Chapter 7. Will Revamped Financial Regulations Work?

Tamim Bayoumi
Published Date:
October 2017
  • ShareShare
Show Summary Details

The need for a thorough revamp of financial regulation was recognized from the outset of the North Atlantic crisis. The first ever G20 leaders’ summit, convened in November 2008, just weeks after the collapse of Lehman Brothers in Washington, pledged to “implement reforms that will strengthen financial markets and regulatory regimes so as to avoid future crises”. A few months later at the April 2009 London G20 summit the leaders issued a Declaration of Strengthening the Financial System that promised extensive changes, including expanding the clubby Financial Stability Forum into the more formal Financial Stability Board whose membership includes every G20 country.

The result has been a massive overhaul of international and national financial regulations and supervision. This has responded vigorously to the immediate problems laid bare by the North Atlantic crisis, including stronger international rules on capital buffers, tighter definitions on what comprises bank capital, and new liquidity standards that have decreased reliance on potentially unstable wholesale funding. At the same time, EU leaders have endorsed the principle of a full banking union while in the United States the Dodd–Frank legislation involved a thorough tightening of regulations, including on shadow banks. Indeed, the current conjuncture may well be a high water mark for bank regulation as the political land-scape has started to turn against tougher rules. Some proposed reforms may well not go forward or be watered down and others that have been passed could be rolled back or weakened. This is therefore a good time to assess how far the regulatory revamp has solved the long-term weaknesses that drove the North Atlantic into crisis discussed earlier in this book, rather than simply responding to immediate symptoms.

The initiatives taken internationally, in Europe, and in the United States all have the same aim of making the banking system sounder using the principles articulated by the G20. However, despite the similarity in purpose, each set of initiatives basically deals with different aspects of a common problem. As such, it is easiest to explain the content and implications of each set of reforms in turn.

* * *

Upgrading the Basel Rules

The crisis laid bare the inadequacy of the Basel 2 international regulatory rules, as numerous banks that adhered to them had to be rescued or given major government support in response to severe market pressures. Rescues included IKB Deutsche Industriebank (Germany), Dexia (France/Belgium), Fortis (Belgium/the Netherlands), Banco Português de Negócios (Portugal), Anglo Irish Bank (Ireland), Bankia (Spain), as well as Citi and the independent US investment banks Bear Stearns (sold to JP Morgan Chase) and, most famous of all, Lehman Brothers (whose post-bankruptcy remnants were sold to Barclays Bank). In addition, many of the Euro area mega-banks came under sustained market pressure, including Deutsche Bank (the largest), ING (which accepted a capital injection from the Dutch government), and several French banks, while in the United States the two largest independent investment banks (Morgan Stanley and Goldman Sachs) saved themselves by becoming regulated banks which gave them access to emergency Federal Reserve funding. The failure of the existing Basel regime could not have been plainer.

The ensuing reforms comprised an emergency fix followed by a new framework. The awkwardly named Basel 2.5 system was an emergency fix that shored up capital charges for market risk. One reason that the internal risk models used in Basel 2 had generated inadequate capital buffers was that they used historical patterns of shocks and correlations across asset prices. Basel 2.5 moved the system from these value-at-risk models (VaRs) to stressed value-at-risk models (SVaRs) which incorporate the larger financial shocks, higher correlations across asset prices, wider margins between the buy and sell prices of assets, and the higher default risks seen in times of market panic. Some specific provisions for securitized assets, whose markets had largely seized up in the immediate aftermath of the Lehman Brothers bankruptcy, topped off this emergency revamp of capital buffers for market risk.

The 2011 Basel 3 agreement was a more fundamental overhaul of international rules on bank safety. Given that the preamble to the new guidance describe the goal as providing a “comprehensive reform package addresses that the lessons of the financial crisis”, it is worth critically assessing the Basel Committee’s assessment of what had gone wrong.1 The Committee ascribed the crisis to the build-up of excessive financial leverage (both on- and off-balance sheet), an erosion in the quality of the capital base, and insufficient liquidity buffers. These weaknesses meant that banks were unable to ride out the temporary trading and credit losses, so that banking problems were rapidly transmitted to the rest of the financial system and the real economy, resulting in unprecedented injections of liquidity, capital support, and guarantees that exposed taxpayers to large losses as well as spillovers to the rest of the globe. Notably absent from this assessment was the manipulation of risk weights by large banks using internal risk models. The Basel Committee was not willing to acknowledge that the switch to internal risk models had been a mistake. That rather than improving bank regulation, the switch had been driven by a view that regulation was not needed as market discipline was more effective, a flawed approach that had allowed the emergence of undercapitalized mega-banks on both sides of the Atlantic.

