How much bank capital best serves the public interest? The answer depends on where the net social value of bank capital is maximized. The benefits of capital include a reduced likelihood of bank failure and broader financial crises while the costs include more expensive credit, leading to slower GDP growth.
In December 2025, the Bank of England’s Financial Policy Committee (FPC) updated its assessment of the optimal level of system-wide tier 1 capital, lowering it from around 14 percent of risk-weighted assets to around 13 percent. The update is a rare example of a major regulator revisiting an earlier estimate of optimal capital using the same broad framework but updated data, improved risk measurement and more than a decade of post-crisis experience. This benchmark refers to the level of system-wide tier 1 capital that the FPC judges to best serve the public interest. It is distinct from the binding capital requirements applied to individual UK banks, which are set through Pillar 1 rules, Pillar 2A add-ons and buffer frameworks. The FPC has also identified several areas of the capital framework, including buffer usability, the leverage ratio and the interaction of domestic capital requirements, for further review through 2026. The revised benchmark does not by itself change the requirements that bind individual banks. Any pass-through to those requirements would come through separate channels, most notably the roughly half a percentage point reduction in Pillar 2A that the FPC expects to follow from the implementation of Basel 3.1.
This note describes the FPC’s analytical framework for assessing optimal capital, how its original 2015 estimate was formed and what changed in the 2025 update and why.
The FPC’s Analytical Framework: 2015 Baseline and 2025 Update
The FPC set out its analytical foundations for its estimate of optimal capital in a supplement to its December 2015 Financial Stability Report.[1] There, it applied a UK-specific version of the standard “optimal capital” framework, rooted in the Basel Committee on Banking Supervision’s 2010 Long-term Economic Impact study, which weighs the lower probability of banking crises associated with higher capital against the resulting increase in the cost of bank credit.[2] A companion analysis, published as Brooke et al. (2015), updated the BCBS framework to reflect post-crisis institutional changes, most importantly credible resolution arrangements and time-varying capital buffers.[3] On that basis, the FPC estimated that the optimal amount of capital in 2015 was 10 to 14 percent of risk-weighted assets, lower than the BCBS’s own central estimate and those of several other studies in the literature.[4]
The FPC was explicit about why its range of estimate sat below much of the earlier “optimal capital” literature. The most important reason was that it gave substantial credit to post-crisis reforms, especially effective resolution arrangements, which the FPC judged would materially reduce the social cost of future bank failures. The FPC estimated that effective resolution alone reduced its estimate of optimal tier 1 capital by about 5 percentage points of risk-weighted assets. It also gave weight to stronger supervision, structural reforms and the intended active use of the countercyclical capital buffer (CCyB).[5]
In the FPC’s 2015 framework, the underlying optimal tier 1 requirement was around 11 percent of risk-weighted assets, assuming shortcomings in risk-weight measurement had been corrected. However, the FPC recognized that risk weights — as measured by banks’ internal models and the standardized approach — may understate the true risk levels of bank assets, meaning that a given capital ratio would provide less resilience than the headline number suggests. To compensate for these shortcomings, the FPC judged that the optimal level of tier 1 capital was about 13.5 percent of risk-weighted assets, with the additional 2.5 percentage points bridging the gap between measured and actual risk.
The FPC’s 2025 reassessment did not replace its original cost-benefit framework for determining “optimal capital” but rather updated it with new evidence on risk measurement, changes in bank balance sheet structure and a decade of experience with the post-crisis regime.
The evidence accumulated by the FPC since 2015 has mixed implications for its optimal capital calibration. On the cost side, the FPC notes that the spread between banks’ cost of equity and debt has narrowed in recent years as risk-free rates have risen. This makes it relatively cheaper for banks to increase their share of equity funding than was assumed in the earlier analysis. Some academic literature also suggests that the Modigliani-Miller offset may be larger than previously assumed (Gimber and Rajan 2019; Clark et al. 2023), which would further reduce the effect of higher capital on lending spreads. On the benefit side, the FPC also points to recent evidence that crisis-related GDP losses may have been more persistent than previously estimated (Romer and Romer 2019; Bonciani et al. 2021), which would increase the benefits of higher bank capital. Other changes point toward lower optimal capital. In particular, the FPC notes that UK household and corporate debt vulnerabilities have fallen since 2015.
The FPC did not publish a full recalibration of the 2015 model with 2025 values, so it does not quantify the net effect of these offsetting developments. Based on its analysis, the FPC determined that the underlying optimal estimate remains unchanged at 11 percent. This conclusion suggests that it views the forces as broadly offsetting each other. It is worth noting that in the 2025 update, the 11 percent underlying estimate is defined inclusive of the neutral UK CCyB, a change from the 2015 framework that affects comparability between the two vintages.
While the updated estimate of the core optimal level of tier 1 capital remains anchored at 11 percent, the FPC’s expected compensation for perceived shortcomings in risk measurement fell from roughly 2.5 percentage points to about 2.0 percentage points. This yields a revised estimate of optimal tier 1 capital of around 13 percent of risk-weighted assets. The FPC says this estimate lies within the range of capital levels likely to maximize expected long-term growth.
