Proportionality in financial regulation: where do we go from here?

Speech by Mr Fernando Restoy, Chairman, Financial Stability Institute, Bank for International Settlements, at the BIS/IMF policy implementation meeting on proportionality in financial regulation and supervision, Basel, Switzerland, 8 May 2019.

Introduction¹

Good morning! And welcome to our joint BIS/IMF global policy implementation meeting on proportionality. That so many senior officials from various parts of the world are gathered here testifies to the topic's universal importance. I would also take this opportunity to thank our IMF colleagues for partnering with us on this event. 

Now that the post-crisis reforms are almost completed, prudential authorities are focused on implementing the agreed global standards. As part of this broader shift towards policy implementation, the question of proportionality has taken centre stage. And, given the increased complexity of Basel III and other regulatory standards, it is only natural to discuss how these standards can be most appropriately applied, particularly in light of the diversity of firms and financial systems operating around the world. 

An equally important debate - and one that often gets drowned out in the discussions around proportionality and regulation - is how day-to-day supervision should be tailored to reflect the systemic importance, complexity and risk profile of regulated entities. 

This meeting provides a timely opportunity to take stock of, and exchange views on the proportionality practices in various jurisdictions, and complement country-specific experiences with the broader analytical work done by various international organisations and academics. At the FSI, we've published several papers on proportionality during the past two years and I'll refer to some of our key findings during these remarks. 

This morning, I want to touch upon three key issues. First, I will outline the concept, objectives and considerations involved with adopting the proportionality principle. Second, I will take stock of how proportionality has been applied at the global level, drawing from our analytical work at the FSI and at other international bodies. Third, I will raise a few implications for policy in the context of proportionality and conclude with brief remarks.

The concept

The concept of proportionality stems from the need to limit public intervention - in the form of rules, sanctions and oversight - to what is actually needed to achieve the desired policy objectives.² 

Financial sector policy objectives typically include financial stability, market integrity and consumer protection. Within this domain, proportionality aims at avoiding policies that could distort the financial services market, for example, by unduly constraining its development; curbing competition or limiting the diversity of market participants. 

Focusing on the financial stability mandate, prudential authorities generally impose, on a proportionate basis, a range of regulatory, supervisory and resolution policies to applicable entities. The proportionality objectives of these three policies, however, are not necessarily the same. 

In regulation, a proportionate approach means tailoring regulatory requirements to a firm's size, systemic importance, complexity and risk profile. Here, the aim is to avoid excessive compliance costs or regulatory burden for smaller and non-complex banks that could unduly dampen their competitive positions without a clear prudential justification (Lautenschläger (2017)). 

In supervision, proportionality has a somewhat different focus: the main aim is to facilitate the efficient allocation of scarce supervisory resources and supervisory activities on firms that are either systemically important or are considered high risk. It is, therefore, closely associated with the concept of risk-based supervision. Naturally, a by-product of this approach is to lighten the supervisory burden on smaller, less complex and healthy banks. This dovetails nicely with the proportionality objectives of regulation. 

In regard to resolution policies, the aim of proportionality is to adjust the requirements for recovery and resolution planning and resolvability to the likelihood that regulated firms will cause systemic stress if they fail; and the expectation that firms can be subject to applicable insolvency procedures without systemic impact. 

Against this backdrop, to apply proportionality necessarily means that different institutions are subject to a differentiated set of requirements. The risk therefore exists that proportionality could be misused to offer a sort of regulatory subsidy to specific institutions, thus creating rather than removing competitive distortions, and preventing (or delaying) an otherwise orderly restructuring of the industry. 

In addition, proportionality - at least in the context of regulation - generally involves the imposition of simpler, less-risk sensitive requirements to applicable entities. One unintended consequence is that it can encourage risky behaviour by firms that benefit from the simplified rules. 

Collectively, these risks suggest that sound proportionality regimes should, ideally, meet at least three conditions: first, the adoption of simplified requirements should not undermine key prudential safeguards, particularly the requisite capital and liquidity backstops needed to promote confidence in regulated institutions and the financial system; second, supervisors should maintain sufficient awareness and control of the overall risk profile of entities that benefit from simplified regulatory rules; and third, the proportionality regime should not seek to overprotect small or medium-sized firms from competition, particularly if there is overcapacity and where consolidation can help to promote a more efficient and viable banking industry.