Based on this diagnosis, the Committee’s new capital rules aimed to “improve the banking sector’s ability to absorb shocks from financial and economic stress, whatever the source”, with subsidiary objectives to “improve risk management and governance as well as strengthen banks’ transparency and disclosures” and “to strengthen the resolution of systematically significant cross-border banks”. Given the size of the changes, provisions are being phased in gradually, so that the entire Basel 3 package will not be in place until 2019, although in practice the impact on banks is quicker as market analysts are already comparing bank performance with these future requirements.

The Committee proposed a major strengthening of the existing risk-based capital framework. The definition of bank capital was tightened, with common equity (the most effective type of capital) raised from one-half to three-quarters of core capital and less effective hybrid instruments that had crept into the definition gradually eliminated. In addition, rules on supplementary “Tier 2” capital were tightened, and the even more supplementary “Tier 3” capital eliminated.

These improvements in the quality of capital were accompanied by an increase in the size of the capital buffers that banks had to hold against risk-weighted assets. Banks are required to hold common equity of at least 7 percent of risk-weighted assets (a baseline figure of 4½ percent is augmented by a 2½ percent “capital conservation” charge that provides an additional buffer than can be eroded in the face of a shock). Regulators can also choose to add a further counter-cyclical charge of up to 2½ percent if they are worried that bank lending is rising too fast during an economic boom. Should a bank allow its capital to fall below the minimum it faces steadily increasing constraints on how much of their profits can be distributed to shareholders. This compares with a pre-crisis requirement to hold only 4 percent of risk-weighted assets in core capital, only half of which had to be common equity.

Large “systemic” banks face a further charge of 1 to 2½ percent, implying a minimum common equity buffer of 8 to 9½ percent of risk-weighted assets. This charge is aimed at offsetting the wider risks that they pose and the funding advantages associated with being viewed by markets as too big to fail. In addition, globally important systemic banks will be subject to holding further capital buffers of 16 to 20 percent of risk-weighted assets (the precise figure has yet to be determined) comprising capital and unsecured bonds, adding a further layer of capital buffers, with the objective of ensuring that there are ample private lenders who can participate in any bank rescue.

To further protect tax payers from bearing the brunt of the costs of bank rescues, the EU has augmented these provisions with rules that require private lenders to participate in bank rescues to discourage generous government bailouts (US rules already required that the Federal Deposit Insurance Corporation choose a rescue plan that minimizes the costs to the federal government).

In a partial acknowledgement of the problems posed by internal models, the Committee also introduced a leverage ratio based on total assets. The acknowledgement of the need for a leverage ratio using total assets is an important step forward by the Basel Committee. However, the Committee made clear that the leverage ratio was only a backstop to the risk-weighted calculations. It requires banks to hold 3 percent of assets as core capital, a looser definition of capital than the common equity required for risk-weighted assets, and there is no supplement for systemic banks. While the Committee will revisit the size and definition of the leverage ratio in 2018 based on the experience to that point, the current pushback against stronger bank regulation does not bode well for a tougher approach.

Basel 3 also introduced two new liquidity standards aimed at ensuring that banks are protected against market contagion. The liquidity coverage ratio requires banks to hold enough saleable assets to survive a severe financial shock, in other words that investment banking operations are sufficiently backstopped with ready money. The net stable funding ratio (NSFR), on the other hand, limits the degree of maturity mismatch by constraining the degree to which long-term lending (loans of a year or more) can be financed through short-term and potentially volatile wholesale funding.

