## Rethinking Macroprudential Capital Buffers ## Source details **Canonical URL:** [Rethinking Macroprudential Capital Buffers](https://www.imf.org/-/media/files/publications/dp/2025/english/rmcbea.pdf) ## Other formats - [Markdown version](/-/media/files/publications/dp/2025/english/rmcbea.pdf.md) - [Structured JSON version](/-/media/files/publications/dp/2025/english/rmcbea.pdf.json) --- ### Executive summary — purpose, motivation, and key lessons - Macroprudential releasable capital buffers (for example, the countercyclical capital buffer, CCyB) are intended to enhance banking-system resilience so that a release supports continued provision of credit through adverse conditions (BCBS 2010a). - Basel Committee supports voluntary adoption of a positive neutral setting of the CCyB (BCBS 2022a), which sets a positive baseline for the CCyB instead of the standard zero-neutral setting (BCBS 2010a). - Key empirical lessons from COVID-19: - No buffers were available for release at the beginning of the COVID-19 shock in many countries because buildup was typically conditioned on excessive credit growth measured by the credit-to-GDP gap. - Credit-to-GDP gaps can stay negative for prolonged periods, leaving buffers at zero despite activation frameworks. - Where releasable buffers had been built up and were released (examples: Hong Kong SAR, the United Kingdom, the euro area, and the European Union), releases increased banks’ capital headroom and supported provision of credit. - Where buffers were not released and banks lacked capital over and above requirements, banks were more prone to amplify shocks. ### Rationale for preemptive buildup and governance benefits - Preemptive buildup provides insurance against stresses that might arise even without prior excessive credit growth and cushions procyclical impacts on banks’ liquidity provision. - Theoretical and empirical work suggests the cost of building such buffers in normal times can be low when implementation proceeds gradually. - Governance benefits: - Availability of releasable buffers reduces the need for supervisors to exercise “forbearance” on rules governing banks’ accounting for nonperforming loans. - Eases the burden of proof on macroprudential policymakers previously needing evidence of excessive credit before implementing buffers. ### Operational design options and practical implementation - Releasable capital buffers can be implemented as: - A positive baseline or “neutral” setting of the CCyB. - Sectoral/releasable buffers (for example, buffers targeting residential real estate exposures). - Simpler binary design for capacity-constrained jurisdictions: a prespecified constant positive buffer in normal times and zero in periods of material stress. - Building buffers can begin as soon as banks are profitable so they can meet new requirements by retaining earnings or using existing voluntary buffers. - Buffer requirements reduce banks’ latitude to distribute earnings or capital to shareholders, but need not affect credit provision when lending is profitable for the banking sector as a whole. ### Calibration considerations and stress-test use - Baseline buffer size could key off vulnerabilities in the existing stock of exposures, such as: - the share of variable rate mortgages, - the share of borrowing in foreign currency, - measures of leverage in the household and corporate sectors. - Stress tests are a natural way of assessing buffer size and can account for gaps in macroprudential toolkits. ### Cross-country insights and recommended orientations - Novel analysis suggests: - Advanced economies stand to benefit from a positive baseline setting of releasable buffers. - Emerging market economies (EMEs) may want to aim for an even larger baseline capital buffer because crises in these countries are less closely tied to prior increases in credit and exposure to external shocks can be more important. - It is useful for both advanced and emerging economies to maintain a share of overall capital requirements as a releasable capital buffer that can be released in periods of stress and rebuilt gradually as stresses abate. ### Interactions with other requirements and offset options - Introducing a positive neutral CCyB or other releasable buffers increases overall capital requirements unless offset by reducing other requirements. - Offsetting can be considered to avoid double-counting where existing nonreleasable buffers (for example, those drawn from stress tests) are meant to protect against macroeconomic changes. - Forward-looking loan loss provisioning frameworks can complement buffers by contributing to resilience against increases in expected default rates. - Activation and pace of buffer tightening should be conditional on the absence of financial stress that leads to curtailed bank credit provision. ### High-level practical framework for buffer evolution - A positive neutral buffer framework implies four phases for the buffer rate: - Accumulation to neutral when banks are not capital-constrained (can start when banks are profitable). - Further increases beyond neutral guided by early warning signals as vulnerabilities build. - Release aimed at preventing capital requirements from becoming binding during periods of financial stress (buffer set to zero during the stress/recovery when banks are digesting losses). - Rebuilding gradually as stresses abate, allowing compliance via retained earnings or voluntary buffers. - Early activation helps ensure a buffer is available for release in the event of exogenous or external shocks. --- ### Section 2 — Availability of releasable buffers at COVID-19 onset (key facts) - Some 85 jurisdictions introduced a framework for the CCyB since 2016. - Many had never lifted the buffer rate up from its default setting of zero. - Only 17 countries had a positive buffer rate and could use it to provide some relief at the onset of the pandemic. - 66 countries had a CCyB framework with no releasable buffer. - The majority of countries that had a positive CCyB rate going into the pandemic released all of their buffer following the COVID-19 shock; only six countries left some of their accumulated buffer in place. - In practice, some countries without a positive CCyB resorted to releasing their CCB or SRB. ### Reasons for the lack of buffers (mechanisms and indicator limitations) - Basel guidance emphasized CCyB’s role tied to excessive credit growth, overlooking other drivers of cyclical systemic risk (interest rates, exchange rates, commodity prices). - Emphasis on the credit-to-GDP gap contributed to prolonged zero settings because applied filtering techniques can produce persistent negative gaps when the historical trend includes prior buoyant credit growth. - Under Basel III guidance the CCyB is the only part of the capital structure explicitly available for release; the capital conservation buffer (CCB) is technically not releasable but only usable. - For EU countries, SRB use has been mainly for noncyclical risks, limiting scope for relaxation in aggregate shocks until recent clarifications. ### Pandemic implications for buffer design and governance - Where buffer space was available, the CCyB was used as intended during COVID-19. - Governance issues did not prevent prompt release where buffers existed, despite tension between macroprudential support and microprudential depositor protection. - Positive neutral setting for the CCyB: - Recognizes difficulty in predicting stress periods and provides insurance. - Gives authorities more time to observe evidence and allows gradual increases. - Eases buffer management given typical phase-in periods for tightening decisions (typically 12 months). - May soften political economy constraints by implying a larger buffer outside periods of excessive credit. --- ### Empirical evidence on buffer releases and lending (selected findings) - Prepandemic and historical evidence: - Following a capital release in a 2008 episode, firms borrowing from banks holding a 1 percentage point higher capital buffer after the release received 11 percent more in credit. - Banks that enjoyed greater relief increased their loan loss provisioning during stress, suggesting a stabilizing role of buffer releases. - Spain: accumulating dynamic provisions ahead of the global financial crisis had only a temporary effect on credit tightening; in bad times higher provisioning cushioned losses and supported firm survival and employment (Jiménez and others 2017). - COVID-19 evidence: - Provision of credit overall held up during COVID-19, unlike during the global financial crisis (BCBS 2021). - Buffer releases were effective in supporting new mortgages (Mathur, Naylor, and Rajan 2023). - CCyB releases had sizable and significant effects on lending especially where loan guarantees were less generous (Dursun-de Neef, Schandlbauer, and Wittig 2023). - Complementarity observed between CCyB release and targeted loan guarantees (Wong and others 2022). - Studies find stronger effects for banks with thin headroom (Berrospide, Gupta, and Seay 2021; Couaillier and others 2022a,b; Dursun-de Neef and others 2023; Bedayo and Galán 2024; Avezum, Oliveira, and Serra 2024). - Galán (2024) using a growth-at-risk framework finds releases of capital buffers in periods of financial stress have immediate and important effects in supporting economic growth in the tails. ### Costs of building buffers (selected quantitative evidence) - BCBS studies: upper bound cost to GDP in range 0.02–0.35 percent for a 1 percentage point rise in the capital ratio (BCBS 2010b, 2019a). - Elliot, Salloy, and Santos (2012): long-term impact of Basel III capital requirements on bank lending rates in Europe, Japan and the United States are less than 30 basis points per 1 percentage point increase in capital. - Basten (2020): a CCyB increase raises mortgage loan rates by eight basis points. - Xiao Chen and Friedrich (2023): a CCyB