The Bank of England’s capital mistake?

On 2 December 2025, the Bank of England’s Financial Policy Committee announced it would be lowering the capital requirement for UK banks. This column argues that higher macro-financial risk and sharply reduced fiscal capacity point to a need for higher rather than lower bank capital requirements. The most likely practical effect of this weakening of resilience will be higher payouts to bank shareholders, rather than increased lending to the real economy.

The financial crisis of 2008 underlined that the equity capital requirement for banks is a policy setting of fundamental importance.  On 2 December 2025, the Financial Policy Committee (FPC) of the Bank of England, which has responsibility for UK financial stability, announced it would be lowering the capital requirement for UK banks (Bank of England 2025). 

To many observers, this might look odd.  Since the last major review of capital settings in 2015, economic and financial risk has clearly increased. And the government’s fiscal capacity to address a future crisis has plainly diminished, if not evaporated.  This suggests that, on economic grounds, the financial stability regulator should, if anything, be tightening, not loosening, policy.  Indeed, using the Bank’s own analytical framework, we explain why we believe that the FPC has got it wrong.  Political and lobbying pressure to relax this key element of regulation should, in our view, have been resisted.  The most likely practical effect of this weakening of financial stability regulation will be higher payouts to bank shareholders, rather than increased lending to the real economy.

Bank capital regulation in a nutshell

Bank capital regulation is in practice a complicated and technical topic, with various types of capital, buffers, risk weightings, supervisory rules, and so on.  But at its core, it answers a simple question: how much of a bank’s assets should be funded by shareholders’ equity rather than borrowed money, including deposits? Much of this complexity, and associated regulatory burden, would be unnecessary if banks were simply required to have plenty of equity capital.  The complexity is unfortunate also because it is a barrier to public understanding of this major policy question.      

In a nutshell, bank capital regulation specifies the minimum level of equity capital funding (i.e. capital provided by shareholders) that a bank must have relative to its assets (loan exposures, etc.) in normal times.  This capital stands ready to absorb losses in times of stress, without public bailout, and with the bank continuing as a going concern.  Capital regulation is commonly expressed as a limit on leverage – that is, the ratio of assets to equity.  If leverage of up to 20 times is allowed, a bank that has lent 100 cannot pay out dividends that would reduce equity below 5.

The backstop maximum leverage limit in the UK framework is 1/(3.25%).Once the various leverage buffers that apply to systemically important banks are taken into account, the leverage limit for them is around 25 times. The primary capital requirement, however, is in terms of risk-weighted assets (RWAs), not simple leverage.  To arrive at RWAs, asset values are deflated according to estimated risk factors, and banks have to maintain normal-times capital requirements as a percentage of the RWAs.  From 2015, the FPC regarded 14% of RWAs as an appropriate benchmark.  It now judges that 13% is appropriate – a reduction of 1/14 = 7%.

Whether the move down to 13% is wise turns on two questions:

1. Was the 2015 benchmark well set?
2. How do events since then affect optimal policy?

In brief, our answers to these questions are: “No”; and “In any case, subsequent developments point to higher rather than lower bank capital requirements”.

The supposed GDP trade-off

Estimates of optimal capital are typically based on a supposed trade-off in terms of expected GDP between higher capital requirements having the benefit of reducing the risk, and hence the cost of future banking crises, and the cost of reducing normal-times output. 

The existence of the benefit is clear; that of the cost is not.  The economic logic of Modigliani and Miller (1958), a cornerstone of corporate finance theory, casts doubt on the proposition that a higher proportion of equity funding increases the overall cost to the economy of bank lending.  More equity (i.e. less leverage) means less risk, which reduces funding cost in a way that offsets the risk premium on equity.  Equity funding might nonetheless be relatively costly to banks for tax, implicit subsidy, and other externality reasons.  Empirical evidence does not show a negative relationship, in the relevant range, between equity capital and lending costs in steady-state, as distinct from short-run transitional periods. So there might well be no long-run trade-off at all, but we will suppose in what follows that there is.

The FPC’s view of the trade-off is depicted in its Chart 11 (reproduced here as Figure 1).  The first thing to notice is that the chart provides no justification at all for the policy change.  The net benefit from unchanged policy of 14% is at least as great as from weakened policy even if one were to accept all the assumptions underlying the analysis.  In its own words, the FPC’s “updated benchmark is within, albeit towards the lower end of, the range of capital requirements that are likely to maximise expected long-term growth” (emphasis added).  It is unclear why a financial stability regulator should choose the lower, and hence riskier, end of its own range.