The evidence

Let me now turn to how the proportionality principle has been implemented in practice. Proportionality is widely applied in tailoring regulatory, supervisory and resolution policies for banks and other financial institutions.³ 

Prudential regulation

The proportionality strategies used to tailor regulatory requirements vary markedly across jurisdictions, including the criteria used to differentiate institutions; the scope of application (eg which requirements are affected); and the methods used to apply proportionality (eg exemptions from rules, modifications to applicable Basel rules, or replacement of Basel rules by domestic standards).

These differences may reflect the lack of international guidance on how to apply proportionality. In banking, beyond the expectation that all Basel standards are applicable for internationally active banks, there is no commitment for prudential authorities to extend their application to other banks operating in their respective jurisdictions. In addition, as an 'internationally active bank' has purposely never been defined by the BCBS, national authorities are free to interpret this term in their own way. 

As for insurance, differences in proportionate solvency regulation across jurisdictions are even more pronounced than in the banking sector, as there is currently no global capital standard for internationally active insurance groups. Moreover, the Insurance Core Principles are not prescriptive in the application of proportionality to solvency regulation, leaving much flexibility to individual jurisdictions to implement their own approaches (Yong and Löfvendahl (2018)).

The evidence on proportionality in banking regulation suggests that this concept is most often applied to the market risk framework,⁴ the quantitative liquidity standards and the large exposures regime⁵ as well as to disclosure and reporting obligations (Castro Carvalho et al (2017), BCBS (2019)). In general, most adjustments in regulation aim at reducing complexity without necessarily diminishing stringency (Restoy (2018)).⁶

In non-BCBS member jurisdictions - which are under no commitment to adopt Basel standards - and where the vast majority of locally incorporated banks are unlikely to be internationally active, the differences in how proportionality is applied are more widespread. 

In a recent FSI paper (Hohl et al (2018)), we review the proportionality practices of 100 non-BCBS jurisdictions regarding their application of Pillar 1 requirements of the Basel framework. One of the most interesting findings is that a wide variety of prudential regimes are implemented in non-BCBS jurisdictions. These include at least three Basel versions of the risk-based capital (RBC) regime (Basel I, II, III); two Basel versions of the large exposure rules (1991 and 2014); and the extensive use of domestic liquidity and large exposure rules in lieu of applicable Basel standards. 

Another insight is that jurisdictions apply different tailoring methods for different iterations of the Basel standards that they have chosen to adopt. For example, despite the simplicity of Basel I, jurisdictions under that standard still tend to make modifications⁷ to reflect country specificities, but these adjustments are generally applied to all banks in the system - with only limited differentiation in rules between smaller versus more systemically important firms.

In contrast, as countries shift to the Basel III RBC regime, greater differentiation and more multifaceted proportionality strategies are applied. This reflects Basel III's additional features and complexity.

Notwithstanding the range of proportionality methods applied in non-BCBS jurisdictions, some have argued against an excessive reliance on proportionality. In a recent paper by the Center for Global Development,⁸ the authors note that an undue reliance on proportionality can undermine the perceived benefits of a common global set of regulatory standards, raising level-playing-field concerns and complicating the task of making cross-country comparisons.

Supervision

With respect to proportionality and supervision, a forthcoming FSI paper (Duckwitz et al (2019)) - based on a survey of 16 BCBS and non-BCBS jurisdictions - indicates that all surveyed authorities apply proportionality, which demands various degrees of supervisory judgment.

To facilitate a proportionate approach in supervision, some authorities rely more on principles-based approaches that emphasise a holistic assessment of a firm. By contrast, others have developed more structured methodologies, which we refer to as "guided discretion". We find that authorities rely more on guided discretion approaches when setting the supervisory intensity of a firm and in determining the amount of capital add-ons under Pillar 2; meanwhile, principles-based approaches, which require a more intricate degree of judgment from supervisors, are used in assessing the quality of a firm's corporate governance.

The key takeaway is that the use of proportionality in supervision is not a choice; it is an intrinsic part of supervision that allows supervisory resources to be better allocated to firms that pose the greatest risks. Further, it also helps supervisors to better align a bank's risk profile with its financial buffers and the quality of its risk management/governance arrangements. These issues simply should not, and cannot, be dealt with by regulation alone.

Resolution

Proportionality is also used in tailoring resolution planning requirements - in terms of both scope and intensity - to achieve policy objectives. A recent peer review report by the Financial Stability Board (FSB (2019)) highlights the variety of proportionality methods FSB jurisdictions have applied in introducing resolution planning requirements.