Reasoning for the FPC’s Updated Assessment of Optimal Required Capital
The FPC highlights three proximate drivers of the lower estimate of optimal capital: (1) improvements in risk measurement, (2) structural changes in banks’ risk profiles and (3) a decade of experience with the post-crisis regulatory framework.
Improvements in Risk Measurement. Basel 3.1, scheduled for UK implementation in January 2027, is central to the FPC’s reasoning on improvements in risk measurement. A revised standardized approach will make risk weights more sensitive to underlying risk, while an output floor will constrain variability across banks in the internal models used to calculate these weights.[6]
Structural Changes in Bank Balance Sheets. Since 2016, average risk weights at major UK banks have declined by roughly 7.5 percentage points, driven by shifts toward lower risk-weighted exposures, changes in modeling approaches and improved underwriting. Lower risk weights reduce risk-weighted assets, which in turn reduce the amount of capital a given ratio requires. Applied to current balance sheets, the FPC’s previous 14 percent benchmark would correspond to approximately £60 billion less nominal capital than when it was first set. Some observers, including Aikman and Vickers (2026), have argued that falling risk weights should not automatically reassure regulators, since the key question is whether the decline reflects genuine de-risking or instead changes in banks’ internal models that reduce reported risk weights.[7] The FPC does not treat falling risk weights as independently reducing the optimal amount of capital, but as part of a broader picture in which the UK financial system’s risk profile has evolved since 2015.
Ten Years of Post-Crisis Framework Performance. The FPC is explicit that its estimate of optimal capital depends on confidence in the effectiveness of post-crisis reforms. In its 2025 review, the FPC reaffirms its reliance on the same three institutional pillars it identified in 2015: (1) credible and effective resolution arrangements, (2) effective supervision and structural reform and (3) active use of the CCyB. On resolution, the FPC points to the Bank’s resolution of SVB UK in 2023 and to the 2022 and 2024 Resolvability Assessment Framework reviews, which found that a major UK bank could enter resolution safely if needed.[8] On broader resilience, UK banks continued lending through the COVID pandemic, the energy shock resulting from Russia’s invasion of Ukraine and the market turbulence surrounding the Silicon Valley Bank and Credit Suisse failures. Finally, the CCyB was cut to zero twice, in 2016 and again in 2020, demonstrating its intended role as a flexible tool that can be released during stress to support lending. The FPC’s 2025 bank capital stress test reinforced that picture, showing significant aggregate headroom over hurdle rates at the trough of the test. The FPC’s update therefore leaves the 11 percent underlying optimal estimate unchanged, redefined to be inclusive of the neutral UK CCyB, and reduces the correction for perceived residual shortcomings in risk measurement from 2.5 to 2 percentage points.
Conclusion
The FPC has published an updated cost-benefit estimate of optimal capital with the underlying analytical work at a point when several jurisdictions are reviewing their capital frameworks. In the U.S., Federal Reserve Vice Chair for Supervision Bowman has described a “broad, careful review” of capital requirements focused on overlap and calibration, and the European Union is debating “capital simplification.”[9] As jurisdictions review the coherence of their own capital frameworks, the FPC’s analysis provides a reference point. It also underscores the related point made by Baer and Newell (2025) that a serious review of capital requirements should be guided by calibration and coherence, not by a predetermined commitment to “capital neutrality.”[10]
[1] Bank of England, Financial Stability Report, December 1, 2015.
[2] Basel Committee on Banking Supervision, An assessment of the long-term economic impact of stronger capital and liquidity requirements (Basel: Bank for International Settlements, August 2010).
[3] Martin Brooke, Oliver Bush, Robert Edwards, Jon Ellis, Benjamin Francis, Raj Harimohan, Katharine Neiss, and Casper Siegert, Measuring the macroeconomic costs and benefits of higher UK bank capital requirements, Bank of England Financial Stability Paper No. 35 (December 2015).
[4] The BCBS’s (2010) study estimated optimal capital at 12 to 15 percent of risk-weighted assets under a range of assumptions. For a survey of the broader literature, see Bank of England, Financial Stability in Focus: The FPC’s assessment of bank capital requirements (December 2025), Box C and Table A.
[5] As Covas (2017) noted at the time, this consideration of post-crisis reforms made the Bank of England’s analysis particularly well-executed relative to other estimates in the literature.
[6] Prudential Regulation Authority, PS1/26 – Implementation of Basel 3.1: Final rules (January 20, 2026); see also Bank of England, Financial Stability in Focus: The FPC’s assessment of bank capital requirements (December 2025), discussion of Basel 3.1 and improvements in risk measurement.
[7] David Aikman and John Vickers, “The Bank of England’s capital mistake?” VoxEU/CEPR, January 15, 2026.
[8] Bank of England, Financial Stability in Focus: The FPC’s assessment of bank capital requirements (December 2025); Bank of England, Resolvability assessment of major UK banks: 2022 (June 10, 2022); Bank of England, Resolvability assessment of major UK banks: 2024 (August 6, 2024).
[9] https://www.ecb.europa.eu/press/pr/date/2025/html/ecb.pr251211~aa5c9271b8.en.html
[10] Greg Baer and Jeremy Newell, “The Problems With ‘Capital Neutral’,” Bank Policy Institute, July 21, 2025 – https://bpi.com/the-problems-with-capital-neutral/
