A One-Way Ratchet: Buffers and the COVID Pandemic Experience

Introduction

A key feature of the post-crisis regulatory regime is the “buffer” framework.  A buffer is, technically speaking, not a strict regulatory requirement, but a strong “suggestion.”  In the case of capital buffers, banks are, in theory, allowed to maintain capital levels that are above the minimum requirement but below the level of the minimum plus buffers.  Regulators have included buffers in the regulatory framework to incentivize banks to “use their buffers” during downturns so that credit is not sharply curtailed, further exacerbating a downturn.  As pointed out by research from the Federal Reserve, the COVID experience shows that these buffers did not work as intended during the pandemic.  That is because the use of buffers by banks is largely seen as a signal of weakness by market participants, a signal that could impair a bank’s ability to maintain funding from its shareholders and creditors.  Because of these incentives, buffers cannot be used flexibly and should be regarded as just another component of “minimum requirements.”  In the remainder of this post, we discuss the large bank buffer regime, the experience with buffers during COVID, the underlying problem with the buffer framework, and implications for the countercyclical capital buffer.

Buffers, Buffers, Everywhere – The Large Bank Regulatory Regime

Financial Services Forum members are required to maintain a minimum amount of Tier 1 common equity capital (“capital”) as a ratio of total risk-weighted assets.  This minimum requirement is 4.5 percent.  A bank that maintains less than a 4.5 percent Tier 1 common equity ratio is in breach of its minimum requirement and would be subject to severe supervisory repercussions.  On top of this minimum requirement, Financial Services Forum members are subject to three additional capital buffers as depicted in Table 1.  One buffer is the GSIB capital surcharge buffer (GSIB), which averages 2.7 percent among Forum members.  Another buffer is the stress capital buffer (SCB), which averages 3.5 percent among Forum members, and the third and final buffer is the countercyclical capital buffer (CCyB).  This buffer varies over time and is currently set at zero percent.  All together, these three buffers amount to an additional 6.2 percent in Tier 1 common equity capital, larger than the minimum requirement itself.  The size and importance of these buffers is underscored by recent increases in the GSIB scores that will result in increased GSIB buffers absent any changes to the way GSIB buffers are calculated or actions by Forum members to significantly reduce these scores.  As we have discussed elsewhere, the increase in GSIB scores largely relates to COVID-related measures to support the economy and are unrelated to systemic risk, yet will have the effect of raising capital standards for large banks.

Source: Federal Reserve Board and FSF calculations. Buffers are reported as averages across FSF members.

To be precise, the additional 6.2 percentage points are not “required” as a strict regulatory matter. Rather, a large bank can maintain a capital level above 4.5 percent and below 10.7 percent (=4.5%+6.2%) and not be in breach of its regulatory capital requirement. At the same time, once a large bank exhibits a capital level below the minimum plus buffers (10.7% in our example), the bank is restricted in its ability to make capital distributions to shareholders. Moreover, the restrictions become more severe as capital levels move closer to the minimum requirement of 4.5 percent.

Buffers and the COVID Experience – Buffers Did Not Work as Intended

Recent research from the Federal Reserve Board, mentioned above, shows that buffers have not worked as intended during the pandemic.  More specifically, these researchers find that “[o]ne feature of the ongoing economic crisis is that large banks, both domestic and foreign, have shown considerable reluctance to lower their capital levels below these buffers, in spite of repeated encouragement by regulators to do so.”  This final point is worth emphasizing.  In the earliest stages of the COVID pandemic, regulators made public statements encouraging banks to make use of their buffers.  The clear lack of buffer usability presents a serious challenge to the large bank prudential framework because buffers play such a central role in the overall system.  A failure of buffers to work as intended represents a significant problem for the entire capital framework.

What Gives? Why Don’t Buffers Work As intended?

A natural question to ask is why buffers have not worked as intended during the pandemic?  One clear and convincing possibility is provided in another recent Federal Reserve Board research paper.  Specifically, the authors consider the fact that outside analysts would view the use of any buffer as a sign of weakness, leading to a ratings downgrade. Such a downgrade would impair a bank’s ability to raise funding and generate unwelcome instability as the entire market reassesses the stability and creditworthiness of a bank that has used its buffer.

These findings are important because they provide a clear and logical explanation for why buffers cannot be relied upon as a source of flexibility in the regulatory capital regime.  Simply put, banks do not operate in isolation and can’t act unilaterally, regardless of the signal that these actions send to the broader market.  Ultimately, bank regulation must recognize and account for the broader ecosystem in which large, global banks operate.  Importantly, during periods of stress market perceptions constrain bank behavior irrespective of regulatory flexibility.  Also, these events should cause regulators to re-evaluate whether markets can truly incorporate a complex buffer scheme into their evaluation of a bank’s financial health.  It may simply be the case that the need for simple rules and benchmarks by market participants and the public render a complex buffer framework impractical.

The implication of these findings is significant.  If it is the case that external market perceptions render buffers unusable, then buffers are simply another “minimum requirement” that serve to ratchet up capital requirements.  Accordingly, the economic cost of the buffer requirements increases as there is no real scope for flexibly using the buffers in practice.  Moreover, to the extent that regulators assessed the costs and benefits of the current framework by assuming that buffers had some benefit due to their potential flexibility in times of stress, recent events suggest that these benefits are overstated and the overall levels of capital required by the buffer framework may not adequately balance costs and benefits.

The COVID Experience: A Lesson for the Countercyclical Capital Buffer?

As shown in Table 1, the Countercyclical Capital Buffer (CCyB) is another capital buffer.  Currently, the CCyB is set to zero, but some support “turning it on” to a positive value.  The idea is that if the CCyB is non-zero, when a period of instability arises, regulators can reduce the CCyB and banks can reduce their capital levels to support lending.  This theory, however, must be considered in light of the recent experience on the non-usability of buffers.  In particular, the incentives created by the expectations of market participants during a period of instability may significantly limit the efficacy of any reduction in the CCyB.  Specifically, a bank that decides to reduces its capital during a period of stress may be perceived as being weak – even if the capital reduction is fully authorized and even encouraged by regulators.  Also, the complexities presented by asking market participants to correctly interpret the use of a buffer during a period of instability would surely be daunting.  The risk of misinterpretation and miscommunication would be large.  Ultimately, investors and analysts may care little about regulator statements just as they seemed not to be swayed by such statements during COVID.  Accordingly, an increase in the CCyB may only succeed in ratcheting up capital requirements for large banks without any attendant benefit because the promised flexibility is not available in practice.

Conclusion

Buffers are a central feature of the large bank regulatory regime.  The buffer system is complex, and for buffers to be effective, banks, investors, and other market participants must have a fine-grained understanding of that regime.  Recent evidence from the COVID pandemic suggests that the buffer system has not worked as intended by regulators because buffer use sends a negative signal to market participants.  Rather than providing flexibility to the capital framework, buffers ratchet up minimum capital requirements.  Regulators should take stock of recent experience with the buffer system and consider the appropriate role of buffers in the large bank regulatory framework – especially in relation to setting the CCyB.