The history of financial regulation is clearly marked by the history of successive crises. Thus, every financial or banking crisis is followed by a re-regulation process that seeks to curb future crises—that is, it attempts to draw lessons from what happened. For example, the 1929 crisis led to the creation of two institutions that remain highly relevant today: the deposit insurance system (the FDIC) and the Securities and Exchange Commission (the SEC), responsible for ensuring investor protection and market integrity.
Inevitably, however, regulation following a financial crisis may contain elements of overreaction, resulting from the logical political climate associated with the social costs of the crisis. For instance, the 1929 crisis led to the prohibition of interest on demand deposits in the United States (Regulation Q), a ban that lasted from 1933 until 2011 and led to the hyper-development of Money Market Funds and the problems subsequently seen in the 2007 crisis. Ultimately, we cannot rule out that the re-regulatory processes following a crisis may contain elements of little utility, or may even prove counterproductive by generating new risks that only become apparent decades later.
The Great Recession of 2007 was to be no exception. Thus, the new regulatory paradigm has brought us new rules and new institutions with new powers. It is not my intention to conduct a detailed review of the new regime for banks, but let us consider some of its elements.
Firstly, requirements for both total capital and high-quality capital (the so-called CET1) are increased. For some portfolios, this increase could represent ten times the capital required before the crisis. Furthermore, a leverage limit is established to avoid situations where risk-weighted capital (RWAs) is high but at the cost of excessively high leverage (in other words, banks now wear both a belt and suspenders). There is also the so-called Pillar 2, which allows a supervisor to impose capital surcharges above the minimums based on the specific risk profile of each entity.
In addition, supervisors now regularly use stress tests to evaluate the capital adequacy of banks; another distinct way of verifying their solvency.
Specific quantitative liquidity requirements have also been established for the short and long term. The short-term requirement, the LCR, establishes the need to have liquidity buffers that allow the entity to survive for one month under an extreme liquidity crisis situation. And the long-term requirement, the NSFR, sets a limit on maturity transformation; that is, although a bank always finances its assets with shorter-term liabilities, there must be a limit to the maturity mismatch (thirty-year mortgages cannot be financed with three-month revolving loans). Furthermore, liquidity stress tests will also be conducted (a liquidity Pillar 2).
Macroprudential considerations also inspire the new regulatory environment, understood as measures that seek to reinforce not only the strength of each bank but that of the system as a whole. There are capital buffers—countercyclical, conservation, and systemic—but also liquidity buffers. And the definitions of risk weightings (or LGDs) are set taking macroprudential considerations into account.
Finally, there is perhaps the most novel element of this new supervisory package: the establishment of rules for the orderly liquidation of banks, the so-called banking resolution. Not only are new principles established, which recognize that “Chapter 11” procedures cannot be applied directly to banks, but specific administrative authorities have been created—the resolution authorities—tasked with this duty. We are talking about “bail-inable” liabilities that can help recapitalize a bank, or a part of a bank that cannot be liquidated in the event of problems, but also about resolution plans (living wills) made by the banks themselves, which are nothing more than the planning of their orderly death in the event of a crisis. And, even more importantly, we are talking about the very broad powers granted to supervisory and resolution authorities to adopt early intervention measures to prevent a struggling entity from reaching such a state of unviability.
In addition to this series of measures, there is another set of additional local regulations that must also be taken into account. For example, we have the prohibition of proprietary trading, the so-called Volcker Rule, in the United States, but also the Vickers reform in the United Kingdom, which requires universal banks to separate (ring-fence) their commercial banking activities from their investment banking activities.
But let us move on to the main purpose of my conference. What are the characteristics of this new regulatory paradigm?
In principle, the architecture of a regulatory system must choose between two options. One can opt for a system based on strict and very detailed rules, in which supervisory authorities focus on verifying compliance (tick the box) with very extensive standards. For the supervised entity, this body of detailed rules increases regulatory compliance costs, but in exchange, they offer certain advantages:
The institutional framework of supervisory authorities before the crisis followed a sectoral model in some cases, with separate supervisors for banking, securities, and insurance; in others, a functional model (known as twin peaks) in which two different authorities coexisted—one for prudential supervision (of banks, insurance companies, and investment firms) and another to handle everything related to conduct of business rules and market integrity; or there was the case of a single supervisory authority grouping all these areas. Standards for each sector and area emanated from three international bodies: IOSCO, the BCBS or Basel Committee, and the IAIS (the International Association of Insurance Supervisors).
The crisis has disrupted this framework, as the number of bank supervisory authorities has multiplied. Thus, we have the emergence of banking resolution authorities (only in the case of the United States is a pre-existing institution, the FDIC, used), but also macroprudential authorities (usually central banks, but exercising a different discretionary power over banks). To give a clearer example, in the European context we now have three European supervisory authorities with regulatory powers (the so-called ESAs) for banking, securities, and insurance; we also have a supervisory body for Eurozone banks (the Single Supervisory Mechanism), another for the resolution of Eurozone banks (the Single Resolution Board), and one more for the European Union, responsible for macroprudential issues (the European Systemic Risk Board, based at the ECB).
