The post-crisis regulatory agenda: What is missing?

English translation of speech in Spanish by Mr Fernando Restoy, Chairman, Financial Stability Institute, Bank for International Settlements, to Círculo Financiero La Caixa, Barcelona, Spain, 19 February 2018.

BIS, FSI speech  | 
21 February 2018


I would like to start by expressing my gratitude to the organisers for their kind invitation to participate again in this prestigious forum - which, like so many other successful initiatives, is linked to Isidro Fainé - and to the President of Caixabank, Jordi Gual, for his kind introduction.

In May 2015, I gave a talk in this very auditorium. On that occasion, as suggested by Fainé I focused on analysing the intense reform process undertaken by the Spanish banking sector in the preceding years, on describing the achievements and on explaining the biggest challenges, particularly in the context of the (at that time recent) joining of the European banking union.

Three intense years have passed, and the topic of my speech today again revolves around the dichotomy of achievements and challenges in the financial sector. I have only exchanged the Spanish or European scope for a global one that is more in line with my current responsibilities. However, I hope that this evident lack of originality in the outline of my remarks will be made up for by the interest that an analysis of the challenges that currently remain for the international regulatory community may awaken.


Allow me to start with the achievements. As you know, last December the Group of Central Bank Governors and Heads of Supervision, which supports the work of the Basel Committee on Banking Supervision, announced that an agreement had been reached which represented the finalisation of the new global framework for prudential regulation, known as Basel III. The announcement rounded off almost 10 years of work in various international forums   to develop what has been called the post-crisis reform, aimed at reducing the likelihood of future systemic crises and minimising their cost.

The agreed revisions to the reforms not only refine the prudential standards for credit institutions, but also include new requirements to facilitate the resolution of non-viable institutions and minimise repercussions for taxpayers, and various measures to strengthen market infrastructures.

There is general consensus that the reforms have already achieved many of their goals. In particular, since the beginning of the crisis larger entities have doubled their high-quality capital ratios, have greatly increased their liquidity buffers and are progressively modifying the structure of their balance sheets and their internal organisation in order to facilitate their resolution in case of non-viability.

At the same time, market transactions are now more secure, as the number and volume of transactions executed and settled outside organised markets and central counterparties (CCPs), respectively, have fallen, at the same time that these infrastructures have been strengthened. In addition, the most opaque forms of financial intermediation not subject to prudential regulation have lost most of their relevance as a result of the regulatory reforms.

In general, the financial system is now better able to support in a stable way the real sector of the economy. In fact, financial flows to firms and households, which were profoundly affected in some jurisdictions during the crisis, have mostly been restored, while financing costs have remained very low.


The fact that the reforms are wide-reaching has made it more acceptable for there to now be a regulatory pause to allow firms and supervisors to adjust to the new rules, properly assess their effects and increase the stability of the relevant framework for strategic decision-making.

The new regulatory framework's profound impact on the banking sector makes a cautious approach to considering new modifications advisable in the short term. Still, it would be a mistake to assume that the reforms dispel all significant risks to financial stability for the foreseeable future.

Indeed, the crisis, far from implying a widespread deleveraging, did not prevent the build-up of debt either by the public sector (in advanced economies) or the private one (in emerging market economies). Additionally, the continuing low interest rate environment encourages risk-taking, is conducive to asset inflation, exposes indebted parties to severe financial losses and dents commercial banks' profitability. This last factor, along with the excess capacity of the banking sector in some jurisdictions - like Europe or Japan - casts doubt on the viability of some institutions' business model, and could eventually affect the sector's stability.

Technological developments - which attract so much attention these days - are significantly changing the nature of the business and the structure of the banking sector and the financial system as a whole. These advances open up a wide array of opportunities for users and suppliers of certain types of digital services, but also give rise to potentially significant risks.

Against this backdrop, the regulatory pause banks asked for cannot hold back the preventive action of the authorities charged with maintaining financial stability. Arguably, in the short term the emphasis previously placed on developing new global regulatory standards may need to give way to concrete actions in each jurisdiction. These should consist of not only rigorous implementation of the new standards, but also the adoption of complementary policies aimed at maximising the positive effects of the new regulatory principles.

In this regard, we must keep in mind that the international reforms generally have three types of fundamental limitations. First, they do not - and cannot - cover in sufficient detail all the areas that are potentially relevant for maintaining financial stability in all jurisdictions. Second, their formal scope of application is generally limited to entities or activities of international relevance. And third, their practical implementation in jurisdictions often faces significant operational hurdles.

