FSIBriefs
No 1
Reflections on regulatory responses to
the Covid-19 pandemic
Claudio Borio and Fernando Restoy
April 2020
FSI Briefs are written by staff members of the Financial Stability Institute (FSI) of the Bank for International
Settlements (BIS), sometimes in cooperation with other experts. They are short notes on regulatory and
supervisory subjects of topical interest and are technical in character. The views expressed in them are those of
their authors and do not necessarily the views of the BIS or Basel-based standard setting bodies.
This publication is available on the BIS website (). To contact the BIS Media and Public
Relations team, please email press@. You can sign up for email alerts at
© Bank for International Settlements 2020. All rights reserved. Brief excerpts may be reproduced or
translated provided the source is stated.
ISSN 2708-1117 (online)
ISBN 978-92-9259-363-6 (online)
mailto:press@
Reflections on regulatory responses to the Covid-19 pandemic1
Highlights
Regulatory policy responses should seek to support economic activity while preserving the financial
system’s soundness and ensuring transparency.
The recommendation for banks to make full use of capital and liquidity buffers should go hand in hand
with restrictions on dividends and bonuses and clarity concerning the process for rebuilding them.
Flexibility in loan classification criteria for prudential and accounting purposes should be
complemented with sufficient disclosure on the criteria banks use to assess creditworthiness.
The publication of detailed guidance on the application of expected loss provisioning rules,
combined with sensible transitional arrangements, may constitute a balanced approach to
mitigating the unintended effects of the new accounting standards.
1. Introduction
The Covid-19 pandemic raises the prospect of a deep recession. Hence the unprecedentedly broad-based policy
response. To cushion the economic blow, the authorities have deployed not only monetary and fiscal policies but
also prudential regulation and supervision and other financial policies.
What types of financial measure have been implemented? What is their rationale? What are their pros and
cons? In this article we address these questions based on a simple framework and a set of principles that can
guide the assessment.
2. The measures and their rationale
Since the start of the Covid-19 crisis, prudential and related authorities have implemented a swathe of measures
to support the supply of credit to the economy (Table 1). Prudential authorities have taken steps that effectively
lighten capital and other regulatory requirements and/or result in a less stringent supervisory stance.
Decisions have either modified or interpreted more flexibly accounting rules and prudential criteria for the
classification and measurement of bank exposures affected by the crisis. And international standard setters and
organisations have provided additional operational relief by postponing the implementation of new standards (Basel
Committee on Banking Supervision (BCBS)) and have publicly supported similar steps at the national level (Group
of Twenty (G20), Financial Stability Board (FSB) and BCBS).
These measures share one characteristic: in order to support economic activity, they enlist policies that are
not primarily focused on economic stabilisation. For the aim of accounting is to provide a faithful
1 Claudio Borio (@) and Fernando Restoy (@), Bank for International Settlements. The authors
are grateful to Patrizia Baudino and Mathilde Janfils for sharing their database on national policy developments and to Christina Paavola
for administrative support.
guarantees classification
use buffers
representation of firms’ financial condition, while that of prudential regulation and supervision is to ensure the
resilience of banks and the financial system.
Selected policy measures Table 1
Jurisdiction
Government
Capital requirements
Asset Expected loss
provisioning
Dividends and other
payouts
Australia Yes Encouragement to
Lower Domestic
New guidance - Expectation to limit
New guidance,
Canada Yes
Stability Buffer,
Encouragement to
use buffers
Release CCyB,
New guidance
Introduction of
transitional
arrangements
Expectation to halt
increases
EU/SSM Yes (*) Encouragement to
use buffers
New guidance New guidance Expectation to halt
Encouragement to
use buffers
Adjust risk weights
of certain loans
United
Kingdom
Release CCyB,
Encouragement to
use buffers
New guidance New guidance Expectation to halt
United
States
Encouragement to
use buffers,
Adjust
supplementary
leverage ratio
New guidance,
Definition of
restructured debt
Optional
suspension,
Extension of
transitional
arrangements
Expectation of prudent
decisions, Smoothening
of automatic
restrictions
(*): conditions vary across member countries.
Source: Publications by national authorities. As of 7 April 2020.
The reason for such an approach is straightforward. The economy is facing a major shock, as
policymakers seek to put it in hibernation to address a public health emergency. Force majeure dictates that all
policies be deployed to attenuate the blow and avoid an economic and social collapse.