In terms of scope of application, the approaches vary from jurisdictions that require resolution plans for all banks to others that impose requirements only on their G-SIBs and/or D-SIBs, and a few jurisdictions that follow other approaches, for example, by imposing requirements on banks that meet an asset size threshold or tailoring rules on banks on a discretionary basis.

However, where jurisdictions, such as the European Union (EU), impose resolution plans for all banks, proportionality is introduced by providing for a tailored application of some aspects of resolution planning requirements (eg the frequency of resolution plan review, data reporting and plan content). Indeed, in the Banking Union and the EU member States, resolution planning for small less-systemic banks that can be liquidated through normal insolvency proceedings focuses mainly on arrangements to assist the liquidation, such as a single customer view to enable swift depositor payouts.

Policy implications

Let me now turn to some of the policy challenges.

On the one hand, the variety of proportionality approaches taken may well reflect different characteristics in national banking markets. On the other, it is unlikely that all proportionality approaches taken in various jurisdictions could deliver the same outcomes with respect to the objective of ensuring a level playing field while protecting financial stability.

By way of example, if we compare how aspects of Basel III have been applied in the United States and the European Union, we find vastly different approaches taken with respect to proportionality. 

In the United States, only a few banks with total assets of $250 billion or more or $10 billion or more in total on-balance sheet foreign exposure are subject to Basel III's risk-based capital and leverage requirements, with additional capital requirements applicable to US G-SIBs. 

In regard to the LCR, the full LCR requirement generally applies to US banking organisations that are under the advanced approaches to regulatory capital measurement and their subsidiary organisations that have consolidated assets of $10 billion or more. 

In contrast, in the European Union, nearly all banks are subject to Basel III, with a few exceptions for smaller banks. 

These vastly different approaches, while well within each jurisdiction's purview, illustrates the need to achieve a common understanding of the pros and cons of the varied proportionality approaches that have been taken or are being considered. 

Against this background, consideration could be given to adopting a categorisation (or tiering) approach, where banks are grouped into several classes (defined by various criteria); and these categories are used as the basis for differentiating requirements. Indeed, this is the approach followed in countries such as Brazil and Switzerland (Castro Carvalho et al (2017), and it is the one that the United States is planning to adopt soon (US Federal Reserve (2018)). Ideally, the defined categories could be used not only to establish specific prudential rules but also supervisory criteria and resolution planning requirements. 

In this context, banks which are considered systemically important to the domestic economy - regardless of the complexity of their business models - should be subject to the most stringent regulatory, supervisory and resolution policies, in relation to other entities operating in those jurisdictions. I mention what appears to be an obvious takeaway, because our recent FSI study suggests that jurisdictions that remain under earlier iterations of the Basel RBC regime tend not to make such distinctions. 

More generally, to the extent that less complex rules imply less risk sensitivity, authorities may consider introducing certain regulatory requirements and supervisory policies to mitigate the potential incentives for firms to take on excessive risks. In particular, there could be merit in imposing more stringent regulatory requirements for banks that are subject to simpler obligations, as an explicit trade-off for adopting less risk-sensitive methodologies. 

One such approach, as currently proposed in the United States, will allow banks with consolidated assets of less than $10 billion to opt into to a community bank leverage ratio framework, provided that they meet other qualifying criteria⁹ and maintain a minimum leverage ratio requirement of greater than 9%. Banking organisations that elect to maintain the leverage ratio requirement of greater than 9% would no longer be subject to other risk-based capital or leverage requirements. It is important to note that the imposition of a 9% leverage requirement would - more likely than not - require banks, on average, to hold more capital than the minimum RBC requirements under Basel III. 

Having said this, we should also acknowledge some market realities. The ability to apply proportionality in financial regulation and supervision - without pushback from rating agencies, institutional investors and other market participants - seems to favour those economies that have "exorbitant privilege" in the structure of the global financial system. 

For economies that do not benefit from this privilege, additional work at the international level to identify good proportionality practices could serve as a reference for the supervisory community. These references may be particularly valuable for emerging market economies as they will need to ensure that their regulatory framework - particularly if it deviates in some respects from Basel standards for small and non-complex institutions - is still perceived internationally as being sufficiently rigorous.

Final remarks

In closing, let me quote the great Irish playwright George Bernard Shaw, who said: "The reasonable man adapts himself to the world: the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." For policymakers (notwithstanding the political challenges), it may well be easier to simply take the global Basel standards and apply them to all banks in their jurisdiction, regardless of the fit. 