Basel 3 likely reflects the apex of tougher bank regulation and the accompanying burden on banks. Recently, attempts to further toughen the Basel 3 rules with an additional package known colloquially as “Basel 4” that was championed by the US Federal Reserve is now in limbo after it was postponed due to strong European objections, particularly from the Germans. This difference in views partly reflects long-standing differences in the philosophy on the role of banking in the economy. The US and the UK, who rely more on direct finance of companies via bond markets, are less concerned about abrupt changes in bank regulation than their French and German counterparts where banks are the major source of finance for firms.2 The “Basel 4” initiative has also been weakened by the resignation of the Federal Reserve official who took the lead in the Basel negotiations and by the Trump administration’s less stringent view on bank regulation.

While the Euro area and the United States have both adopted the Basel 3 framework, there are important and revealing differences that loosen the rules for Euro area banks while tightening them on US banks.3 The most important difference is that the European Central Bank decided not to adopt the leverage ratio as a binding constraint, which means that large Euro area banks remain only subject to capital charges based on internal risk models. While it seems likely that the European Commission proposal to make the leverage ratio binding in 2018 will come into effect despite the current backlash against stricter regulation, the devil will be in the details. It is quite possible that the final implementation of the leverage ratio will include tweaks of rules that make it less stringent, as has occurred with other elements of the Basel 3 package. By contrast, the US has embraced the leverage ratio for its banks and, for the eight globally significant US banks, it increased the requirement to a more meaningful 5 percent of assets (6 percent for regulated banking backed by deposit insurance).4 In addition, as discussed further below, the US-specific Dodd–Frank Act imposed a further backstop on the way that banks can calculate risk-weighted assets that also constrains the ability of US banks to manipulate risk weights using internal risk models.

There are also important differences in the types of banks covered by Basel 3 rules on either side of the North Atlantic. In the Euro area, the framework applies to all banks, and hence continues to favor the competitive position of large banks that can use internal risk models to calculate their capital buffers versus small banks that have to use standardized weights. By contrast, in the United States, Basel 3 applies only to large banks. Smaller banks are subject to older rules that fit more easily with US accounting rules, including the US version of the leverage ratio. Overall, the subtractions from the Basel guidance led the Basel Committee to assess that the European Union was “materially noncompliant” with Basel 3 while the additions by the United States left them with a “largely compliant” status. By contrast, most of the rest of the world adopted the suggested package without significant changes and hence are simply “compliant”.

The key question is how well these changes in the Basel framework have addressed the pre-crisis weaknesses in bank regulation. The first part of this book argued that the fundamental flaw in the system was that large European and US investment banks could manipulate internal risk models to gain a competitive advantage over the rest of the system. This advantage was especially potent in boom times when high asset prices justified extremely thin capital buffers. Looked at from this perspective, the Basel reforms provide only a partial solution to the underlying issues. On the positive side are more stringent rules on capital buffers, including surcharges for systemic banks, and the new rules on liquidity. In addition, the introduction of the leverage ratio is a useful acknowledgement of the deficiencies of only monitoring risk-weighted assets. All of this makes the North Atlantic financial system safer against a new crisis even if some analysts still worry that capital buffers should be still higher.5

On the negative side, however, the fact that at this time Euro area banks remain only subject to internal risk models perpetuates some of the pre-crisis distortions. It provides the large Euro area banks with a competitive advantage over smaller banks, and hence perpetuates mega-banks. Indeed, it is striking that an assessment of the leverage ratio by the European Bank Authority (EBA) found that, in contrast to smaller banks:

The quantitative benchmarking results give indications of a potentially elevated exposure to REL [risk of excessive leverage] in the case of the largest and most complex credit institutions, in particular those that operate the business model of a ‘cross-border universal bank’ and that are, at the same time, GSIIs [Global Systematically Important Institutions] (emphasis added).6

This implies that capital remains thin compared to total assets despite additional capital charges for systemic banks, and the EBA goes on to recommend a higher leverage ratio for such institutions. Recent market jitters with regard to Deutsche Bank also suggest that capital buffers for large Euro area banks remain problematic. The fact that the Euro area banks are not subject to the leverage ratio also gives them a potential competitive advantage over their international competitors, especially systemic US banks that are under additional constrains over and above baseline Basel 3 rules. At present, the advantage conferred by internal risk models is limited since depressed asset prices imply a more pessimistic assessment of credit worthiness and higher capital charges. With asset prices starting to recover, however, these models could again promote a cyclical boom in bank lending based on an inappropriate thinning of capital buffers. This risks restarting the cross-Atlantic financial drift that helped to fund the US housing bubble. How far can better supervision coming from moves to a Euro area banking union offset these risks?