Figure 1 The net macroeconomic costs of reducing capital could increase sharply below the estimated optional range

The 2015 benchmark

To question (1) above, our view is that the benchmark that the Bank derived from Brooke et al. (2015) is not a sound basis for policy because it was founded on very questionable assumptions that had the effect of substantially lowering the estimate of optimal capital (e.g. Vickers 2016).  First, the net benefit of higher capital was considered as:

{Reduction in probability of crisis due to higher capital × Net present cost of a crisis} – {Reduction in output due to higher lending spreads}.

There should also be a term to reflect the reduction in crisis severity, not just crisis probability, from higher capital.  Second, the analysis adopted a risk-neutral view of gains and losses of GDP, whereas in fact there is risk aversion.  Without risk aversion there would be no equity risk premium in the first place.  Third, the 2015 study was calibrated on average risk conditions, not elevated risk conditions, which is akin to basing flood defences on average weather conditions.  Fourth, it assumed unrealistic optimism about the effectiveness of counter-cyclical capital policy.  Fifth, the analysis placed much faith in the assumption that resolution regimes would work well if called upon, which it was assumed would reduce the expected net present value (NPV) cost of a future crisis to 43% of GDP, compared with 63%, almost half as much again, in a then-recent Bank for International Settlements analysis (Basel Committee on Banking Supervision 2010).

Insofar as the FPC’s 2025 policy judgement is based on the Bank’s 2015 benchmark, it inherits these weaknesses.  

The 2025 reduction

To question (2) above, the main reason given for the lowering of capital requirements is: “the evolution in the financial system since the FPC’s first assessment, including a fall in banks’ average risk weights, a reduction in the systemic importance of some banks, and improvements in risk measurement”. 

But the fall in risk weights gives no basis for reducing capital relative to RWAs.  The level of capital in the system has already come down with the RWAs.  Moreover, big UK banks have not evidently become less systemically important.  The FPC also notes that banks in fact tend to operate with headroom over the required minimum, but that is little reason to reduce the regulatory minimum.

The Bank continues its faith in resolution working well, despite what happened in Switzerland (and regionally in the US) in 2023.  A striking feature of the FPC’s Chart 11 is the orange dotted line for a world without effective and credible resolution.  If extended up and to the right, that curve would peak at a much higher level of optimal capital.

On developments since 2015, the FPC notes analysis suggesting that the cost of bank capital has declined and that “various developments may have impacted the macroeconomic benefits of bank capital”.  This puts it very mildly indeed.  Macroeconomic risks, including now even to the global trading order, have increased markedly, and the UK’s fiscal position is much more stretched.  Rather than take comfort from the resilience of the banking system through the pandemic, we would draw the opposite lesson that the fiscal consequences of the government bailing out the economy, which shielded the banks, have substantially increased the likely cost of a future crisis. 

In sum, compared with ten years ago, bank equity capital would appear now to have considerably greater benefit, in which case its optimal level has gone up not down, and perhaps to a significant extent. 

Independent studies

The FPC’s benchmark is also “at the bottom end of the range of optimal capital estimates” in external studies, as the Bank itself notes.  Some of the estimates reported in its Table A are double the comparable FPC figure, and most are at least 5 percentage points higher.  Far from supporting policy loosening, the body of independent analysis therefore indicates that substantial tightening would be better.

The policymaking process

The difference between the policymaking processes for monetary and financial stability policy is striking.  The Monetary Policy Committee (MPC) has a high degree of transparency and individual accountability for committee members, and it attracts intense public scrutiny.  The FPC lacks these desirable attributes.  There are no votes, and so no dissents, and relatively little public scrutiny.  The published record of the meeting gives little sense of the nature of the discussion or whether any of the external members of the FPC resisted the decision.  This needs to change.

The FPC does, however, welcome the views of stakeholders, including academics, on the issues covered in its paper, and this column, which we plan to develop with further analysis, is in that spirit.

Conclusion

The primary effect of this relaxation in capital requirements is likely to be to boost bank payouts to shareholders (Aikman 2025).  The effects on lending and growth are likely to be small.  If banks wished to expand lending, they could have done so using their existing capital headroom. 

We see no compelling economic reason for the FPC’s loosening of bank capital policy.  On the contrary, the Bank of England might have made a capital mistake.

Source : VOXeu