increase results in a decline of 12 to 17 basis points in the growth rate of cross-border lending between Canadian banks and borrowers abroad. - Gambacorta and Shin (2016): a 1 percentage point increase in the equity-to-total-assets ratio reduces overall cost of debt funding and leads to a 0.6 percentage point increase in annual loan growth. - BCBS (2022c) evaluation found no meaningful evidence of negative side effects of Basel reforms on banks’ lending and capital costs in the long term. - Short-term adjustment costs documented when increases occur in periods of stress and actual capital is thin (Gropp and others 2019). - Some studies find increases in capital requirements can be more binding for smaller banks and reduce lending to smaller firms (Jiménez and others 2017; Behn, Forletta, and Rhegezza 2024). ### Policy trade-offs and implementation guidance - Timing and pace of buffer increases matter: allow transition time, account for bank profitability and voluntary buffers to minimize short-term credit costs. - Buffer releases can be effective depending on crisis context and complementary policies (dividend restrictions, fiscal support, loan guarantees). - Where banks are profitable or hold voluntary buffers, buildup costs are likely small and can be met via retained earnings given phased implementation. --- ### Section 4 — Operational framework: phases, triggers, build/release guidance - Framework overview: buffer moves through four phases—Recovery, Normal-risk (CCyB—neutral at 2 percent), Risk build-up (beyond neutral), Financial stress (partial/full release). - Main difference with PNCCyB: positive buffers are available in normal times and can be released for external or exogenous shocks. A. When to release buffers? - Objective: support continued provision of credit when stress makes regulatory capital requirements binding (BCBS 2010a). - Indicators guiding a release: - Near-contemporaneous indicators of stress and judgment. - Incipient increases of nonperforming loans and early signs of deteriorating asset quality. - Market-based indicators (sharp drops in bank equity prices, increases in credit default swap premiums). - Indicators of liquidity stress on borrowers (e.g., drawing down credit lines). - Surveys of bank lending standards indicating constrained provision. - If survey data signal the CCyB is becoming binding on a substantial share of the financial system and this affects aggregate credit flows. - What not to use a release for: - Not to manage the business cycle or respond to reduced demand for credit (IMF 2014a). - Communication: - Clarify release is precautionary and avoid implying new information about broader economic risk; timebound commitments can increase banks’ willingness to use released space. B. When and at what pace to build? - Preconditions: - Supply of credit not constrained by capital requirements. - Absence of signals that would support a release. - Banks’ profitability restored sufficiently. - Pace considerations: - Balance protection benefits and adjustment costs. - When monetary policy tightens, building buffers can still be useful; tighter monetary policy often boosts bank profitability initially. - Make pace contingent on absence of detrimental side effects (for example, enforce baseline pace after one or two years while retaining flexibility). C. What level for the neutral rate? - Approaches: - Aim buffer large enough relative to expected losses in stress so a release can reduce amplification. - Stress-test-based calibration: target resilience to milder adverse scenarios, account for nonreleasable buffers (CCB and SRBs). - Historical peak-loss approach: use peak accumulated losses in past stress episodes as yardstick. - Country-specific factors arguing for larger neutral buffers: - Substantial stock vulnerabilities (high household leverage, large share of variable-rate or foreign-currency debt). - Constraints on monetary policy (pegs, currency unions). - Small open and emerging market economies exposed to external shocks may want permanently larger buffers released only in large shocks. D. When to increase CCyB beyond neutral? - Triggers: any information implying probability of future financial stress has increased. - Use a range of complementary indicators (credit gaps, asset price deviations, market risk measures, debt service burden, leverage, external imbalances). - Growth-at-risk and countercyclical stress testing can inform sizing. - Apply judgment rather than mechanical rules; do not reduce CCyB below neutral unless release conditions are fulfilled. E. Assessing and addressing leakage effects - Leakage risks: - Domestic leakage to bank-affiliated nonbanks. - Cross-border leakage via foreign branches absent reciprocity. - Leakage when capital requirements increase during stress. - Addressing leakage: - Expand scope of macroprudential tools across banks and affiliated nonbanks. - Apply borrower-based and liquidity tools to nonbank providers. - Use reciprocity arrangements (as in BCBS CCyB agreement) to address cross-border leakage. - Consider applying a higher domestic rate to outward exposures while reciprocating host authority actions. --- ### Box 1 — Using past crises to guide calibration (indicative estimates) Data and method highlights: - Sample: 27 advanced economies (AE), 29 emerging market economies (EME), and 11 low-income countries (LIC). - Peak losses as share of RWA computed as (Peak NPL × Average Share of Total Gross Loans to RWA). - Average Share of Total Gross Loans to RWA calculated as (Total Gross Loans × 15 percent)/RWA for AEs and (Total Gross Loans × 25 percent)/RWA for EMEs. - Recovery assumptions: assume 85 percent of AE NPLs recovered; assume 75 percent of EME and LIC NPLs recovered. Indicative average buffer ranges to cover losses from peak NPLs: - Advanced economies: a buffer between 3 to 7.5 percent would be needed to cover losses from low or high credit growth crisis respectively. - EMEs: a 6 to 7 percent buffer would have covered losses in low or high credit crisis periods. - LICs: a buffer between 5 and 9 percent would be needed. Policy implications: - Advanced economies: buffers should respond strongly to credit conditions; allow increases in releasable buffers in response to strong credit flows. - EMEs: emphasize larger neutral (permanent) buffers rather than relying on cyclical responsiveness to credit growth. - LICs: well-capitalized banks and potentially permanently higher capital buffers may be warranted given incidence of currency and sovereign crises. - Releasable buffers preferred to higher minimums given sizable post-crisis credit slowdowns, especially in EMEs. - Need for country-specific analysis and more granular stress tests; absolute numbers are indicative. --- ### Section 5 — Relationship with other capital requirements and tools A. Provisioning and dividend restrictions - Expected credit loss framework (IASB 2014) produces forward-looking provisioning that smooths profits and reduces likelihood of hitting capital constraints; may not cover all losses in downturns. - Releasable CCyB can absorb unexpected losses from sudden exogenous shocks and support financing provision. - Dynamic provisioning (Spain example) built general provision buffers for drawdown in bad times; where ceilings are too low, a releasable capital buffer can complement dynamic provisions. - Aggregate dividend/distribution restrictions can preserve capital but legal authority can be constrained; for preparing for future stress, increases in capital buffer requirements may be preferred to distribution restrictions. B. CCyB and sectoral or alternative releasable buffers - Basel Committee (BCBS 2019c) provides principles for sectoral CCyB implementation. - Sectoral CCyB and sectoral buffers are viable alternatives or complements (examples: Swiss sectoral CCyB against housing; SRB used on real estate exposures in Belgium and Germany). - Combining sectoral buffers with broad-based CCyB can balance targeted risk mitigation and systemwide protection. - Simpler binary CCyB design recommended for data- or capacity-constrained jurisdictions. C. CCyB and Capital Conservation Buffer (CCB) - CCyB and CCB are complements; CCyB protects aggregate activity, CCB allows dipping for idiosyncratic losses and triggers distribution constraints. - During COVID-19 some countries released CCB (maximum of 2.5 percent of RWA), contrary to Basel guidance; releasing CCB removes supervisory instrument to control individual banks. - Outside aggregate stress, a positive neutral CCyB adds to combined buffer requirement (CCyB + CCBs); combined buffer can function as an expanded CCB when levels exceed 2.5 percent. D. CCyB and bank-specific buffers from stress tests - Stress tests and Pillar 2 guidance should account for a positive neutral CCyB to avoid double-counting macro risks. - Supervisory guidance can remain bank-specific; stress scenarios in normal times may assume release of CCyB in the stress scenario. - Pillar 2 buffers remain legitimate even during broad-based buffer release for idiosyncratic reasons. E. CCyB, leverage, and resolution requirements - Release of capital buffers can cause unweighted capital requirements (leverage ratio) or resolution-related requirements to become binding, undermining release effectiveness. - Presence of unweighted capital requirements should be considered when calibrating releasable buffers; a larger releasable buffer may be chosen if unweighted requirements would hamper effectiveness. - International work needed to ensure buffer release is not hampered by unweighted requirements. --- ### Box 2 — Positive neutral CCyB, reciprocity, and early adopters (selected guidance and statistics) - Basel reciprocity: CCyB determined nationally; reciprocity mandated up to 2.5 percent of RWA; above 2.5 percent reciprocation voluntary. - Positive neutral CCyB increases reciprocated capital levels allocated by foreign firms relative to zero-neutral framework (assuming reciprocation below 2.5 percent). - For non-Basel members with significant foreign branches, establish reciprocity arrangements or require subsidiaries for enforceability. - Implementation guidance: - Do not wait for elevated credit gap readings; start implementation when release conditions are not met and supply of credit is not constrained. - Phase in new requirements to build buffers gradually from retained earnings. - Focus neutral buffer analysis on vulnerabilities in existing stock of exposures and use stress tests to size buffers. - Larger buffers likely useful for small open and emerging market economies exposed to external shocks or monetary-policy constraints. Key statistics — early adopters of a positive neutral CCyB rate (information as of January 2025): - Armenia 1.5 percent - Australia 1 percent - Chile 1 percent - Cyprus 0.5 percent - Czech Republic 1 percent - Estonia 1 percent - Georgia 1 percent - Greece 0.5 percent - Hong Kong SAR 1 percent - Hungary 1 percent - Iceland 2–2.5 percent - Ireland 1.5 percent - Latvia 1 percent - Lithuania 1 percent - New Zealand 1.5 percent - The Netherlands 2 percent - Poland 2 percent - Portugal 0.75 percent - Slovenia 1 percent - South Africa 1 percent - Spain 1 percent - Sweden 2 percent - United Arab Emirates 0.5 percent - United Kingdom 2 percent --- *Source: IMF Departmental Paper — “Rethinking Macroprudential Capital Buffers.”