This multiplicity of authorities tasked with monitoring banks may have its advantages: after all, redundant systems guarantee, in all industries, that if one system fails, the others will not.
But the problem arises when these redundant systems operate with different parameters, as the probability that they will end up reaching different conclusions is quite high. No one would like to fly in an airplane with that redundancy architecture.
At the international level, we also have new and powerful actors in the field of financial activity regulation. The FSB (Financial Stability Board) has not only emerged strongly as the inspiration for regulatory reform after the crisis (although it is the sectoral committees that do the concrete work) but, more importantly, it is the transmission belt between the G20, governments, and sectoral regulatory bodies.
In short, new and powerful actors have burst into the new post-crisis regulatory environment, increasing the risk that banks will be subjected to requirements that are, if not contradictory, at least occasionally inconsistent.
One of the main problems with the new regulatory paradigm is its enormous complexity. This stems from the complexity of the industry itself, and as much as we would like to have simple, straightforward rules that apply to all situations, a certain degree of regulatory sophistication is inevitable.
That being said, it is no less true that the new regulatory paradigm is probably excessively complicated. Let us consider a bank’s capital planning in the medium term. It will have to consider:
It is not obvious what a bank can do to plan its capital five years ahead given the redundancy of rules that affect it in one way or another. It is especially difficult to plan what the result of the stress tests will be: indeed, as the supervisor’s emphasis may differ each year depending on how macroeconomic risks evolve, it is impossible to know what scenarios will be used, for example, to evaluate capital adequacy three years ahead. We are, de facto, facing time varying capital requirements.
Another type of complexity, this time organizational, is that which arises from so-called structural reforms, whereby either certain activities considered speculative are prohibited (such as the Volcker Rule, which prohibits proprietary trading), or traditional deposit banking is required to be separated (ring-fenced) from notably riskier investment banking activities (the Vickers reform). This type of reform presents special problems for banks. Thus, they represent a new class of regulation that directly interferes with the way businesses are managed (instead of providing incentives through regulation to manage risks efficiently), or that directly prohibits certain types of operations. In many cases, they also impose extraterritorial effects on those banks operating in multiple jurisdictions (which increases not only compliance costs but, above all, legal risks). These are regulations that are concerning not only for their current consequences but because they open the door to future regulatory developments that invade the field of free corporate organization, which is part not only of management’s prerogatives but also of the free and legitimate choice of business model inherent in a market economy.
Those with a weakness for Tolkien will remember the successive volumes of his new works that came to light every few years. Just when it seemed the vein had been exhausted, a new volume would appear. That same feeling arises with the successive phases of the reform.
Thus, it seemed that Basel III, with its 7% CET1, represented the new regulatory capital standard. But no, then came the systemic buffers, the countercyclical capital buffers, TLAC and MREL, SCAR, etc.
It is because of this incrementalist derivative that markets do not quite believe regulators when they say there are no more rules in the oven, that the cake is already baked. For example, the main lines of Basel III have been defined for years, and regulators from the Basel Committee announce that the process is finished, except for a few loose ends: interest rate risk in the banking book (IRRBB), the fundamental review of the trading book, the review of the standardized approach, the imposition of floors on model parameters, and the review of the operational risk framework (which will eliminate advanced models), among others. It is no wonder that, given these loose ends, the market speaks of Basel IV.
And the problem is not only the impact of these reforms on regulatory capital, but above all that they induce market uncertainty, which scares off potential investors in bank capital. Similar to the well-known “peso problem” in exchange rates (the impact of a sudden, unforeseen, and idiosyncratic event on the exchange rate), the potential emergence, for unforeseen reasons, of another “loose end” in Basel III or in banking resolution that requires, once again, another turn of the screw in standards weighs on the stock market performance of banks. This is why regulators, who now say that the re-regulation effort is over (and everything indicates that this is indeed the case), have such difficulty convincing the market.
Another element of the new paradigm is the multiplicity of capital and liquidity buffers. Within the former, we have the capital conservation buffer, the countercyclical buffer, the systemic buffer, and probably some that escape mention. The new regulation embraces the concept of buffers—that is, having an additional safety margin in case problems arise. This is the new macroprudential view of regulation: it is not enough to guarantee the individual stability of banks, but also that of the system as a whole, which requires understanding the interrelationships in the financial system and incorporating potential externalities into the design of the rules.
But buffers only make sense both economically and prudentially if they are used in case of need. In other words, a car with airbags is of no use if they do not deploy in the event of an accident. And there are doubts that, when the time comes, authorities or the market will allow banks to use these when the situations for which they were precisely created arise.
Indeed, let us consider the short-term liquidity buffer. From a regulatory perspective, it is logical for the buffer to be used when liquidity tensions in the market increase. But from the market’s perspective, the use of the buffer may be seen as a sign of weakness in the bank, so in practice, the bank will most likely not use that buffer, even when it is rational to do so from a supervisory and management standpoint.