In doing so, however, the objectives of financial policies should influence the type and extent of the
adjustments. For one, if taken too far, the adjustments could backfire: a weaker banking sector would weaken the
economy. A solid and sound financial system is a prerequisite for sustainable growth. In addition, financial
policies could lose credibility. This could only be counterproductive once the emergency has passed. In fact,
asymmetric policies that simply ease standards in bad times but do not tighten them in good times could generate
excessive risk-taking in the long run (“moral hazard”).
This suggests three principles that could guide the assessment of the adjustments.
Principle 1: the adjustments should be effective in supporting economic activity. That should apply at
least to the crisis period, and preferably even beyond, when establishing the basis for a solid recovery.
Principle 2: the adjustments should preserve the health of the banking (financial) system. Banks should
remain sufficiently well capitalised, liquid and profitable to underpin sustainable growth.
Principle 3: the adjustments should not undermine the long-run credibility of financial policies.
Credibility is hard to gain and easy to lose. Compromising the policies excessively in the short run can create
serious long-term damage. From this perspective, adjustments should be, and seen to be, temporary.
Transparency is key in meeting this principle.
Japan Yes
Ye
s
Ye
s
- -
3. The principles at work
Inevitably, putting these principles to work will require judgment. Even so, they can help guide the
assessment of the measures. Consider, in turn, those pertaining to prudential regulation/supervision and accounting.
Prudential regulation and supervision
There are good reasons why prudential regulation and supervision may seek to encourage banks to support
lending in a downturn. The most compelling one is a “fallacy of composition”. It is entirely reasonable for a
bank to pull back its horns in a downturn: the fragility of borrowers is more apparent and, by itself, the individual
bank can have little impact on the economy at large. But if all banks retrench at once, they may amplify the
downturn, and possibly end up being even worse off, as expenditures and output are unduly constrained.
Moreover, such fallacies aside, banks may become too conservative in bad times, in reaction to being too aggressive
in good times. As the saying goes, as long as the music is playing, the dance goes on, until it suddenly stops. This
captures the essence of the excessive procyclicality of the financial system.
Such considerations sit at the heart of the systemic (or “macroprudential”) approach to regulation and
supervision, which has risen to prominence since the Great Financial Crisis (GFC). The idea is that regulation
should not consider the riskiness of institutions on a standalone basis, but rather as a group, and that the
institutions’ own actions can collectively influence the overall risk in the system.
In a first-best world, these considerations should be built into the prudential framework, avoiding
unforeseen ex post discretionary adjustments. Otherwise, there is a risk that the adjustments are taken too far,
undermining the soundness of banks and weakening the framework’s credibility. And, indeed, some provisions
have such a built-in character. But in an emergency, discretionary adjustments may be justified as a second best.
Various types of adjustment have pros and cons, fulfilling the three guiding principles to varying degrees.
The regulatory instrument best suited to supporting lending during a downturn is the Basel III
countercyclical capital buffer. The instrument is designed to induce banks to accumulate capital in good times so
that they can draw it down in bad ones. The aim is to strengthen banks’ resilience, but the tool may also have the
side benefit of dampening lending cyclicality. And while the adjustments are discretionary, based on
indicators of increasing vulnerabilities, they are part and parcel of the framework. One problem is that, unless the
authorities have activated the buffer with sufficient vigour in good times, the resources that can be released are
limited, taking the wind out of the measure’s sails. In fact, few countries utilised the buffer actively or anywhere
close to its full potential before Covid-19.
Basel III has introduced another tool: the conservation buffer. While designed to apply to banks on a
standalone basis (microprudential perspective) this buffer is also intended to be drawn down in bad times so as to
allow banks to keep up their intermediation function. This feature is consistent with the aim of supporting the
Arguably, a more borderline case is allowing banks to draw down, even if partially, bank-specific capital
add-ons. These add-ons include Pillar 2 requirements, whose breach would normally trigger severe penalties. Indeed,
if these add-ons were calibrated to compensate for deficiencies in Pillar 1 minima, their use would clearly be
inconsistent with our principles: in that case, the minimum requirement would roughly
2 Basel III includes also a liquidity buffer. This Liquidity Coverage Ratio is in the form of high-quality liquid assets (HQLAs) in relation
to potential short-term withdrawals. The requirement raises issues similar to those regarding the conservation buffer, except that it does not
involve automatic distribution restrictions. If the LCR dips below 100%, supervisors are expected to set banks a timeline for rebuilding it.