* ### Executive Summary.................................................................................................v ### Executive Summary ### Rethinking releasable macroprudential capital buffers: purpose and motivation - Macroprudential buffers were conceived after the global financial crisis and experience with their use has grown, prompting a rethink of their design and deployment across the financial cycle. - A releasable capital buffer (for example, the countercyclical capital buffer, CCyB) is intended to enhance banking-system resilience so that a release supports continued provision of credit through adverse conditions (BCBS 2010a). - The Basel Committee on Banking Supervision supports voluntary adoption of a positive neutral setting of the CCyB (BCBS 2022a), which sets a positive baseline for the CCyB instead of the standard zero-neutral setting (BCBS 2010a). ### Key empirical lessons from COVID-19 - In many countries, no buffers were available for release at the beginning of the COVID-19 shock because buildup of releasable buffers had typically been conditioned on excessive credit growth as measured by the credit-to-GDP gap. - Credit-to-GDP gaps can stay negative for prolonged periods, especially when past credit booms push up the historical trend, so frameworks that rely on such gaps often left buffers at zero despite having activation frameworks in place. - Where releasable buffers had been built up and were released (examples cited include Hong Kong SAR, the United Kingdom, the euro area, and the European Union), the release supported provision of credit to the real economy by increasing banks’ capital headroom over remaining requirements. - Where buffers were not released and banks lacked capital over and above requirements, banks were more prone to amplify shocks. ### Rationale for preemptive buildup of releasable buffers - Preemptive buildup provides insurance against stresses that might arise even without prior excessive credit growth, cushioning the procyclical impact of shocks on banks’ liquidity provision. - The growing body of theoretical and empirical research suggests the cost of building such buffers in normal times can be low when implementation proceeds gradually. - Governance benefits: - Availability of releasable buffers reduces the need for supervisors to exercise “forbearance” on rules governing banks’ accounting for nonperforming loans. - It eases the burden of proof on macroprudential policymakers who previously needed evidence of excessive credit before implementing buffers. ### Operational implications and design options - Releasable capital buffers can be implemented as: - A positive baseline or “neutral” setting of the CCyB. - Sectoral/releasable buffers (for example, buffers targeting residential real estate exposures). - For jurisdictions with limited data or capacity, a simple binary design can be considered: a prespecified constant positive buffer in normal times and zero in periods of material stress. - Building buffers need not wait for credit gaps to signal a future crisis; accumulation can begin as soon as banks are profitable so banks can meet new requirements by retaining earnings or using existing voluntary buffers. - While buffer requirements reduce banks’ latitude to distribute earnings or capital to shareholders, they need not affect credit provision when lending is profitable for the banking sector as a whole. ### Calibration considerations - Baseline buffer size could key off vulnerabilities in the existing stock of exposures, such as: - the share of variable rate mortgages, - the share of borrowing in foreign currency, - measures of leverage in the household and corporate sectors. - Stress tests are a natural way of assessing the size of buffers and can account for potential gaps in macroprudential toolkits. ### Cross-country insights and recommendations - Novel analysis in the paper suggests: - Advanced economies stand to benefit from a positive baseline setting of releasable buffers. - Emerging market economies (EMEs) may want to aim for an even larger baseline capital buffer, because crises in these countries are less closely tied to prior increases in credit and exposure to external shocks can be more important. - It is useful for both advanced and emerging economies to maintain a share of overall capital requirements as a releasable capital buffer that can be released in periods of stress and rebuilt gradually as stresses abate. ### Interactions with other requirements and offset options - Introducing a positive neutral CCyB or other releasable buffers increases overall capital requirements unless offset by reducing other requirements. - Some offsetting can be considered to avoid double-counting where existing nonreleasable buffers (for example, those drawn from stress tests) are meant to protect against macroeconomic changes. - Forward-looking loan loss provisioning frameworks can complement buffers by contributing to resilience against increases in expected default rates. - The activation and pace of buffer tightening should be conditional on the absence of financial stress that leads to curtailed bank credit provision. ### Practical framework for buffer evolution (high level) - A positive neutral buffer framework implies four phases for the buffer rate: - Accumulation to neutral when banks are not capital-constrained (can start when banks are profitable). - Further increases beyond neutral guided by early warning signals as vulnerabilities build. - Release aimed at preventing capital requirements from becoming binding during periods of financial stress (buffer set to zero during the stress/recovery when banks are digesting losses). - Rebuilding gradually as stresses abate, allowing compliance via retained earnings or voluntary buffers. - Early activation helps ensure a buffer is available for release in the event of exogenous or external shocks. *Source: Executive Summary of "Rethinking Macroprudential Capital Buffers."* ### 2. Lack of Buffers ahead of the COVID-19 Crisis ### 2. Lack of Buffers ahead of the COVID-19 Crisis ### Availability of releasable capital buffers at the onset of COVID-19 - Some 85 jurisdictions introduced a framework for the CCyB since 2016. - Many of these countries had never lifted the buffer rate up from its default setting of zero. - Only 17 countries had a positive buffer rate and could use it to provide some relief at the onset of the pandemic. - 66 countries had a CCyB framework with no releasable buffer. - The majority of countries that had a positive CCyB rate going into the pandemic released all of their buffer following the COVID-19 shock; only six countries decided to leave some of their accumulated buffer in place. - In practice, some countries that did not have a positive CCyB resorted to releasing their CCB or SRB, which were in place in many European countries and also outside the European Union. ### Reasons for the lack of buffers - Basel Committee guidance (2010a) emphasized the CCyB’s main objective as supporting banks in providing credit during stress, while also framing the CCyB as a tool to mitigate risks from excessive credit growth. The guidance defined the CCyB objective as protecting the banking system against potential future losses when excess credit growth is associated with an increase in systemwide risk. - This framing overlooked that cyclical systemic risk may arise from drivers other than private sector credit growth, including adverse changes in interest rates, exchange rates, or commodity prices. - The emphasis on the credit-to-GDP gap as the main trigger likely contributed to prolonged zero settings: - The Basel buffer guide (BCBS 2010a) identified the credit-to-GDP gap as a key indicator for CCyB activation and tightening. - The parameters for the applied filtering technique were chosen to mimic long financial cycles, producing a tendency for negative gaps to arise and persist when the statistical trend includes prior periods of buoyant credit growth. - Policymakers often felt unable to act when the credit gap was deeply negative, contributing to long periods with CCyB rate at zero. - Under Basel III guidance, the CCyB is the only part of the capital structure that is available for release. The capital conservation buffer (CCB) is technically not releasable but only usable, meaning the CCB requirement remains in place for all banks even as individual banks may use the buffer. - For EU countries, the SRB has been