Table 10 — Mean, Rank):
  - Strengthening models used for stress testing: Country Authorities Mean 2.5 Rank 1; Executive Directors Mean 2.9 Rank 3
  - Deepening the analysis of contagion across domestic financial institutions: Country Authorities Mean 2.6 Rank 2; Executive Directors Mean 2.6 Rank 2
  - Strengthening the analysis of cross-border exposures: Country Authorities Mean 2.8 Rank 3; Executive Directors Mean 2.3 Rank 1
  - Deepening the analysis of inward/outward spillovers: Country Authorities Mean 3.1 Rank 4; Executive Directors Mean 3.0 Rank 4
  - Better align the baseline and stress scenarios with those used in Article IV consultations: Country Authorities Mean 3.9 Rank 5; Executive Directors Mean 4.3 Rank 5

### Improving assessments of supervisory principles (Figure 33)
- Country Authorities (percent distribution shown in Figure 33):
  - Streamline the standards assessments to focus on those principles of most relevance to the country: 50%
  - Continue with assessments of all principles for each ROSC: 33%
  - Continue with assessments of all principles but focus on a reduced number of ROSCs: 11%
  - Streamline the standards assessments to focus on a subset of principles applied to all countries: 6%
- Executive Directors (summary reported in text):
  - Saw as equally preferable (50 percent) to continue with the assessment of all principles and using a reduced set of principles to be applied to all countries.

*Source: Survey results and tables/figures presented in the cited content unit.*

### 99.      This note describes the calculations conducted to assess trends in the cost of the FSAP

### _081814b - 99.      This note describes the calculations conducted to assess trends in the cost of the FSAP

### Methodology and data issues
- Challenges in estimating FSAP costs:
  - Costs can spread over 2 or more fiscal years.
  - FSAPs entail work by FSAP teams and departmental support staff; support-staff time was not tracked adequately prior to FY2011.
  - Result: a break in the time series of FSAP costs that complicates interpretation.
- Two cost series constructed:
  - Cash-basis cost series:
    - Merges ACES-based series (available since FY2011) with a pre-ACES series for FY2006–2010.
    - ACES series includes direct mission costs (including travel), work by FSAP teams prior to/after missions, and departmental support staff activities in headquarters.
    - Pre-ACES series does not include work by support staff.
    - Focuses on annual outlays on all FSAPs for which staff reported time in a given year.
  - Accrual-basis cost series:
    - For each FSAP, adds all costs incurred in various years and allocates the resulting figure to the fiscal year in which the FSAP was discussed at the Executive Board.
    - The beginning of an FSAP marked by the first mission and the end by the Board discussion of the FSSA.
    - Costs reported up to two years prior to or after this interval were also taken into account.
- Rationale:
  - Cash-basis series suitable for assessing the costs of running the FSAP in general.
  - Accrual-basis series better captures changes in underlying costs of FSAPs when costs vary significantly across countries.

### Aggregate cost trends and series comparison
- Both series indicate FSAPs have become more expensive in recent years, particularly after FY2011 and FY2012.
- Statement on magnitude and fluctuations:
  - "The cost of running the FSAP in the last 4 years has been around US$14 million on a cash basis, and has fluctuated between US$8–16 million on an accrual basis."
- Both series show costs rising further in FY2012; accrual series remains significantly above FY2011 levels even after a subsequent decline—this cannot be attributed solely to methodological changes.

### Cost composition and drivers
- Cash-basis figures: FSAP costs are explained mainly by personnel costs.
  - Regular staff costs account for nearly 90 percent of personnel costs; contractual and external experts explain the difference.
  - Personnel costs have grown at a somewhat faster pace than total costs.
  - Travel costs have declined.
- Fluctuations in the cash series are driven by changes in costs of FSAPs in emerging and advanced economies; costs in low-income countries vary significantly less.
  - This pattern is evident in sub-periods FY2006-2010 and FY2011-2014 and is independent of the methodology used.

### Accrual series: role of S25 and country composition
- Accrual basis series shows increased FSAP costs in recent years driven by:
  - An increase in the number and average cost of FSAPs in countries with systemically important financial sectors (S25).
- High correlation noted between overall FSAP costs and the number of FSAPs in S25 countries (referenced as Table 11).
- Average FSAP cost observations:
  - FSAP costs tend to rise with a country’s level of development.
  - FSAP costs rise significantly if the country is an S25 country—complexity of financial sector identified as main determinant of FSAP costs.
- The 2010 decision to make assessments mandatory in S25 countries changed the country composition of FSAPs and substantially increased FSAP costs.

### Key statistics (cash-basis breakdown by fiscal year)
- Aggregate totals and components (FY 2006 through FY 2014) — figures presented exactly as in source:
  - Total Cost: 9,067,877; 6,598,027; 10,298,747; 6,932,133; 8,737,849; 13,560,272; 14,406,613; 13,009,222; 12,699,900
  - Total Personel: 6,452,193; 4,942,343; 7,619,195; 5,109,016; 7,069,627; 10,321,394; 11,394,243; 10,127,994; 10,394,687
  - Staff: 5,521,857; 4,528,463; 6,940,347; 4,487,330; 6,251,007; 8,855,712; 9,189,998; 8,483,672; 8,784,771
  - Experts: 880,519; 362,396; 653,306; 594,835; 750,256; 961,628; 1,390,345; 1,130,686; 1,171,429
  - Contractuals: 49,817; 51,484; 25,541; 26,851; 68,364; 504,054; 813,900; 513,636; 438,487
  - Travel: 2,615,684; 1,655,683; 2,679,552; 1,823,117; 1,668,222; 3,238,877; 3,012,370; 2,881,228; 2,305,212
  - Personnel in % of Total Cost (rhs): 71; 75; 74; 74; 81; 76; 79; 78; 82