correspond to the point of non-viability. That said, certain components of Pillar 2 could be considered as buffers
designed to absorb losses in adverse scenarios while the bank remains viable; in this respect, they would be akin to
the conservation buffer. This general principle is embedded in the Basel standards (Internal Capital Adequacy
Assessment Programme). Moreover, in many jurisdictions, banks are required to build a supplementary buffer to
prevent the capital ratio from falling below a pre-specified threshold in a stress test scenario, equivalent to an
informal or supplementary Pillar 2
While useful, none of these measures are sufficient in themselves to achieve the desired goal. To varying
degrees, they increase the resources banks can draw on to lend, but they do not provide the incentive to do so.
Indeed, banks are typically rather reluctant to draw down the buffers.
One disincentive may stem from a lack of clarity in supervisors’ expectations. Despite regulators’ explicit
recommendations to use the buffers, banks may need to be confident that deadlines for rebuilding them will be
sufficiently flexible and consistent with the full normalisation of economic activity and capital markets. Greater
clarity would help. This is especially the case in jurisdictions where the banking sector was facing profitability
challenges before the Covid-19 crisis and has found it difficult to raise capital in financial markets on reasonable
terms.
Another, closely related, disincentive, which applies to the conservation buffer, is the associated
distribution restrictions. These pertain to the amounts banks can hand out to shareholders (eg dividends or share
buy-backs), to some quasi-debt holders (eg coupon payments on other capital instruments such as convertible bonds
eligible as Additional Tier 1) or to employees (variable remuneration). The first two restrictions could potentially
affect market valuations and impede issuance activity.
One way of mitigating some of these disincentives is to include blanket distribution restrictions,
unrelated to the size of the buffer. To be sure, such restrictions have limitations. In particular, they would still
reduce banks’ attractiveness to investors. However, across-the-board restrictions, independent of individual
banks’ condition, ensure that the buffer’s size does not influence incentives and limit the risk of signalling a bank’s
relative weakness. They are the price to pay for improving the consistency of programmes that seek to
ensure banks have both the resources and some incentive to put them to work (Carstens (2020)).
Even then, faced with a deteriorating environment, banks would have an incentive to retrench, rather
than to finance the economy. This suggests that forms of government guarantee are a crucial complement to
measures that bolster banks’ capital. Indeed, a number of countries have implemented them (Table 1). Such
guarantees need to be designed carefully, so as to limit moral
In addition, supervisors can adjust, and have adjusted, the guidance concerning the assessment of bank
assets’ quality. This involves removing possible regulatory disincentives, mainly related to the prudential
classification of exposures. The guidance provides a more flexible interpretation of the conditions for the
reclassification of loans upon contractual modifications as forborne or troubled debt, and for the assets to be
declared non-performing when such reclassifications are made. Similarly, supervisors have also clarified that
exposures benefiting from government guarantees or “payment holidays” need not be classified as non-
performing. That said, banks would still have to continue making a case-by-case assessment of borrowers’
creditworthiness when considering
3 That buffer is denominated Pillar 2 Guidance (P2G) in the euro area and Pillar 2 B (P2B) or the Prudential Regulator Authority (PRA)
buffer in the United Kingdom.
4 In addition, lightening the burden of supervision and regulation during the crisis can help reduce compliance costs and allow banks to
focus on the task at hand by releasing resources. Examples implemented in various jurisdiction include suspension of stress tests and
rescheduling of online inspections. As long as these measures are exceptional, they do not compromise banks’ soundness (Principle 2) or
the credibility of the prudential framework (Principle 3). The BCBS has applied the same logic to justify the decision to postpone the
implementation of the newest Basel III standards.
5 The BCBS has also issued guidance along the same lines (BCBS (2020)).
Such clarifications are certainly useful. They provide a reference for reclassification that prevents
excessive conservativism and can thus boost overall credit availability. Moreover, they can hardly damage the
prudential framework’s credibility to the extent that the criteria used are a reasonable interpretation of existing
rules as applied to an exceptional situation. The main challenge is to ensure that the guidance does not obfuscate
the reflection of the crisis on asset quality as reported on bank balance sheets (Principle 3). When supervisory
criteria rely extensively on banks’ own assessment of borrower solvency, it could be useful for supervisors to request
banks to disclose the concrete criteria employed (as done eg in Canada, OSFI (2020)). This would allow their
consistency with the authorities’ guidance to be assessed.