used mainly for noncyclical risks, limiting scope for relaxation in aggregate shocks; use for cyclical and sectoral risks has only recently been clarified in European legislation. ### Pandemic experience and implications for buffer design - Countries that had built sufficient policy space used it in the COVID-19 pandemic; where buffer space was available, the CCyB was used as intended. - The pandemic evidence suggests governance issues did not prevent prompt release of buffers where available, despite potential tension between macroprudential support for credit extension and microprudential concerns to protect depositors. - A releasable capital buffer can reduce incentives to resort to regulatory forbearance in crises (for example, relaxation of microprudential minima or accounting standards for nonperforming loans), which can mask losses and reduce transparency. - A positive neutral setting for the CCyB: - Recognizes difficulty in predicting periods of stress and provides insurance against unforeseen stress periods. - Better positions countries to reach desired resilience when vulnerabilities build, by giving authorities more time to observe evidence and allowing more gradual timing of buffer increases. - Eases buffer management given typical phase-in periods for tightening decisions (typically 12 months). - May soften political economy constraints by implying a larger buffer outside periods of excessive credit and reducing the burden of proof on authorities to demonstrate excessive credit expansion. - Greater use of a positive neutral CCyB and other releasable elements of the capital structure could reduce the need for regulatory forbearance and help preserve the integrity of the prudential framework. ### Empirical and theoretical evidence on benefits of releasable buffers - The credit gap was identified in earlier research as the best single early warning indicator of a banking crisis two to three years out, but it does not signal all future crises. - Lang and Menno (2023) provide a theoretical argument: when the impact of capital requirements on lending is state-dependent, gradual increases in capital in normal times have small lending effects via a “pricing channel,” while releases when banks are capital constrained trigger a larger “quantity channel.” Bank profitability is a key determinant of desirability of buffer accumulation in this framework. - Empirical findings on bank behavior and buffer effects: - Banks maintain positive management buffers or “headroom” because breaching requirements entails costs (including enhanced supervisory scrutiny and market stigma). - Berrospide, Gupta, and Seay (2021) and Couaillier and others (2022a) find banks closer to breaching buffer requirements behaved more defensively and reduced lending more during the pandemic than banks with more headroom. - Releases of capital buffers created additional headroom, supporting continued provision of credit when needed; effects were stronger for banks with thin headroom. - Studies documenting supportive effects of CCyB releases include: - Wong and others (2022) for Hong Kong SAR. - Couaillier and others (2022b) for the euro area. - Mathur, Naylor, and Rajan (2023) for the United Kingdom: banks with lower headroom maintained tighter mortgage lending standards, and banks receiving greater capital relief from CCyB reduction were more likely to maintain lending to riskier loan categories. - Dursun-de Neef, Schandlbauer, and Wittig (2023); Bedayo and Galán (2024); and Avezum, Oliveira, and Serra (2024) for broader EU samples find sizable effects of CCyB releases, especially for banks with low headroom. - Galán (2024) using a growth-at-risk framework finds that releases of capital buffers in periods of financial stress have immediate and important effects in supporting economic growth in the tails of the distribution. - Sivec and Volk (2022) use an event study around a Slovenian prudential measure that reduced bank capital requirements (of nearly 1 percentage point of risk-weighted assets on average) during a financial crisis. *Source: IMF Departmental Paper — “Rethinking Macroprudential Capital Buffers,” section 2, “Lack of Buffers ahead of the COVID-19 Crisis”; canonical PDF: https://www.imf.org/-/media/files/publications/dp/2025/english/rmcbea.pdf* ### 2008. They find that following the capital release banks that benefited more from the release were able to ### rmcbea - 2008. They find that following the capital release banks that benefited more from the release were able to ### Evidence on buffer releases and lending during stress - Following a capital release, firms borrowing from banks holding a 1 percentage point higher capital buffer after the release received 11 percent more in credit (study of 2008 episode). - Banks that enjoyed greater relief increased their loan loss provisioning during the period of financial stress, suggesting a stabilizing role of buffer releases in helping banks absorb losses. - Prepandemic evidence indicates that where banks face loan losses, the cutback in lending is less where available capital buffers are larger (for example, Nier and Zicchino 2008). - Spain (dynamic provisions): accumulating provisions ahead of the global financial crisis had only a temporary effect in tightening credit supply and did not affect firm survival because firms could substitute to less affected banks; in bad times higher provisioning cushioned losses and supported firm survival and employment (Jiménez and others 2017). However, dynamic provisions were eventually depleted in the large shock. - Empirical evidence from the COVID-19 crisis: - Provision of credit overall held up during COVID-19, unlike during the global financial crisis (BCBS 2021). - Buffer releases were effective in supporting provision of new mortgages (Mathur, Naylor, and Rajan 2023). - CCyB releases had sizable and significant effects on lending especially where loan guarantees were less generous (Dursun-de Neef, Schandlbauer, and Wittig 2023). - Complementarity observed between CCyB release and targeted loan guarantees (Wong and others 2022). - Studies indicate macroprudential and fiscal responses played a positive role in reducing crisis impact (Bergant and Forbes 2021). - Reasons why banks may still lend in stress even if less profitable: - Contractual obligations to satisfy preexisting credit lines to avoid legal costs and preserve reputational capital. - Providing liquidity to existing clients can prevent defaults on existing loans to the bank. - Preserving relationships with existing (especially large) clients to secure future profitable business. - Avoiding loss of market share to competitors by passing up lending opportunities. - Caveats on generalizing COVID-19 findings: - COVID-19 generated strong demand for credit; other crises may similarly raise liquidity needs but patterns could differ. - Generous official support (monetary, fiscal) and dividend restrictions during COVID-19 may have reduced banking losses and maintained capital buffers, affecting the observed effectiveness of buffer releases. - If losses had been higher, banks might have been more reluctant to use released buffers; alternatively, more banks might have been constrained by buffers becoming binding, potentially amplifying the effect of releases. - Pre-pandemic buffer-release research (e.g., Jiménez and others 2017; Sivec and Volk 2022) suggests impacts could plausibly have been larger in other contexts. ### Costs of building buffers and long-term effects - Basel Committee studies establish an upper bound for the cost to GDP in the range of 0.02–0.35 percent for a 1 percentage point rise in the capital ratio (BCBS 2010b, 2019a). - Elliot, Salloy, and Santos (2012): long-term impact of Basel III capital requirements on bank lending rates in Europe, Japan and the United States are less than 30 basis points per 1 percentage point increase in capital. - Basten (2020): a CCyB increase raises mortgage loan rates by eight basis points. - Xiao Chen and Friedrich (2023): a CCyB increase results in a decline of 12 to 17 basis points in the growth rate of cross-border lending between Canadian banks and borrowers abroad. - Gambacorta and Shin (2016): a 1 percentage point increase in the equity-to-total-assets ratio reduces the overall cost of debt funding and leads to a 0.6 percentage point increase in annual loan growth. - BCBS (2022c) evaluation found no meaningful evidence of negative side effects of the Basel reforms—including higher capital requirements—on banks’ lending and capital costs in the long term. - Short-term adjustment costs are documented, especially when increases occur in periods of stress and actual capital is thin (Gropp and others 2019). - Some studies find increases in capital requirements can be more binding for smaller banks with less headroom and reduce their lending to smaller firms (Jiménez and others 2017; Behn, Forletta, and Rhegezza 2024). - The effect of capital requirement increases on lending is likely small where banks are profitable or have built voluntary buffers (Lang and Menno 2023; Munoz and Smets 2024; Bedayo and Galán 2024; Behn, Forletta, and Rhegezza 2024). - If banks are profitable, they can meet new buffer requirements by retaining earnings given sufficient transition time. - If banks have high voluntary buffers, they may reduce voluntary buffers when a new macroprudential buffer is imposed, trusting authorities to release the buffer in aggregate stress. - If banks are unprofitable and hold low voluntary buffers, a rapid buildup could have more pronounced negative effects, suggesting phased implementation and additional time for adjustment. - Extreme increases in capital requirements could constrain deposit funding and impair banks’ profitability and payment activities, though most systems are likely to operate below that point. ### Theory, trade-offs, and policy implications - The literature aligns