It is much more difficult to analyse the design of Basel standards, deconstruct them and decide which aspects make sense for which group(s) of regulated entities. And all this has to be done in a way that reinforces key prudential objectives, and that upholds a competitive, level playing field in both domestic and international markets. 

In the wake of the post-crisis reforms, applying proportionality begins with a more humble endeavour: knowing your regulated firms in a manner that allows for a sensible tiering of regulation, supervision and resolution. 

I trust this meeting will shed light on these and other related policy issues over the next two days. Thank you.

References

Angeloni, I (2018): "Another look at proportionality in banking supervision", speech at the 13th Asia-Pacific High-level meeting on Banking Supervision, Singapore, 28 February. 

Basel Committee on Banking Supervision (2019): "Proportionality in banking regulation and supervision - a survey of current practices", March. 

Castro Carvalho, A, S Hohl, R Raskopf and S Ruhnau (2017): "Proportionality in banking regulation: a cross-country comparison", FSI Insights, no 1, August. 

Center for Global Development (2019): "Making Basel III work for Emerging Market and Development Economies", A CGD Task Force Report, co-chaired by Thorsten Beck and Liliana Rojas-Suarez, 23 April. 

Coelho, R, J Mazzillo, J Svoronos and T Yu (2019): "Regulation and supervision of financial cooperatives", FSI Insights, no 15, January. 

Duckwitz V, S Hohl, K Weissenberg and R Zamil (2019): "Pillar 2, risk-based supervision and proportionality", FSI Insights, no 16, forthcoming. 

Financial Stability Board (2019): "Thematic review on bank resolution planning", 29 April. 

Hohl, S, M Sison, T Stastny and R Zamil (2018): "The Basel framework in 100 jurisdictions: implementation status and proportionality practices", FSI Insights, no 11, November. 

Lautenschläger, S (2017): "Is small beautiful? Supervision, regulation and the size of banks", speech at an IMF seminar, Washington DC, 14 October. 

Restoy, F, (2018): "Proportionality in banking regulation", speech at Westminster Business Forum Keynote Seminar, London, 4 July. 

US Federal Register (2018): "Proposed changes to applicability thresholds for regulatory capital and liquidity requirements", vol 83, no 245, 21 December. 

--- (2019): "Regulatory Capital Rule: Capital simplification for qualifying community banking organisations", vol 84, no 27, 8 February. 

Yong, J and G Löfvendahl (2018): "Proportionality in the application of insurance solvency requirements", FSI Insights, no 14, December.

¹ I am grateful to Raihan Zamil and Juan Carlos Crisanto for their helpful comments and to Cissy Mak for her useful support. The views expressed are my own and do not necessarily reflect those of the BIS.
² See Angeloni (2018) for the historical context of proportionality.
³ See Coelho et al (2019) for the regulatory and supervisory approaches used to oversee financial cooperatives and Yong and Löfvendahl (2018) for an overview of proportionality regimes in insurance regulation.
⁴ The proportionality approaches taken typically involve either exempting all banks from the market risk capital requirements (Basel I countries) or to impose market risk capital requirements to only those banks that meet a minimum materiality threshold.
⁵ The most frequent proportionality strategy used for the large exposures standard and the Liquidity Coverage Ratio (LCR) - at least in non-BCBS jurisdictions - are the adoption of domestic rules in lieu of applicable Basel standards.
⁶ For example, several jurisdictions have imposed domestic liquidity rules in lieu of adopting the LCR. These rules frequently involve the calculation of a simple liquid assets/short-term liabilities ratio (or some variation thereof). However, in several cases, the implied liquid asset holdings under the domestic liquidity rules may be similar to, if not greater than, the applicable LCR requirement.
⁷ Most of the modifications relate either to making certain changes to the denominator or increasing the minimum risk-based capital ratio.
⁸ To address these concerns, the authors recommend a regional approach, particularly in emerging market and developing economies (EMDEs), where regulators agree on a common set of proportionality rules in their respective regions. See Center for Global Development (2019) for further discussion of the challenges faced by EMDEs in applying Basel III.
⁹ Under the proposal, in addition to the $10 billion size threshold, eligible institutions must also have limited amounts of off-balance sheet exposures, trading assets and liabilities, mortgage servicing assets and deferred tax assets arising from temporary differences that a banking organisation could not realise through net operating loss carrybacks. Banks using the advanced approaches, even if they meet the size and above requirements, are not eligible to apply this methodology.