Accounting standards for expected losses
The aim of accounting standards is to provide for an accurate representation of a firm’s economic condition
in public financial statements. Meanwhile, that of prudential regulation is to promote a firm’s financial
soundness. These two objectives need not be always fully consistent. For example, when accounting
standards rely heavily on market valuations, which may be justified on transparency grounds, they can induce
excessive procyclicality in the financial system (eg Borio et al (2001), FSF (2009)). Mutually reinforcing liquidity-
price spirals come to mind. In turn, because prudential indicators, such as regulatory capital, rely on accounting
valuations, they may end up encouraging banks to behave more procyclically. Covid-19 has revived such concerns.
The tension between accounting and prudential regulation has been at the core of the debate on loan loss
provisioning (Restoy and Roldán (2009), Borio and Tsatsaronis (2004)). For a long time, prudential authorities
have encouraged accounting standard setters to modify their incurred loss provisioning rules. There was a concern
that the rules failed to recognise the increasing vulnerability of banks’ financial condition during economic
expansions, thereby encouraging further lending, only to record losses when borrowers were on the verge of
defaulting, thereby encouraging banks to retrench when it was too late. Partly in response, the two main accounting
codes, the International Financial Reporting Standards (IFRS) and the US Generally Accepted Accounting
Principles (US GAAP), have recently adopted a more forward- looking, expected loss approach for loan loss
provisioning. The two variants of the approach entered into force in January 2018 and December 2019, respectively.
Despite some disagreement on a few of their technical specifications,6 the new, more forward- looking
standard is a welcome step towards mitigating excessive It requires banks to start provisioning for
future expected losses before the borrower defaults, reducing the need to raise provisions when the event materialises.
This should smooth the profile of provisioning over the credit cycle.
However, by construction, the new scheme cannot smooth this profile in the case of a truly
unexpected shock such as the Covid-19 pandemic. Since provisions are for expected losses, none could have
been made prior to the event. And by requiring in current conditions that provisions be taken ahead of defaults and
hence be front-loaded, the approach actually amplifies procyclicality, very much as market
6 In particular, the classification of exposures in three categories, depending on their credit quality, with different rules to calculate expected
losses in each, could generate cliff effects that could generate procyclicality. See eg ESRB (2019).
7 Significant procyclicality would however still remain. As applied, the standard has an excessive point-in-time nature that induces
procyclicality. Prudential authorities can partly offset this through appropriate backstops or filters. See eg Borio (2019) and Restoy and
Zamil (2017) and references therein. Arguably, such backstops are a better and more transparent way of reconciling the objectives of
prudential regulation and accounting (Borio and Tsatsaronis (2004) and Restoy (2010)). In fact, in some respects accounting and good
risk management are not reconcilable given the underlying principles (Borio and Tsatsaronis (2006)).
prices do. The transitional arrangements that most jurisdictions have adopted under the original BCBS guidance
mitigate the problem only to some
Thus, there could be a case for exploring measures to further mitigate such unintended effects at this
exceptional juncture. Such measures, however, should not come at the expense of transparency. This could
weaken trust in financial statements, thereby aggravating, rather than alleviating, financial market pressure on
banks. The GFC has shown how disruptive such a loss of trust can be.
The authorities’ response has included one or more of the following initiatives. First, allowing banks
to temporarily suspend the application of the new standard. Second, enhancing existing arrangements so
as to temporarily sterilise the effect on regulatory capital. Finally, issuing pragmatic implementation guidance to
avoid excessively rigid interpretations that could unduly boost provisions.
Concretely, the guidance option establishes references for a pragmatic calculation of expected losses in a
context of a sharp but presumably transient contraction of economic activity and of bold policy responses by fiscal
authorities. Moreover, this may also include useful references for the calculation of expected losses for
exposures that benefit from government measures such as guarantees or payment holidays. Specifically, the
authorities have clarified that these support measures, by themselves, should not automatically trigger a
reclassification across accounting categories and that public guarantees should moderate expected losses. The
reclassification would depend on the extent to which the crisis damages the borrower’s
The measures vary in merit, as they strike a different balance between the impact on procyclicality and a
transparent recognition of bank asset valuations. Temporary suspensions of the new standard would be most
effective in moderating the potentially disruptive effect of higher provisions, followed by transitional
arrangements and pragmatic guidance, in that order. By the same token, the ordering in terms of transparency is
This conclusion is based on the realistic assumption that the new standard provides a more faithful
picture of firms’ financial condition than the previous one.