with Modigliani-Miller theorem: weighted average cost of capital is a function of the net present value of assets and invariant to shifts in equity-debt share in liabilities (Modigliani and Miller 1958). - In models with defaultable debt, higher equity share reduces probability of default and cost of debt, consistent with Gambacorta and Shin (2016). - Debt overhang (Myers 1977) explains shareholders’ reluctance to commit additional equity even when socially beneficial (Admati and others 2012): shareholders may oppose equity increases that benefit other stakeholders. - Stock market reactions to buffer increases may be negative while bond prices increase (Gambacorta and Shin 2016). - Policy trade-offs: - Capital buffers can improve banks’ ability to continue lending through adverse shocks, benefiting constrained borrowers and society. - Policymakers should consider timing and pace of buffer increases, allowing transition time and accounting for bank profitability and voluntary buffer levels to minimize short-term credit costs. - Buffer releases can work as intended and need not be “pushing on a string,” but effectiveness depends on crisis context, complementary policies (dividend restrictions, fiscal support, loan guarantees), and banks’ balance-sheet conditions. *IMF DEPARTMENTAL PAPERS • Rethinking Macroprudential Capital Buffers* ### 4. How Would the Overall Framework Operate? ### 4. How Would the Overall Framework Operate? ### Framework overview and evolution of the CCyB - A positive neutral CCyB framework guides banks to accumulate additional capital when the cost of accumulating this capital is relatively low (for example, when profits are high), and makes that capital available when its value is greater, typically during periods of financial stress.17 - Under this framework the buffer rate moves through four distinct phases, involving three decision points: - Recovery phase - Normal-risk phase (CCyB—neutral at 2 percent) - Risk build-up phase (raise CCyB beyond neutral when cyclical risks build) - Financial stress phase (partial or full release when stresses arise) - Early activation of the CCyB is the main difference between the PNCCyB and the existing CCyB framework: positive buffers are available in normal times and can be released in response to external or exogenous shocks. ### A. When to release buffers? - Objective: - Support continued provision of credit when the financial system experiences stress and regulatory capital requirements are binding (BCBS 2010a). - Reduce the risk that credit provision becomes constrained and turns procyclical, amplifying real-economy downturns and credit losses.19 - Indicators that should guide a release: - Near-contemporaneous indicators of stress and judgment. - Incipient increases of nonperforming loans and early signs of deteriorating asset quality. - Market-based indicators (sharp drops in bank equity prices, increases in credit default swap premiums). - Indicators of liquidity stress on borrowers (e.g., drawing down credit lines). - Surveys of bank lending standards indicating constrained provision of lending to households or firms.20 - If survey data signal the CCyB is becoming binding on a substantial share of the financial system and this affects aggregate credit flows, this can signal a release.21 - What not to use a release for: - Not to manage the business cycle or respond to reduced demand for credit (IMF 2014a). If banks are not constrained by capital requirements, a release cannot generate benefits in funding costs or credit expansion.22 - Communication and credibility: - Clarify that the release is a precaution to help banks function normally and avoid implying new information about broader economic risk. - Authorities should communicate that buffer reactivation will not be premature; timebound commitments can increase banks’ willingness to use released space (example: Bank of England announcement). Clear guidance should emphasize reactivation only when the banking system is profitable enough to rebuild buffers.23 ### B. When and at what pace would buffers be built? - Preconditions to start building toward the neutral rate: - The supply of credit should not be constrained by capital requirements. - Absence of signals that would support a capital buffer release. - Banks’ profitability restored sufficiently to rebuild and maintain voluntary capital buffers.24 - Bank lending surveys and market indicators can help assess conditions. - Pace considerations: - Balance benefits (protection against vulnerabilities) and adjustment costs (procyclical effects, costs of accumulating capital). - Factors affecting costs: banking system profitability, existing voluntary buffers, ease of accessing market capital. - When monetary policy is tightening, building buffers can still be useful: - Tighter monetary policy often boosts bank profitability initially (transmission to loan rates tends to lead deposit rates), aiding retained earnings accumulation.25 - Evidence shows policy rate increases can substantially boost bank profitability through interest margins, though they can also cause losses on long-term securities and increased provisioning (ECB 2024; IMF 2023e). - Locking in profits through higher buffer requirements can be prudent (IMF 2024c). - Policymakers can make pace contingent on absence of detrimental side effects (for example, set a baseline pace to be enforced after a set period such as one or two years and retain flexibility to reconsider).26 - Monitor lending surveys and financial market stress indicators to assess whether to pause or slow buildup. ### C. What level for the positive neutral rate? - The appropriate neutral rate remains under research; initial approaches for authorities include: - Start from the desired benefit in generating policy space for a release: the buffer should be large enough relative to banks’ expected losses in stress so that its release can reduce amplification. - Simple stress-test-based calibration: - Aim for resilience to stress events more frequent than extreme FSAP-level stresses; calibrate buffer to absorb losses in a milder adverse scenario without causing contraction in loan growth. - Account for all nonreleasable buffers (CCB and structural SRBs) in the capital hurdle.28 - Historical peak-loss approach: - Use peak accumulated losses in past stress episodes as a yardstick for neutral and peak buffer calibration.29 - Distance between neutral and peak buffers depends on country capacity to identify cyclical buildups in real time and the ability to increase buffers when indicators warrant. - Country-specific factors arguing for larger neutral buffers: - Substantial stock vulnerabilities (high household leverage, high share of variable-rate or foreign-currency debt).30 31 - Constraints on monetary policy (pegs, currency unions) limit scope for monetary accommodation. - Small open and emerging market economies exposed to external shocks and volatile capital flows may want permanently larger buffers released only in large negative shocks.32 - Empirical findings: - Novel analysis finds losses in past banking crises can be sizable even without prior strong credit increases (Box 1); in advanced economies prior credit expansion is associated with higher peak losses, suggesting a neutral base with scope for further increases when cyclical indicators strengthen. - In emerging markets and low-income countries, no positive relationship is found between prior credit strength and crisis losses; twin or triple crises (currency or sovereign shocks) may imply higher neutral buffers in these countries, consistent with IMF (2014c). ### D. When to increase the CCyB beyond the neutral level? - Triggers for increase beyond neutral: - Any information implying the probability of future financial stress has increased relative to the standard risk environment (including qualitative information, e.g., expectations of future policy rate increases, structural changes, or sectoral reforms). - Quantitative guidance and indicators: - Early warning indicators: use a range of complementary indicators rather than a single measure (credit gaps, alternative gap measures, credit growth, asset price deviations, market risk measures, debt service burden, leverage, reliance on flighty funding, external imbalances).33 34 35 - Growth-at-risk models: capture tails of the output distribution and financial amplification effects; useful in mapping dynamic interactions between credit, financial conditions, and real sector risk. Models show easing financial conditions increase tail risks to output growth. (Adrian, Boyarchenko, and Giannone 2019; Adrian and others 2018; Brandão-Marques and others 2020) - Countercyclical stress testing: vary severity with economic and financial conditions to inform CCyB calibration (examples: Central Bank of Ireland 2022; Bank of England 2016).36 37 - Decision process: - Do not rely mechanically on indicators; apply judgment informed by a range of indicators and qualitative information (guided discretion). IMFs prior guidance supports this approach. - Limits on reductions: - A moderation of cyclical risk indicators should not prompt a decrease of the CCyB below the neutral level unless release conditions are fulfilled. The neutral level is a floor to avoid needlessly giving away resilience and causing overly frequent releases/increases that reduce future release effectiveness.38 ### E. Assessing and addressing leakage effects - Leakage risks: - Domestic leakage from banks to bank-affiliated nonbanks when shareholders shift lending to circumvent higher buffer requirements. - Regulatory tightening on banks can favor nonbanks, causing leakage (Buchak and others 2018). - Cross-border leakage via foreign branches in absence of reciprocity agreements. - Leakage can also occur when capital requirements increase during stress (Bednarek and others 2023). - Addressing leakage: - Expand scope of macroprudential tools to capture activities across banks and affiliated nonbanks (consolidation of buffer requirements by activity). - Apply borrower-based and liquidity tools to nonbank providers of credit to maintain a level regulatory playing field. - Use reciprocity arrangements (as in BCBS CCyB agreement) to address cross-border leakage and negotiate arrangements where needed. - Domestic authorities could consider applying a higher domestic rate to outward exposures while reciprocating host authority actions in response to local conditions. - Examples: - Croatia addressed increased bank-affiliated leasing company credit by regulations capturing that activity within macroprudential scope (IMF 2014a). *IMF Departmental Paper excerpt: “Rethinking Macroprudential Capital Buffers,” Section 4.