### Key statistics (accrual-basis FY2010–FY2014, from Table 11)
- Total Cost (FY 2010; FY 2011; FY 2012; FY 2013; FY 2014):
  - 5,677,532; 8,156,517; 16,989,765; 14,743,054; 14,300,957
- Advanced Economy (AE):
  - 917,911; 3,299,904; 6,308,941; 8,322,697; 6,511,305
- AE - S25:
  - ...; 2,615,542; 5,639,385; 6,975,182; 6,511,305
- Middle-Income Countries (MIC):
  - 1,980,863; 4,023,944; 8,813,252; 5,180,227; 6,845,599
- MIC-S25:
  - ......; 6,020,371; 749,220; 2,321,738
- Low-Income (LIC):
  - 2,778,758; 832,669; 1,867,572; 1,240,130; 944,053
- LIC-S25:
  - ...............
- Average Cost (FY 2010; FY 2011; FY 2012; FY 2013; FY 2014):
  - 283,877; 543,768; 738,685; 867,238; 1,021,497
- Advanced Economy (AE) average cost:
  - 229,478; 824,976; 1,051,490; 1,387,116; 1,302,261
- AE - S25 average cost:
  - ...; 2,615,542; 1,127,877; 1,743,795; 1,302,261
- Middle-Income Countries (MIC) average cost:
  - 247,608; 502,993; 734,438; 575,581; 855,700
- MIC-S25 average cost:
  - ......; 1,204,074; 749,220; 1,160,869
- Low-Income (LIC) average cost:
  - 347,345; 277,556; 373,514; 620,065; 944,053
- LIC-S25 average cost:
  - ...............
- Number of FSAPs (FY 2010; FY 2011; FY 2012; FY 2013; FY 2014):
  - 20 15; 23 17 14 (source table formatting shows concatenated counts: "2015  231714" — retained as in source)
- Number of FSAPs by category (values as presented in source table):
  - Advanced Economy (AE): 4 4 6 6 5
  - AE - S25: 0 1 5 4 5
  - Middle-Income Countries (MIC): 8 8 12 9 8
  - MIC-S25: 0 0 5 1 2
  - Low-Income (LIC): 8 3 5 2 1
  - LIC-S25: 0 0 0 0 0

### Accrual-basis sample composition by fiscal year (Table 12)
- Fiscal Year 2007 — Number of Countries: 17
  - Andorra, Australia, Bosnia & Herzegovina, CEMAC, Denmark, Georgia, Guatemala, Guyana, Ireland, Mexico, Panama, Poland, Portugal, Sierra Leone, Spain, Tunisia, Uruguay
- Fiscal Year 2008 — Number of Countries: 20
  - Algeria, Botswana, Canada, Costa Rica, Egypt, Gibraltar, Haiti, Kyrgyz Republic, Latvia, Lithuania, Mauritius, Moldova, Montenegro, Qatar, Samoa, Slovak Republic, Sri Lanka, Switzerland, Turkey, United Arab Emirates
- Fiscal Year 2009 — Number of Countries: 28
  - Austria, Barbados, BCEAO/WAEMU, Belarus, Bermuda, Bulgaria, Burkina Faso, Cameroon, Central African Republic, Croatia, Estonia, Iceland, Kazakhstan, Liechtenstein, Macedonia, Malawi, Mali, Monaco, Mongolia, Morocco, Niger, Pakistan, Russian Federation, South Africa, Syria, Tajikistan, Thailand, Ukraine
- Fiscal Year 2010 — Number of Countries: 20
  - Bangladesh, Burundi, Cape Verde, Cayman Islands, Côte d'Ivoire, Cyprus, Djibouti, Dominican Republic, Honduras, Isle of Man, Jersey, Jordan, Kenya, Macao, Mozambique, Philippines, Romania, San Marino, Serbia, Zambia
- Fiscal Year 2011 — Number of Countries: 15
  - British Virgin Islands, Cambodia, El Salvador, Fiji, Finland, Guernsey, Indonesia, Kuwait, Mongolia, Nicaragua, Oman, Peru, Tanzania, Trinidad & Tobago, United States
- Fiscal Year 2012 — Number of Countries: 23
  - Argentina, Belize, Bolivia, Chad, Chile, China, Germany, Ghana, India, Israel, Luxembourg, Mexico, Netherlands, Panama, Papua New Guinea, Paraguay, Russian Federation, Rwanda, Saudi Arabia, Sweden, Turkey, Uganda, United Kingdom
- Fiscal Year 2013 — Number of Countries: 17
  - Angola, Armenia, Australia, Bahamas, Brazil, Colombia, Czech Republic, France, Japan, Kosovo, Malaysia, Nigeria, Slovenia, Spain, Sri Lanka, Tunisia, Uruguay
- Fiscal Year 2014 — Number of Countries: 14
  - Albania, Algeria, Argentina, Austria, Barbados, Belgium, Canada, Hong Kong, Italy, Korea, Kyrgyz Republic, Poland, Singapore, Vietnam

*Source: Office of Budget and Planning and staff calculations (figures and tables reproduced exactly as in the content unit).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_081814b.pdf_