Finance: Research, Policy and Anecdotes

Important disclaimer: I am a member of the Advisory Scientific Committee of the ESRB and was co-chair of the ESRB task force on restrictions of distributions that helped draft this recommendation. However, the views expressed below are exclusively mine and do not necessarily reflect the official stance of the ESRB or its member institutions.

The ESRB issued a recommendation today extending and amending the existing Recommendation on pay-out restrictions, after the discussion at the General Board meeting on Tuesday (the same day the ECB’s Supervisory Board issued its recommendation on pay-out restrictions). It amends the previous recommendation from 27 May, which would have lapsed on 31 December.    This recommendation comes at the end of an intensive consultation process and among a lively debate between bankers, regulators and academics on the usefulness of such restrictions.  

There are valid and important arguments on both sides: on the one hand, the COVID-19 crisis is still ongoing, and uncertainty remains about the future impact on the economy and financial institutions, with a risk of further worsening of health and economic conditions. Markets and authorities lack information on the long-term impact of the crisis on the financial sector and credit markets. Financial institutions also remain strongly dependent on public policy support. This calls for an approach that aims at maintaining a sufficiently high level of capital to mitigate systemic risk and contribute to economic recovery. On the other hand, the uncertainty has a different character now than it had in the spring: we have moved from unknown unknowns to known unknowns. Further, regulators are aware of the importance of distributions in enabling financial institutions to raise capital externally, as rewarding investors for their investment is critical for the long-term sustainability of financial institutions and markets.  Extending the pay-out restriction with an expectation that it will be lifted in a few quarters could result in perverse incentives in that banks will hold back on lending and risk-taking to save capital space to pay out later, the opposite to what is the intention. Finally, restricting proper market functioning to allow for reallocation of capital across sectors and within the banking sector is certainly not in line with the need for further restructuring and consolidation in the sector. So, there is a clear trade-off or – in line with a long-standing tradition in the dismal science – good arguments on both sides. And beyond the economic arguments, there is the issue that the recommendation is not legally binding, so a degree of moral suasion is needed, which might carry only so far.

This trade-off if not tension has resulted in an amended recommendation by the ESRB, with the following highlights:

First, it is recommended that relevant authorities request financial institutions under their supervisory remit  to refrain until 30 September 2021 from (i) dividend payments, (ii) buy-back ordinary shares and (iii) creating an obligation to pay variable remuneration to a material risk takers, which has the effect of reducing the quantity or quality of own funds.  So, far this is in line with the previous recommendation.

Second, if, however, financial institutions decide to distribute profits, they are asked to apply extreme caution and that the resulting reduction in own funds does not exceed the conservative threshold set by their competent (i.e., supervisory) authority.

Third, the recommendation offers three criteria to competent authorities when setting this threshold: one, ensuring that financial institutions maintain a sufficiently high level of capital - including taking into account their capital trajectory - in order to mitigate systemic risk and to contribute to economic recovery; two, ensuring that the overall level of distributions of financial institutions under their supervisory remit is significantly lower than in the recent years prior to the COVID-19 crisis; three, an appropriate and possibly differentiated approach across the different sectors, banks, insurers and investment firms.

Fourth, CCPs are no longer included in the recommendation, as the stress test exercise regarding CCPs in the Union conducted by the European Securities and Markets Authority following the outbreak of the COVID-19 pandemic confirmed the overall operational resilience of Union CCPs to common shocks and multiple defaults for credit, liquidity and concentration stress risks.

So, here we are.  I am sure the debate on the merits and risks of such pay-out restrictions will not end.  While a rigorous analysis of the effects of these regulatory measures might be a far way off (and might not include all the different dimensions of this multi-faceted challenge), I am sure such analysis will be forthcoming at some point in the future.

The initial restrictions on profit distribution in the spring has already led to some analyses.  Here are two papers I find worth mentioning.  Leonardo Gambacorta, Tommaso Oliviero and Hyun Shin show in a recent BIS working paper that banks with a low price-to-book ratio have a greater propensity to pay out dividends, a correlation that is especially strong for banks with price-book ratios below 0.7.  This also explains why the ratio of cumulated dividends to retained earnings is particularly large for banks in the euro area, reaching 60% over 2007–19, against about 30% for banks in other advanced economies. Using data on dividend payments of previous years and based on capital-lending elasticities from the literature, the authors then show that a complete suspension of bank dividends in 2020 during the Covid-19 pandemic would have added, under different scenarios, an additional US$ 0.8–1.1 trillion of bank lending capacity in their sample, equivalent to 1.1–1.6% of total GDP across 30 advanced economies.  Obviously, this estimate is based on certain assumptions on the elasticity of lending to capital buffers and does not take into account bank reactions to dividend restrictions, so is certainly an upper limit.

My former colleague Steven Ongena co-authored an article for the European Parliament, assessing the potential increase in lending in the EU following different regulatory measures, including relaxing capital and liquidity requirements and restricting pay-outs. They document an overall reduction in capital requirements by 1.7 percentage points, which would translate into a 2.0–2.6% increase in lending to the real economy (as the previous paper these are based on capital-lending elasticities from the literature). Interestingly, the actual increase in lending was even stronger, driven by drawdowns of pre-committed credit lines and government loan guarantee schemes. The article argues for either a lifting of the pay-out restriction or a clear exit strategy from these restrictions as the policy uncertainty created by it might counter the positive effects of the other policies on lending.

Overview