* * *

Moving Toward a Euro Area Banking Union

In June 2012, even as the Euro crisis continued to rage, European heads of state agreed in principle to move to a banking union in which all Euro area banks would be overseen and supported by the European Central Bank (ECB) rather than national governments. This decision holds the promise of eliminating competition across national regulators and hence solving a major weakness in the design of the Euro area. Indeed, the announcement had an almost immediate dramatic effect on the Euro area crisis although in a somewhat indirect way. This was because it paved the way for President Draghi of the ECB to pledge that he would do “what it takes” to preserve the Euro, a move that rapidly ended the acute phase of the European crisis. Indeed, the promise of ECB support was so successful in calming market fears that the central bank was never required to activate the associated new Outright Monetary Transaction program designed to provide support to stressed Euro members.

The agreement to move towards a banking union reflected frustration at the failure of national regulators to “break the vicious circle between banks and sovereigns”. Over the crisis, uncertainty over the costs of bailing out banks led to questions about the financial stability of the governments of Euro area members in crisis. Simultaneously, questions about the ability of these governments to provide banking support raised the riskiness of banks. The result was a vicious circle as the two processes reinforced each other and borrowing costs spiraled upwards. Much of the problem stemmed from the fact that markets assumed the worst because national regulators were unable to paint an accurate picture of the true state of the Euro area banking system. In particular, nationally led European stress tests in 2009, 2010, and 2011 were not deemed credible by the markets because individual national regulators were seen as providing low estimates of the true problems in their banking system since they did not want to disadvantage their own banks and sovereigns compared to those in other members. The failure of the European stress tests to calm market jitters over the state of the banking system was in sharp contrast to the success of the US stress tests of May 2009 conducted by the Federal Reserve Board and thrift supervisors that almost immediately calmed markets as they were seen as a realistic assessment of the amount of capital needed by US banking system.

The agreement to move to a banking union resulted in centralized supervision of Euro area banks at the European Central Bank. ECB supervision could be implemented relative easily because it had been contemplated as part of the Maastricht Treaty, and hence did not involve opening the Pandora’s Box that a major treaty revision would entail. While Maastricht clearly gave the prudential supervision of credit institutions and the stability of the financial system to the “competent national authorities”, a clause allowed the European Council to transfer these functions from national regulators to the European System of Central Banks if it voted unanimously on a proposal coming from the Commission, after having consulted the ECB, and if the European Parliament agreed. The inability of national regulators to ensure the stability of the banking system by providing credible stress tests over the crisis allowed these hurdles to be overcome.

The ECB now directly supervises 129 significant Euro area banks that include over 80 percent of regional assets.7 Inspections of these banks occur through Joint Supervisory Teams headed by a coordinator from the ECB (who cannot be a national of the country involved) plus a sub-coordinator (who is such a national). These two lead a team comprising ECB staff and supervisors from the home country as well as any other Euro area countries in which the bank has major operations. This team reports to the ECB’s Supervisory Board, comprising a chair and vice-chair from the ECB Executive Board, four more ECB employees, and representatives from all countries. The remaining three thousand-odd Euro area banks are supervised at the national level under the oversight of the ECB, whose job is to maintain consistency and provide technical support.

The 129 significant Euro area banks supervised by the ECB can be stratified into three groups depending on their international reach. Three Euro area banks are truly international—Deutsche Bank of Germany and Santander and BBVA of Spain—with over 30 percent of their assets outside of the EU, so that effective supervision implies the need for close collaboration with non-EU supervisors. A further seven banks can be thought of as pan-European, in that more than 30 percent of their assets are in the rest of the European Union. These are the banks whose supervision benefits most from the switch from national to centralized banking supervision since it allows European supervisors a fuller assessment of their overall operations. The remaining 119 banks are primarily national, with over 70 percent of their assets in their home country. Strikingly, while the ten international/European banks are owned by private shareholders, many of the 119 nationally orientated banks are owned by less transparent cooperatives or by the government. These arrangements create a strong national bias that limits their ability to expand overseas or to be acquired by other banks, and suggest that there remain major institutional limits on how much further the integration of the European banking system can proceed.