Another element of the new regulation is a growing skepticism regarding the implementation of internal models by entities to calculate risk-weighted assets (for example, IRBs for the credit portfolio). The term “internal models” is a bit misleading, as it would be more accurate to say models designed by supervisors but using private inputs from each bank. In any case, it is the banks that, based on those models, compute their risk (their RWAs or risk-weighted assets) and establish their capital according to supervisory scales.
Among those loose ends yet to be defined in the reform mentioned earlier, which the industry calls Basel IV, is the possibility of imposing either a ban on the use of internal models (for example, for operational risk) or limitations on parameters (PDs or LGDs). In short, it is about making internal models less internal and more subject to the supervisor’s criteria, thereby reducing one of the main problems revealed after their implementation: the high dispersion existing in the calculation of RWAs, not only between banks but also between the credit portfolios of different banks.
But this limitation of internal models also has unintended consequences and presents some contradictions. Regarding unintended consequences, risk sensitivity of capital requirements is lost, which may increase incentives for regulatory arbitrage—that is, assuming more risk without having to put up more capital. As for contradictions, it turns out that while international accounting standards (IFRS) move, after years of debate in the FSB, toward covering expected losses with provisions (which requires the use of internal models), solvency rules are moving away from these internal estimates. Continuing with the contradictions, in Europe, the regulatory package called Solvency II accepts the use of full portfolio models for insurance companies to calculate their internal capital, while in banking, there is a retreat from much more restrictive internal models.
Not infrequently, learned interlocutors mention that the objective of the reform is to turn banks into utility companies (like electric companies)—that is, into less profitable but also less risky companies. The banking business would be, in short, boring but safe. I believe there is no idea more false, pernicious, and dangerous than this.
Because if we want banking to fulfill its function of financing the productive economy, financing investment projects for companies, home acquisitions for families, or their consumption, the banking business will by definition be cyclical. That is, ROE will rise in good times of the economic cycle and suffer in bad times. Or, in other words, only a shrunken banking sector, with non-existent balance sheets, dedicated solely to transactional operations and not to credit risk (in line with academic proposals for narrow banking) could behave similarly to a utility.
The new regulatory paradigm has focused on increasing the stability of the financial system in general, and of banks in particular. But this revolution in rules has completely ignored a very profound change facing the sector: the emergence of new technologies in finance and their impact on it.
There is, therefore, a lack of a regulatory effort that allows for the implementation of fintechs without jeopardizing the stability and integrity of the system. It is not a matter of protecting the banking sector from competition from new digital operators, but of allowing it to compete on equal terms with these new agents. It is no longer just a question of financial stability—same activities and risks, same regulation—but of consumer protection. That is, the development of fintechs cannot be based on the degradation of conduct rules and retail investor protection.
One of the causes of the crisis lay in the processes of regulatory or capital arbitrage that led entities outside the banking sector to assume credit risk on their balance sheets. Indeed, both AIG and Lehman Brothers were not traditional deposit-taking banks, but an insurance company and what in Spain is called a securities firm (the term investment bank is confusing). And their respective crises jeopardized the stability of the global financial system. The danger of the shadow financial system is not theoretical, but very practical and anchored in the experience of the recent crisis.
With capital requirements that, according to the Bank of England, are ten times higher than those in force before the crisis (for, for example, certain positions in the trading book), the question for the future is not whether regulatory arbitrage will exist and, therefore, whether credit risks will shift from banking to other sectors, but where, how, and when that risk shift will take place. After all, if requirements increase tenfold, the incentives for regulatory arbitrage also increase tenfold.
Two clarifications should be made here. Firstly, when speaking of shadow banks, we should not think of a non-bank entity with all the payment and transactional services that a bank provides. We refer specifically to financial entities that invest in credit risk, doing so with third-party (and redeemable) resources and, furthermore, financing themselves at shorter terms than the risk of the asset. And secondly, that supervisory and macroprudential authorities, such as the FSB or the ECB, are especially aware of this regulatory arbitrage problem and are improving both the regulations to be applied to these operators and the surveillance to detect new forms and activities of capital arbitrage in shadow banking.
It is unthinkable that a financial crisis like the one experienced, with its high social cost and global political consequences, would not lead to a profound review of the regulatory environment. It happened after the 1929 crisis, it has happened after the 2007 crisis (already called the Great Recession), and it will happen again after the next global financial crisis. It is not, therefore, a matter of questioning the rationality of the regulatory reaction.
But we must also not ignore the unintended consequences and side effects of the approved reforms. We must understand the weaknesses of the new regulatory paradigm well in order to adapt, as regulators and operators, to the new environment and understand where the new system of created incentives leads us. In the case of regulators, to guarantee financial stability, and in the case of banks, to be able to adapt the business model, guarantee its viability in the short term, and thus continue to fulfill their function of financial intermediation, channeling the savings of families and companies toward productive investment. Because a safe banking sector is of no use if it does not fulfill its function of contributing to the improvement of society’s well-being.
Thank you very much.
José María Roldán, Chairman of the Spanish Banking Association