The choice of options depends on views regarding the right balance between reducing
procyclicality and maintaining the credibility of the financial statements and accounting standards more generally.
To our mind, the guidance option, coupled with an adjustment of transitional arrangements – in line with the BCBS
guidance (BCBS (2020))11 – could provide an adequate balance between effectiveness and transparency. True, this
approach would arguably amplify the distance between accounting and prudential measures of capital. But it
could alleviate market pressure on banks, provided that market participants do not exclusively base their banks’
assessment on fully loaded regulatory capital (ie net of the additions allowed by the transitional arrangements).
Moreover, to the extent that transitional arrangements do not extend for too long, the adjustments should not
have a major impact on the regulatory framework’s ability to preserve financial system soundness. And they are
superior to temporary suspensions of a standard that has already been implemented, which runs the risk of being more
damaging to its credibility.
8 The transitional arrangements foresee the option of partially sterilising the impact of higher provisions stemming from the new standard
along a declining path over five years. Since IFRS 9 entered into force in 2018, the arrangements are already halfway in IFRS
jurisdictions. In the United States, however, the CECL standard has only been implemented in 2020. As a result, the sterilising impact
would in principle be greater than in IFRS jurisdictions.
9 In particular, neither public guarantees nor payment holidays need be considered as signalling a significant increase in credit risk,
triggering a reclassification from Stage 1 to Stage 2 in the credit risk hierarchy. This is important: provisions for loans in Stage 1 are
based on expected losses over one year while those for loans in Stage 2 cover the whole lifetime of the loan.
10 Temporary national suspensions, unless uniform, would also impair the comparability of financial statements across countries.
11 The new guidance opens the possibility for a two-year period, comprising the years 2020 and 2021, to allow banks to add back up to 100% of
the transitional adjustment amount to CET1.
4. Concluding remarks
Policymakers have often described the historic challenges the pandemic has raised as a war that must be fought and
won. Hence the need for bold and extraordinary approaches, measures and instruments. Regulatory and
supervisory responses are no exception. But the war cannot be won if, in the process, policy actions
compromise the vital intermediary role that financial institutions perform. It is more important than ever today
not to put at risk the stability and the transparency of the financial system. This sets a limit on how bold and
extraordinary the necessary steps can be. As the famous Chinese strategist Sun Tzu put it more than 25 centuries
ago in his Art of War, in order to win a war:
“…there are roads which must not be followed, armies which must not be attacked, towns which must not be
besieged, positions which must not be contested…”.
References
Basel Committee on Banking Supervision (2020): “Measures to reflect the impact of Covid-19”, April.
Borio, C (2019): “New loan provisioning standards and procyclicality“, Financial Stability Review, Bank of
Spain, Spring 2019. Also available in BIS Speeches.
Borio, C, C Furfine and P Lowe (2001): “Procyclicality of the financial system and financial stability: Issues and
policy options” in “Marrying the macro- and micro-prudential dimensions of financial stability”, BIS Papers, no
1, March, pp 1–57.
Borio, C and K Tsatsaronis (2004): “Accounting and prudential regulation: from uncomfortable bedfellows to
perfect partners?“, Journal of Financial Stability, vol 1, issue 1, September, pp 111–35.
——— (2006): “Risk in financial reporting: status, challenges and suggested directions”, BIS Working Papers,
no 213, August.
Carstens, A (2020): “Bold steps to pump coronavirus rescue funds down the last mile“, Financial Times, 30
March.
European Systemic Risk Board (2019): “The cyclical behaviour of the ECL model in IFRS 9”, March.
Financial Stability Forum (2009): “Report of the Financial Stability Forum on addressing procyclicality in the
Financial System”, April.
Office of the Superintendent of Financial Institutions (2020): “OSFI actions to address operational issues
stemming from COVID-19”, 27 March.
Restoy, F (2010): “Accounting, convergence, governance and regulation“, Speech at the IASC Foundation IFRS
Conference, June.
Restoy, F and J Roldán (2008): “Dynamic provisioning and accounting: can a consensus be reached?”, The Banker,
July.
Restoy, F and R Zamil (2017): “Prudential policy considerations under expected loss provisioning: lessons from
Asia”, FSI Insights on policy implementation, no 5, October.