How do these changes in the Euro area banking system compare with its pre-crisis deficiencies? As outlined in the first part of this book, the combination of universal banks (as a result of the single market program), regulatory competition (as a result of the Maastricht Treaty), and internal risk models for investment banking (as a result of the Basel market risk amendment) created a small number of nationally oriented mega-banks that were large as a proportion to their own economies and generally had bloated investment banking operations. Over the boom, these mega-banks manipulated the flexibility of the internal risk models to thin their capital buffers, and reallocated the capital that was freed to make increasingly unwise loans that led to a banking crisis. The move to ECB supervision clearly solves the problems associated with competition across national supervisors, including the incentives of national supervisors to support national champions, discourage foreign entry, and undermine stress tests of the European banking system. Indeed, banking union holds the promise of creating a truly integrated Euro area banking system.

Crucially, however, the banking union remains only half formed.8 This is because responsibility for bank rescues remains largely with national governments. Bank deposit insurance is still the province of individual members within European Monetary Union and, while a European-wide bank resolution system was agreed in July 2014 funded from charges on banks, it will take seven years to mature and even when complete will be too small to cope with any major bank rescues. In theory, there is also a (limited) option to provide centralized support for Euro area banks using European Stability Mechanism (ESM), a bailout fund for countries in crisis. Indeed, in 2012 there was an expectation that EU funds would be used to directly recapitalize and restructure the Spanish banking system. If the EU had indeed taken the lead, this would have cemented the use of centralized funds to support Euro area banks and would have been a major move toward banking union. In the end, however, funds were only lent to the Spanish government, which bore responsibility for the actual bank restructuring. In practice, therefore, bank support currently remains in essence a national responsibility within the Euro area. Since the mega-banks are so large compared to national output, this also implies that some banks remain too big to fail.

The split between centralized bank supervision and national responsibility for bank rescues risks undermining the effectiveness of centralized ECB supervision of Euro area banks. This is because it replaces one set of misaligned incentives with another set. The problem with the pre-crisis system was that national regulators were responsible for both bank supervision and support, creating incentives for a regulatory race to the bottom. The new system generates a new misalignment of incentives since national regulators, who have responsibility for bank support, will generally want to minimize their assessment of banking problems and the associated costs which will put them at odds with the ECB, which is responsible for supervision. Indeed, this already seems to be occurring in the case of the Italian banks where the ECB is keener on a comprehensive approach to cleaning up the problems of the system than the Italian authorities. Unless this dichotomy is resolved by creating a Euro area bank rescue fund, the gap between the centralized responsibility for supervision and national responsibility for the costs of bank cleanup will continue to create unhelpful tensions between the ECB and national authorities. This underlines the importance of completing the half-formed Euro area banking union. We now turn to the quite different imperatives involved in US financial reform.

* * *

Taming the US Shadow Banks

US financial reformers faced a fundamentally different issue from those in Europe. Banks in Europe were under a flawed single system. The issues in the United States, on the other hand, centered on the dual nature of the pre-crisis banking system that contained a relatively tightly regulated core and a loosely regulated shadow banking system. Post-crisis changes in US regulation were thus focused on revamping the regulation of shadow banks and altering the dividing line between the two sectors.

The most important reform since the crisis has been to move the investment banks into the regulated banking sector. Most of this change largely occurred by necessity over the crisis, as the independent US investment banks were either rescued and absorbed by regulated banks (Bear Stearns, Merrill Lynch, and Lehman Brothers) or voluntarily converted to regulated banks to gain access to Federal Reserve emergency funding (Morgan Stanley and Goldman Sachs). There was also one deliberate action, which was the decision to bring foreign-owned investment banks into the US safety net by forcing them to become bank holding companies.