* ### Box 1. Using Past Crises to Guide the Calibration of the Countercyclical Capital ### Box 1. Using Past Crises to Guide the Calibration of the Countercyclical Capital Buffer in Advanced Economies and Emerging Market Economies ### Data and methodology - Sample: - 27 advanced economies (AE), 29 emerging market economies (EME), and 11 low-income countries (LIC). - Data sources: - Laeven and Valencia (2020) for past crises’ peak shares of NPLs. - IMF Monetary and Financial Statistics database, Bank for International Settlements, and IMF Financial Soundness Indicators database for credit growth. - Latest available observations for Total Gross Loans (range of dates 2013:Q4 to 2023:Q3) and risk-weighted assets (range of dates 2018:Q2 to 2023:Q3) from the IMF Financial Soundness Indicators database. - Measurement and grouping: - Real credit deflated by the consumer price index. - Maximum year-on-year real credit growth in the five years before the crisis computed for each country. - Countries classified as "low credit growth" if maximum credit growth before the crisis is below (or at) the median; otherwise "high credit growth". - For countries appearing more than once in the Laeven and Valencia (2020) dataset, the maximum peak NPL level is selected; average peak NPLs are then calculated for each country group. - Loss and RWA calculations: - Peak losses as a share of RWA are calculated as (Peak NPL × Average Share of Total Gross Loans to RWA). - Average Share of Total Gross Loans to RWA is calculated as (Total Gross Loans × 15 percent)/RWA for AEs and (Total Gross Loans × 25 percent)/RWA for EMEs. Latest observations for Total Gross Loans and RWA are used. - Recovery assumptions (loss-given-default assumptions): - Assume 85 percent of advanced economy NPLs will be recovered. - Assume 75 percent of EME and LIC NPLs will be recovered. ### Empirical findings - Peak NPL patterns: - Peak NPLs in past crises were lower on average for advanced economies than for EMEs or LICs. - For advanced economies, higher credit growth ahead of the crisis is associated with higher peak NPLs; the association is much weaker for EMEs. - For LICs, peak NPLs have been higher in crises following low credit growth compared to crises following high credit growth. - Among LIC low credit growth episodes, three out of five relate to simultaneous banking and currency or banking and sovereign crises. - Explanations include unobserved increases in bank exposures to the sovereign and high inflation reducing measured real credit growth. - Correlations: - Across all crisis episodes, the correlation between past credit growth and the share of NPLs in total loans is positive 0.3. - For EMEs the correlation is just above zero and not statistically significant. - Credit growth dynamics around crises: - The drop in credit growth after banking crises is sizable in advanced economies and even larger in EMEs. - The difference in average credit growth during the four years before the crisis and three years after the crisis was calculated for 27 AE, 28 EME, and 9 LIC observations. ### Buffer estimates (indicative peak buffer needs) - Procedure: - Multiply average peak NPL percentage (from Laeven and Valencia) for low and high credit growth groups with averaged gross loans to RWA for AEs, EMEs, and LICs; adjust for recovery assumptions to estimate losses as a share of RWA. - Indicative average buffer ranges to cover losses from peak NPLs: - Advanced economies: a buffer between 3 to 7.5 percent would be needed to cover losses from low or high credit growth crisis respectively. - EMEs: a 6 to 7 percent buffer would have covered losses in low or high credit crisis periods. - LICs: a buffer between 5 and 9 percent would be needed. - Calculation note: - Peak losses as a share of RWA use the formula and assumptions described above, and rely on latest available Total Gross Loans and RWA observations and peak NPL from Laeven and Valencia (2020). ### Policy implications and recommendations - For advanced economies: - It can be useful to assume buffer levels should respond strongly to credit conditions; the capital buffer framework could allow for increases in releasable capital buffers in response to strong credit flows. - For EMEs: - Emphasis could be more on calibrating larger neutral (permanent) buffers rather than relying on strong cyclical responsiveness to credit growth, given the weaker link between pre-crisis credit growth and peak NPLs. - For LICs: - Given higher incidence of currency and sovereign crises and instances where crises with low measured credit growth produced high NPLs, well-capitalized banks and potentially permanently higher capital buffers may be warranted to withstand large shocks. - Releasable buffers vs higher minimums: - The benefit of releasable buffers (rather than higher minimum requirements) is underscored by the sizable post-crisis credit slowdowns, especially in EMEs. - Need for country-specific analysis: - Absolute numbers are indicative and depend on assumptions and data; further analysis with country-specific data and more granular stress tests (using current balance sheet composition, default rates, and recovery values) is needed to arrive at robust buffer estimates. - Data limitations and caution: - If reliable signals of incipient vulnerabilities are scarce, building resilience with permanently higher capital buffers that can be released in extreme shocks may be preferable. *Source: Box 1, "Using Past Crises to Guide the Calibration of the Countercyclical Capital Buffer" (IMF Departmental Papers — Rethinking Macroprudential Capital Buffers).* ### 5. Considering the Relationship with ### 5. Considering the Relationship with Other Capital Requirements ### A. Provisioning and Dividend Restrictions as a Complement to Capital Buffer Requirements - Two categories of measures to counter procyclicality: - Measures targeting capital ratio requirements (e.g., releasable capital buffer requirements). - Measures targeting banks’ capital levels (provisioning and capital distribution rules / dividend restrictions). - Expected credit loss accounting regime: - Introduced post-global financial crisis to recognize credit losses earlier. - In the expected credit loss framework (IASB 2014), banks account for expected losses in addition to historical and current losses. - Banks use forward-looking models to forecast expected future losses; forecasts are translated into provisions subject to some management discretion. - This produces profit and loss impacts ahead of aggregate stress, smoothing profits through the financial cycle and reducing likelihood of hitting regulatory capital constraints in downturns (BCBS 2022a). - Provisioning for expected loss may not cover all losses in a downturn; releasable capital buffers can cover unexpected losses from sudden exogenous shocks. - A releasable CCyB can help absorb sudden increases in expected credit loss provisions during exogenous shocks, supporting banks’ ability to provide financing. - Dynamic provisioning (Spain example): - Builds countercyclical macroprudential buffers through provisioning rules; forward-looking “dynamic-provisioning” approach built general provision buffers in expansion for drawdown in bad times (Jiménez and others 2017). - Dynamic provisions can be released to cover specific provisions for incurred losses, either via predetermined formula or discretion (example: Panama during the pandemic). - Parameters often set a ceiling for accumulation of provisions; where ceiling is too low, a releasable capital buffer such as CCyB can complement dynamic provisions. - Dividend/distribution restrictions: - Aggregate restrictions on dividend payouts can preserve capital during stress and complement buffer release to prevent opportunistic payouts (Nier and Olafsson 2020). - Distribution restrictions can raise legal concerns if supervisors lack authority to restrict dividends; power often rests with bank boards (Awad and others 2020). - For preparing for future stress, increases in capital buffer requirements may be preferred to distribution restrictions. ### B. The Countercyclical Capital Buffer and Alternative Releasable Buffers - Basel Committee’s revised Core Principles (BCBS 2024a) recommends authorities be empowered to require banks to build releasable capital; authorities can choose the most suitable tool. - Main alternatives to a broad-based CCyB are sectoral tools or tools that build larger buffers for particular exposures. - Sectoral CCyB: - Basel Committee (BCBS 2019c) provides principles for sectoral CCyB implementation. - Swiss authorities used a sectoral CCyB against housing market vulnerabilities (Swiss National Bank 2022). - A sectoral CCyB can be released in aggregate shocks and implemented with a positive neutral setting. - Sectoral capital buffers more generally: - Common in Europe; have been released during COVID-19 where available (example: Poland). - Affect capital needed to support lending to a specific sector; can make lending to targeted sectors less attractive. - Examples: SRB applied to all real estate exposures in Belgium and Germany; SRB used