Allow me to spend the second part of my remarks giving some examples of how domestic policies should bypass these limitations of the international financial standards. In describing them I will refer to reports that have been or will be published by the Financial Stability Institute (FSI).

Non-performing loans

Let us start with an important area in which the reach of international prudential standards is relatively limited: the identification and valuation of non-performing loans on bank balance sheets.

As you know, the abundance of non-performing assets, whether they be non-performing loans or foreclosed properties, is hurting the profitability of the banking sector in Europe and, increasingly, in other jurisdictions as well. It may explain the slack in the credit supply in the most affected economies. Certainly, the emergence of this phenomenon is partly linked to policies governing the classification and valuation of impaired assets.

The identification and measurement of assets are generally governed by accounting standards. Until this year, these principles established that provisions for non-performing loans be calculated so as to reflect the losses incurred by the entity as at the reference date of its financial statements. Since the entry into force of the new asset valuation principles of the International Financial Reporting Standards (IFRS 9), institutions located in countries that follow this accounting code will have to calculate the provisions according to their own forward-looking estimates of the expected loss from the loans.

Incorrect classification of assets and insufficient provisions not only distort the accounting reflection of the economic reality of banks, but also generate biases in the calculation of capital and consequently hinder monitoring of solvency by the prudential authority. As a consequence, accounting standards often interact with prudential regulations or heterogeneous supervisory practices that regulate at the national level aspects such as the definition of non-performing loans, provisioning requirements and criteria for estimating the value of collateral and its use for classifying and valuating assets.[1] In fact, the application of some type of supplementary control by the prudential authority appears particularly warranted in a context in which, as a result of the change in the accounting standard, the calculation of provisions by banks will entail more complexity in the future.

In the area of the identification of problem assets, in 2017 the Basel Committee issued a guide which attempts to harmonise the definitions of non-performing and forborne assets.[2] These definitions will surely help indicators of the asset quality to be more comparable between jurisdictions in the future.

In the area of asset measurement, there are no international principles that promote the adoption of specific prudential tools. In some jurisdictions, the regulator may establish specific provisioning requirements for credit entities as a complement to the accounting standards. However, prudential authorities do not generally have the ability to intervene on loan-loss accounting provisions, which adhere to the applicable accounting principles and are monitored by auditors. However, in some jurisdictions - especially in Asia - where the regulator does not have accounting responsibilities, it can establish benchmarks for minimum provisions and require adjustments to regulatory capital when the provisions are below the benchmark.[3] In the jargon this is referred to as a prudential backstop.

In this way, direct intrusive action by the supervisor which, for example, imposes minimum accounting provisions that would affect an entity's balance sheet and the profit and loss statement, turns out to be incompatible with the legal framework in most jurisdictions. By contrast, the use of prudential backstops means that when the prudential supervisor believes that provisions are insufficient, it imposes adjustments to regulatory capital but not to the entity's public financial statements. Therefore it is an option that respects the division of accounting and prudential responsibilities.

Recently the European Central Bank and the European Commission[4] have proposed a system of prudential backstops for accounting provisions related to non-performing loans, which follows the outline I have just described. This initiative is part of the strategy to reduce incentives for banks to accumulate impaired assets on their balance sheets, which is currently one of the main factors darkening the outlook for the European financial system.

In more general terms, the global regulatory community must reflect on whether this type of prudential controls over accounting provisions, which already exist in various jurisdictions, should become one of the supervisory tools promoted by international standards.


Let us move on to examine the scope of application of the international regulatory principles, where there are also major differences across jurisdictions.

As you know, the Basel accord is in principle designed to ensure that internationally active banks comply with common prudential standards. Nevertheless, in practice the Basel principles have served as a reference for the regulation of the whole banking sector in many jurisdictions. In Europe, for example, prudential regulation - contained in the Capital Requirements Directive (CRD) and Capital Requirements Regulation (CRR) - is derived directly from the international standards and is applied to all credit institutions, no matter their degree of cross-border activity.