to cover sectoral or broad-based time-dimension risks after EU CRD revision. - Can be designed as increasing in concentration of exposures (France) or to increase resilience to sovereign exposures (recommended in IMF FSAPs for Italy, Romania, South Africa). - Use cases and combinations: - Sectoral buffers helpful where borrower-based tools are unavailable or insufficient (example: Germany implemented sectoral SRB on residential real estate loans in addition to CCyB). - Sectoral buffers can target FX loans where borrower-based tools for indirect FX credit risk are hard to apply. - Combining sectoral buffer and broad-based CCyB can balance targeted risk mitigation and systemwide protection (examples: Germany; recommended in Belgium and Luxembourg FSAPs). - Simpler binary CCyB design for capacity-constrained jurisdictions: - Where data/capacity limit active CCyB management, a binary design could set the buffer at a positive neutral level most of the time and zero in periods of substantial aggregate stress. - Authorities can change the positive buffer level in line with perceived changes in the risk environment. - Implementation notes: - Buffers must be expressed as a fraction of risk-weighted assets. - Implementation does not require sophisticated RWA approaches; it requires supervisory enforcement mechanisms and a decision-making process for release. ### C. The Relationship between the Countercyclical Capital Buffer and the Capital Conservation Buffer - CCB and CCyB are complements; both absorb losses in stress. - CCyB: protect economic activity in aggregate stress. - CCB: allow dipping into buffers for bank-specific/idiosyncratic losses, enabling laddered supervisory responses. - Practice during COVID-19: - In some countries, CCBs (usually amounting to a maximum of 2.5 percent of risk-weighted assets) were released at the onset of the pandemic, contrary to Basel Framework guidance (BCBS 2019d). - Releasing the CCB removes an important supervisory instrument to control individual banks during long stress periods. - Interaction and combined buffer concept: - Outside aggregate stress, a positive neutral CCyB adds to the combined buffer requirement (CCyB + CCBs). - The combined buffer functions as an expanded CCB when required combined levels exceed 2.5 percent, triggering dividend restrictions when banks dip into the combined buffer. - A larger combined buffer can protect against idiosyncratic risks and reduce run risk for weaker institutions. - Theory and policy implications: - Drechsler and others (2023) model: banks hedge interest rate risk with deposit franchise value; rising rates can increase run risk on uninsured deposits when asset values decline; issuing more capital as rates rise can address this. - Positive neutral CCyB can complement CCB in normal times and potentially reduce the need for increases in deposit insurance coverage. - Replacing parts of high minimum requirements with releasable buffers: - In some emerging market and developing economies, minimum requirements beyond Basel serve macrofinancial objectives; authorities could consider replacing some of these with explicitly releasable capital buffers. ### D. Neutral Countercyclical Capital Buffers and Bank-Specific Buffers Driven by Stress Tests - Stress tests inform supervisory guidance on bank-specific capital expectations (examples: Pillar 2 Guidance in euro area; Stress Capital Buffer in United States). - BCBS (2019b): stress test results commonly used to assess capital adequacy across Basel jurisdictions, though integration into supervisory capital expectations differs. - Supervisors should account for a positive neutral CCyB when calibrating bank-specific expectations informed by stress tests: - Where stress tests cover cyclical variation of capital, it is natural to consider the positive neutral CCyB rate in calibrating bank-specific expectations. - Supervisory guidance can remain bank-specific and stress-test guided, but stress scenarios applied in normal times may assume release of the CCyB in the stress scenario. - The “hurdle rate” in stress tests could assume banks would still meet CCBs and any additional requirements for systemically important institutions, but not the CCyB during aggregate stress. - Rationale: - This avoids double-counting macroeconomic risks between Pillar 2 requirements and CCyB, keeping their targets distinct as intended by the Basel Framework. - Pillar 2 buffers remain heterogeneous and positive even if aggregate CCyB has been released; bank-specific increases in Pillar 2 buffers for idiosyncratic reasons remain legitimate even during broad-based buffer release. ### E. Neutral Countercyclical Capital Buffers and Leverage or Resolution Requirements - Release of capital buffers can cause unweighted capital requirements to become binding constraints, undermining buffer release effectiveness. - Example: In advanced economies where banks have portfolios with low risk weights (e.g., mortgages), lowering a CCyB could make the leverage ratio requirement binding. - Resolution-related requirements (total loss absorbing capital requirements) can similarly become binding for systemically important institutions. - Calibration implications: - Presence of unweighted capital requirements should be considered when calibrating releasable capital buffers. - A larger releasable buffer may be chosen if unweighted requirements would otherwise hamper effectiveness (see Herrera, Scalone, and Pirovano 2024). - Need for international work: - More work is required at the international level to ensure buffer release is effective and not hampered by unweighted requirements (BCBS 2022a; ESRB 2021, 2022; Leitner and others 2023). *Source: IMF Departmental Paper — Rethinking Macroprudential Capital Buffers, Chapter 5.* ### Box 2. Positive Neutral Countercyclical Capital Buffers and Reciprocity ### Box 2. Positive Neutral Countercyclical Capital Buffers and Reciprocity ### Basel framework and reciprocity rules - The Basel framework stipulates that the countercyclical capital buffer (CCyB) should be determined at the national level for all exposures to borrowers in that country. - The framework mandates international reciprocity of CCyB rates up to 2.5 percent of risk-weighted assets, such that the same CCyB rate applies to exposures of foreign banks. - Authorities may choose CCyB rates above 2.5 percent; reciprocation of such CCyB decisions could take place on a voluntary basis. - The European Union framework, operated based on recommendations by the European Systemic Risk Board, is presented as an example of voluntary reciprocity arrangements. ### Effects of a positive neutral CCyB on reciprocated capital - A positive neutral CCyB framework gives rise to higher reciprocated capital levels allocated by foreign firms than the conventional zero-neutral CCyB framework because: - Foreign firms can be expected to fully reciprocate the positive neutral level (assuming this is below 2.5 percent). - If the rate chosen by the host country tracks above 2.5 percent, foreign banks operating in the country must hold at least 2.5 percent of capital for the exposures in the destination country, even if voluntary reciprocity is not forthcoming. - If the CCyB starts from a lower level, the reciprocated capital will therefore ultimately always be less than in a positive neutral framework. ### Considerations for non-Basel members and cross-border presence - Countries that are not Basel members and have a significant presence of foreign branches want to make sure that reciprocation arrangements are well established. - When there is a significant presence of foreign branches or cross-border lending, authorities may want to make sure that reciprocity arrangements are in place to sustain a level playing field in the banking sector following CCyB decisions. - Supervisors often require foreign banks to establish a presence as a subsidiary to maintain close scrutiny; this removes the need for reciprocation, since chosen capital buffers can be directly enforced on these banks. ### Role of a positive neutral CCyB in building releasable capital buffers (from the paper’s conclusion) - A positive neutral CCyB framework is a key means to build releasable capital. In such a framework, the default setting of the CCyB is positive rather than zero. - Cyclical increases in systemic risk, as measured using early warning indicators, or other indications that risks will increase in future, could lead to increases in the buffer rate from its neutral setting. - When financial conditions deteriorate and banks face capital constraints that hinder lending, a release of the buffer is appropriate and should be maintained until such stresses have resolved. - As authorities release capital buffer requirements, they could also consider advising banks to refrain from profit distributions. ### Alternative and complementary approaches to releasable buffers - Authorities may prefer a sectoral buffer, generated by the sectoral CCyB or an SRB, useful where vulnerabilities emanate from particular exposures (examples given: loans collateralized by residential real estate, loans in foreign currency, or exposures to the sovereign). - Targeted use of such buffers can reduce the vulnerability over time. - Where active management of the CCyB based on a cyclical indicator (such as the credit gap) is not feasible or indicators are unreliable, consideration can be given to a simpler broad-based buffer that would not be varied actively unless the conditions for release are met. ### Implementation guidance and sizing of neutral buffers - Implementation of releasable buffers should not wait for elevated readings of the credit gap; reliance on the credit gap risks buffers being unavailable or insufficient in periods of stress. - Implementation can start as soon as conditions for a release are not or are no longer met, and the supply of credit by the banking system is