The new Basel accord has meant a significant increase in regulatory requirements, due to greater sophistication of the banking business and the need to improve control over different types of risk (such as credit, liquidity, market, operational and derivatives counterparty risk). This involves an important increase in the cost of complying with the regulations for institutions, which have to adjust their systems according to the regulatory requirements and deal with new and more exhaustive information requirements.[5]

Undoubtedly, the costs associated with compliance with the new regulations increase disproportionately for those entities that, due to their size, are less able to take advantage of the economies of scale that characterise the systems and processes of regulatory compliance. As I said before, the greater complexity of the new standards derives partly from the sophistication of the business models of some institutions, in many cases those that are large and internationally active. It thus could be argued that smaller institutions that specialise in more traditional intermediation activities in the domestic market face excessive costs due to the burden of complying with regulations that are not calibrated to the risks they generate.

In this way, in parallel with the development and implementation of Basel III, regulators have started to consider the need to adjust the regulatory requirements for small and unsophisticated entities. The debate revolves around the interpretation of the principle of proportionality in the regulation, and concrete proposals have emerged which incorporate various formulas for the application of that principle.[6]

For example, in the United States regulatory agencies have recently announced plans to ease requirements for small and medium-sized institutions.[7] Switzerland and Brazil have opted to develop a system for classifying institutions based on their size and business model, so that different requirements can be applied for different types of entities. In the European Union, the European Commission's planned update of the CRR and CRD includes for a group of entities lower requirements in terms of published information, communications to the supervisor and liquidity ratios, among other requirements.

Even when simplifying requirements may be justified in order to mitigate the distortions that regulation inevitably gives rise to in the banking services market, there are at least two important restrictions on the application of the principle of proportionality.

First, simplified requirements should not entail lowered solvency and liquidity requirements. In fact, institutions' size and risk profile are already taken into account in the indicators used to determine the Basel regulatory requirements. In addition, as the crisis has shown in some jurisdictions, systemic crises are not always caused by problems with the largest firms.

Second, proportionality should not be used to protect a certain type of entity from the normal competitive environment. If formulated correctly, proportionality can help level the playing field by keeping small institutions from being unduly penalised due to the complexity of the regulations. On the other hand, poorly calibrated application could deactivate the competitive forces that foster additional consolidation of the industry and help correct in an orderly way the excess capacity observed in some jurisdictions.

Accordingly, to facilitate compliance with the two restrictions I just mentioned, some jurisdictions are considering accompanying the introduction of simplified obligations for small entities - eg in regard to the contents and frequency of the confidential financial statements reported to the supervisor or to the solvency self-evaluation requirements - with more demanding requirements in terms of capital ratios. This formula may facilitate a balanced application of the principle of proportionality by simplifying the prudential controls on small and simple institutions while requiring them to have somewhat higher loss-absorbing capacity.

New resolution framework

Allow me to conclude my list of the areas in which action by national authorities is needed to complement the new international framework by discussing the implementation of the new resolution standards.

As you know, an essential component of the post-crisis reform has been the adoption of new resolution mechanisms for non-viable credit institutions that minimise the need for them to be bailed out by the public sector. This new framework is captured in various publications by the Financial Stability Board (FSB).[8]

A crucial ingredient of the new framework is administrative capability by national authorities to require, outside the regular insolvency proceedings, that creditors of institutions in resolution absorb losses and participate in the recapitalisation of those entities. To do so they must be able to write off debt securities or require them to be converted to common shares (ie bail-in).

Evidence available until now points to major difficulties for the practical implementation of this essential element of the new resolution framework. Currently, around half of the jurisdictions represented on the FSB do not include in their regulations these bail-in powers for resolution authorities.[9]

This is not the case in the European Union, where the new resolution regulations contained in the Bank Recovery and Resolution Directive (BRRD) give competent authorities the power to ensure that creditors - be they holders of uncollateralised debt or unprotected deposits - are part of the resolution process. However, even though the BRRD entered into force 2015, this power has not been fully utilised in any of the crisis episodes since then.

The difficulties with applying bail-in stem from economic, institutional and operational factors. From the economic point of view, bail-in risks may generate contagion through the losses suffered by institutional investors exposed to the bank under resolution and through a general increase in risk perception for the type of instruments affected, with the subsequent corrections in their market value.

From the institutional perspective, bail-in powers are peculiar with regard to the traditional principles of commercial law. They involve accepting the administrative capacity to alter the rights and obligations of creditors, regardless of the established contractual terms and without the judicial control that characterises normal insolvency proceedings.

From the operational perspective, bail-in requires developing the ability to valuate different kinds of debt instruments within a short period of time, agile methods to modify the formal rights and obligations of affected creditors and shareholders, and swift procedures that allow the trading, clearing and settlement of those instruments after the write-off or conversion imposed by the authorities.

Many of the challenges for the implementation of bail-in procedures are alleviated when entities in resolution have issued enough instruments whose specific characteristics, due to their subordinated nature or the existence of specific provisions for conversion into equity, include the possibility that they will eventually have to absorb losses. Thus, at both the international level - through the total loss-absorbing capacity (TLAC) standard for globally systemic institutions - and the European one - through the minimum requirement for own funds and eligible liabilities (MREL) - regulations have been established that aim for institutions to have enough instruments on the liabilities side of their balance sheet that could serve as bail-in debt with as little disruption as possible.

Naturally, these requirements are not binding for small institutions. To the extent that their failure would not have systemic repercussions, these institutions should be able to disappear from the market through regular insolvency proceedings, which would not require the new resolution powers to be exercised. And for their part, the TLAC/MREL requirements should also not present major difficulties for more complex entities with experience issuing debt instruments in the market, although they can raise their financing costs to a certain degree.

The main problem is for the middle class: medium-sized institutions whose failure could have systemic repercussions, so that they must be subject to the new resolution framework, but whose business model - based on financing their lending activity primarily through capital and deposits - is not consistent with large issuance in the market for TLAC- or MREL-eligible instruments (such as subordinated or convertible debt). These intermediate institutions could come under significant pressure in the future.

It is thus possible that, in the long term, stringent MREL requirements may foster a restructuring of the sector into two well defined segments. First would be systemically important institutions that are able to issue on the market the liabilities required by the resolution regulations. Second would be a group of smaller institutions that would not perform essential functions themselves and could be subject to the established insolvency proceedings without generating adverse systemic effects.

In this context, the role of the authorities could include two types of action. First, they could facilitate corporate operations that would encourage a smooth transition to the new bipolar structure of the industry. Second, given the expected increase in the proportion of non-viable institutions that will be subject to insolvency proceedings, they could target an improvement of those proceedings for credit institutions, with the aim of minimising the destabilising risk associated with their length and the excessive destruction of value to which they tend to lead.

Concluding remarks

At the beginning of these remarks, I told you that, despite the significant progress that has been made, the regulatory community still has a lot of work to do. The tasks ahead are derived not only from the various sources of risk that affect the international financial system, but from the need to rigorously translate international principles into the regulations and supervisory practices of each jurisdiction. As I have said, this means adopting specific policies when international guidelines are not prescriptive enough - as in the case of identifying and measuring non-performing loans. It also means accurately defining the scope of application of the international standards, aiming for careful application of the principle of proportionality. Finally, it entails designing complementary measures that maximise the benefits and reduce the risks associated with the practical application of the international principles - such as those required by the new powers to resolve credit institutions.

All these tasks should be carried out by national authorities (or European ones in the case of the EU). Still, experience shows that in this area of implementation cooperation between supervisors can add significant value rooted in the exchange of experiences, mutual learning and the identification of effective ways of overcoming challenges. This is precisely the area of focus of the FSI.

Thank you very much.

[1]  See P Baudino and R Zamil, "The identification and measurement of non-performing loans: a cross-country comparison", FSI Insights on policy implementation, forthcoming.

[2]  See Basel Committee on Banking Supervision, Prudential treatment of problem assets, April 2017.

[3]  See F Restoy and R Zamil, "Prudential policy considerations under expected-loss provisioning", FSI Insights on policy implementation, no 5, October 2017.

[4]  See European Commission, Consultation document: statutory prudential backstops addressing insufficient provisioning for newly originated loans that turn non-performing, November 2017; and ECB, Addendum to the ECB guidance to banks on non-performing loans, October 2017.

[5]  See A Dombret, "Addressing proportionality in Europe", speech at the 13th BCBS-FSI High-level Meeting for Africa, 24 January 2018.

[6]  See A Castro Carvalho, S Hohl, R Raskopf and S Ruhnau, "Proportionality in banking regulation: a cross-country comparison", FSI Insights on policy implementation, no 1, August 2017.

[7]  See US Department of the Treasury, Federal Reserve System and Federal Deposit Insurance Corporation, Simplification of capital rules pursuant to the Economic Growth and Regulatory Paperwork Reduction Act of 1996, September 2017.

[8]  See, in particular, Financial Stability Board, Key attributes of effective resolution regimes, October 2014.

[9]  See Financial Stability Board, Implementation and effects of the G20 financial regulatory reforms, July 2017.

Related information