not otherwise constrained by a lack of capital. - Where the system is profitable, or voluntary buffers are already available, releasable capital buffers can often be built at little cost. - Careful consideration should be given to phasing in new requirements so buffers are built gradually from retained earnings, avoiding procyclical effects from an increase in buffer requirements. - The analysis of the appropriate “neutral” buffer rate would focus on vulnerabilities in the existing stock of bank exposures rather than cyclical early warning indicators. - Examples of stock vulnerabilities: prevalence of variable rate loans sensitive to local currency interest rate tightening; large share of existing debt denominated in foreign currency sensitive to exchange rate changes. - Stress tests are recommended to assess the desirable size of releasable buffers quantitatively; they can condition on the range of vulnerabilities in current exposures and model risk factors affecting debt service burdens of borrowers (households, firms, and potentially the sovereign). - Larger buffers would likely be useful for small open and emerging market economies exposed to external shocks (volatile commodity prices, volatile capital flows), and where monetary policy is constrained (peg or currency union). - Emerging market economies may want to aim for larger neutral capital buffer requirements compared to the average advanced economy, based on evidence on losses in past banking crises. ### Key statistics and early adopters of a positive neutral CCyB rate (Annex Table 1.1) - Note: Information as of January 2025. CCyB = countercyclical capital buffer. - Armenia 1.5 percent - Australia 1 percent - Chile 1 percent - Cyprus 0.5 percent - Czech Republic 1 percent - Estonia 1 percent - Georgia 1 percent - Greece 0.5 percent - Hong Kong SAR 1 percent - Hungary 1 percent - Iceland 2–2.5 percent - Ireland 1.5 percent - Latvia 1 percent - Lithuania 1 percent - New Zealand 1.5 percent - The Netherlands 2 percent - Poland 2 percent - Portugal 0.75 percent - Slovenia 1 percent - South Africa 1 percent - Spain 1 percent - Sweden 2 percent - United Arab Emirates 0.5 percent - United Kingdom 2 percent *Source: IMF DEPARTMENTAL PAPERS • Rethinking Macroprudential Capital Buffers.* ### 1. https://www.bafin.de/SharedDocs/Veroeffentlichungen/EN/Pressemitteilung/2022/ ### Rethinking Macroprudential Capital Buffers ### Major thematic areas in the referenced literature - Operational and policy frameworks for the countercyclical capital buffer (CCyB) - Basel Committee on Banking Supervision (BCBS). 2010a. “Guidance for National Authorities Operating the Countercyclical Capital Buffer.” - Bank of England. 2016. “The Financial Policy Committee’s Approach to Setting the Countercyclical Capital Buffer: A Policy Statement.” - DeNederlandsche Bank. 2022. “Analytical Framework for Setting the Countercyclical Capital Buffer in the Netherlands.” February. - European Central Bank (ECB) and European Systemic Risk Board. 2025. “Using the Countercyclical Capital Buffer to Build Resilience Early in the Cycle.” Joint ECB/ESRB Report on the Use of the Positive Neutral CCyB in the EEA, January 2025. - BCBS. 2019c. “Guiding Principles for the Operationalisation of a Sectoral Countercyclical Capital Buffer.” - National Bank of Belgium. 2022. “Introduction of a Sectoral Systemic Risk Buffer for IRB Belgian Residential Real Estate Exposures on the Basis of Art 133 CRD: Additional Explanation.” - Evaluation, reform, and empirical evidence on capital buffers and Basel III - BCBS. 2010b. “An Assessment of the Long-Term Economic Impact of Stronger Capital and Liquidity Requirements.” - BCBS. 2017. “High-Level Summary of Basel III Reforms.” - BCBS. 2022b. “Buffer Usability and Cyclicality in the Basel Framework.” - BCBS. 2023. “Report on the 2023 Banking Turmoil”. - BCBS. 2022c. “Evaluation of the Impact and Efficacy of the Basel III Reforms.” - BCBS. 2024a. “Core Principles for Effective Banking Supervision.” - BCBS. 2024b. “Range of Practices in Implementing a Positive Neutral Countercyclical Capital Buffer.” - Empirical studies on CCyB effects on credit, lending, and the macroeconomy - Bedayo, Mikel, and Jorge E. Galán. 2024. “The Impact of the Countercyclical Capital Buffer on Credit: Evidence from Its Accumulation and Release before and during COVID-19.” - Behn, Markus, Marco Forletta, and Alessio Rhegezza. 2024. “Buying Insurance at Low Economic Cost—The Effects of Bank Capital Buffer Increases since the Pandemic.” ECB Working Paper 2024/2951. - Couaillier, Cyril, Marco Lo Duca, Alessio Reghezza, and Constanza Rodriguez d’Acri. 2022a. “Caution: Do Not Cross! Capital Buffers and Lending in COVID-19 Times.” - Couaillier, Cyril, Alessio Reghezza, Constanza Rodriguez d’Acri, and Alessandro Scopelliti. 2022b. “How to Release Capital Requirements During a Pandemic? Evidence from Euro Area Banks.” ECB Working Paper 2720. - Dursun-de Neef, Özlem, Alexander Schandlbauer, and Colin Wittig. 2023. “Countercyclical Capital Buffers and Credit Supply: Evidence from the COVID-19 Crisis.” - Jiménez, Gabriel, Steven Ongena, José-Luis Peydró, and Jesús Saurina. 2017. “Macroprudential Policy, Countercyclical Bank Capital Buffers, and Credit Supply: Evidence from the Spanish Dynamic Provisioning Experiments.” - Buffer usability and releases during crises (COVID-19 focus) - Bank of England. 2020. “Bank of England Measures to Respond to the Economic Shock from COVID-19.” London, March 11. - Mathur, Aakriti, Matthew Naylor, and Aniruddha Rajan. 2023. “Useful, Usable, and Used? Buffer Usability During the COVID-19 Crisis.” Bank of England Working Paper 1011. - Berrospide, Jose M., Arun Gupta, and Matthew P. Seay. 2021. “Un-used Bank Capital Buffers and Credit Supply Shocks at SMEs during the Pandemic.” Finance and Economics Discussion Series 2021-043, Federal Reserve Board. - Wong, Eric, Kelvin Ho, Andrew Wong, and Vincent Lo. 2022. “The Effects of COVID-19 Support Measures on Bank Lending: Lessons from the Release of the Countercyclical Capital Buffer and Loan Guarantee Schemes in Hong Kong.” Research Memorandum 03/2022, Hong Kong Monetary Authority. - Sivec, Vasja, and Matjaž Volk. 2022. “Empirical Evidence on the Effectiveness of Capital Buffer Release.” Banka Slovenije Working Paper 1/2022. - Conceptual, measurement, and indicator work for CCyB activation and calibration - Drehmann, Matthias, Claudio Borio, Leonardo Gambacorta, Gabriel Jimenez, and Carlos Trucharte. 2010. “Countercyclical Capital Buffers: Exploring Options.” BIS Working Papers 317. - Drehmann, Matthias, Claudio Borio, and Kostas Tsatsaronis. 2011. “Anchoring Countercyclical Capital Buffers: The role of Credit Aggregates.” - Detken, Carsten, Olaf Weeken, Lucia Alessi, Diana Bonfim, Maguel M. Boucinha, Christian Castro, Sebastian Frontczak, and others. 2014. “Operationalising the Countercyclical Capital Buffer: Indicator Selection, Threshold Identification and Calibration Options.” ESRB Occasional Paper 5. - Tölö, Eero, Helinä Laakkonen, and Simo Kalatie. 2018. “Evaluating Indicators for Use in Setting the Countercyclical Capital Buffer.” International Journal of Central Banking 14 (2): 51–111. - Behn, Markus, Ana Pereira, Mara Pirovano, and Alessandra Testa. 2023. “A Positive Neutral Rate for the Countercyclical Capital Buffer—State of Play in the Banking Union.” ECB Macroprudential Bulletin 24: 21. - Cross-sector and non-bank considerations, shadow banking, and substitution effects - Buchak, Greg, Gregor Matvos, Tomasz Piskorski, and Amit Seru. 2018. “Fintech, Regulatory Arbitrage, and the Rise of Shadow Banks.” Journal of Financial Economics 130 (3): 453–83. - Claessens, Stijn, Guilo Cornelli, Leonardo Gambacorta, Francesco Manaresi, and Yasushi Shiina. 2023. “Do Macroprudential Policies Affect Non-Bank Financial Intermediation?” International Journal of Central Banking 19 (5): 185–236. - Cizel, Janko, Jon Frost, Aerdt Houben, and Peter Wierts. 2019. “Effective Macroprudential Policy: Cross-Sector Substitution from Price and Quantity Measures.” Journal of Money, Credit and Banking 51: 1209–35. - IMF. 2014d. “Shadow Banking Around the Globe: How Large and How Risky?” (Chapter 2). In Global Financial Stability Report: Risk Taking, Liquidity, and Shadow Banking—Curbing Excess while Promoting Growth. Washington, DC, October. - Country and institutional applications, assessments, and mission notes - IMF country and FSAP reports listed include: Romania 2018; Italy 2020; South Africa 2022a; Germany 2022b; Finland 2023a; Belgium 2023b; Maldives 2023c; Jordan 2023d; Iceland 2023f; Luxembourg 2024a; Spain 2024b. - Central Bank of Ireland. 2022. “The Central Bank’s Framework for Macroprudential Capital.” Dublin. - Swiss National Bank. 2022. “Swiss National Bank Proposes Reactivation of Sectoral Countercyclical Capital Buffer at 2.5%.” Bern, January 26. ### Representative document types and years present in the reference list - Policy statements, working papers, technical notes, ECB/BCBS/ESRB reports, country Financial System Stability Assessments, IMF Departmental Papers, and academic journal articles dated across 2005–2025, including specific items from 2010, 2011, 2012, 2014, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023, 2024, and 2025. ### Notable exact figures and phrases extracted verbatim - “pm_2022_01_12_antizyklischer_Kapitalpuffer_en.html” - Swiss National Bank proposes sectoral CCyB “at 2.5%.” - ECB Macroprudential Bulletin 24: 21. - BCBS Working Paper 37. - Bank of England Measures to Respond to the Economic Shock from COVID-19. London, March 11. - ECB Working Paper 2024/2951. - IMF Departmental Paper series title present as: IMF DEPARTMENTAL PAPERS • Rethinking Macroprudential Capital Buffers *Source: Rethinking Macroprudential Capital Buffers (reference list provided in the content unit).* --- _Source: https://www.imf.org/-/media/files/publications/dp/2025/english/rmcbea.pdf_