The conversion of the investment banks into regulated banks has largely tamed the pre-crisis shadow banking system since the lightly capitalized investment banks were at its core. Bringing the investment banks into the regulated banking system with its much more stringent capital buffers has made the investment banks themselves safer. Equally important, however, is that it has largely eliminated the funding advantage of the rest of the shadow banking system. The investment banks were the main providers of funds to other parts of the shadow banking system such as hedge funds. The additional costs of higher capital buffers in the investment banks have been passed on to their clients in the shadow banking sector through higher interest rates on loans.

Bringing the investment banks into the regulated banking sector leveled the playing field between the two halves of the banking system. However, the broker-dealers at the heart of the investment banks are still subject to some differences in regulation compared to commercial banks, although how these procedures would work in the case of a new crisis is unclear given the creation of a system-wide oversight council which can label firms as systemic. In addition, while the FDIC can extend support to broker-dealers in the face of a crisis, it may take a more cautious approach to such nonbanks compared to its traditional bank clients.

This major change in the US dual banking system has been reinforced by numerous other tweaks, most of them contained in the 2010 Dodd– Frank Wall Street Reform and Consumer Protection Act.9 Among its many provisions, the Dodd–Frank Act ended the silos across regulators that had allowed the rapid pre-crisis expansion of shadow banking system. More specifically, the act created the Financial Stability Oversight Council (FSOC), chaired by the Secretary of the Treasury and including all major financial regulators, with a mandate to oversee the entire US financial system for systemic risks and regulatory gaps. For example, the FSOC can require Fed oversight of nonbanks that are viewed as posing a potential risk to financial stability and facilitates information-sharing across agencies. While there are concerns about how effective the somewhat large and unwieldy FSOC would be in the event of a crisis, its creation clearly signaled a desire to end the destructive pre-crisis compartmentalization across US financial regulation.

Another important aspect of the Dodd–Frank Act was the Collins amendment, which limits the ability of large US banks that adhere to the Basel rules to use internal risk models to manipulate risk weights. The amendment requires these banks to calculate risk-weighted assets using the standardized risk weights designed for smaller banks as well as their own internal risk models and use whichever risk-weighted asset calculation is larger. The large US banks are thus subject to three regulatory capital approaches: The opaque internal risk model calculations of risk-weighted assets, the simpler and more verifiable standardized approach to risk weighted assets, and the Basel 3 leverage ratio on total assets discussed earlier. The use of standardized weights and the leverage ratio largely eliminates the ability of large US banks to use their internal risk models to reduce capital buffers, just as before the crisis the simple leverage ratio protected the US regulated banks from the erosion of capital buffers seen in the Euro area and US investment banks.

Numerous other aspects of US financial regulation were strengthened by the Dodd–Frank Act. Among the most important is the Volcker rule, which disallows banks from most trading in markets on their own behalf and raised the capital buffers on the reminder. Capital requirements were also tightened on repurchase agreements (repos) and securitizations, the latter of which has essentially ended private securitizations. In addition, complex institutions such as universal banks are also required to create living wills that specify how they will be resolved in the event of bankruptcy, many shadow banks such as hedge funds are now registered with the Securities and Exchange Commission (SEC), which also now oversees the credit agencies, and the Consumer Financial Protection Bureau was created. On the other hand, the international effort to move trading of derivatives to centralized platforms to make the market more transparent and avoid the confusion associated with the pre-crisis spaghetti bowl of bilateral trades has met with only partial success. In particular, while the US repo market has shrunk and the length of the typical loan has been extended, a lot of trading has moved from the more transparent and more easily regulated trilateral market, in which a third party holds the collateral, to the hazier bilateral market in which the lender and borrower dispense with third-party assistance.

A final major element of US regulatory reform was the nationalization of the two major housing government-sponsored enterprises (GSEs), Fannie Mae and Freddie Mac. Before the crisis, the status of these institutions was ambiguous as they were private institutions with a public purpose. The markets treated them as having a pseudo-government guarantee, on the (accurate) assumption that in the event of a crisis the federal government would take control of them. With the private mortgage-backed asset markets moribund, essentially all new mortgage-backed securities are now being issued through Fannie and Freddie, and hence have federal guarantees on the quality of the underwriting. In short, the US mortgage-backed securities market has in effect been nationalized.

How do these changes respond to the problems observed in the pre-crisis US financial system? Three major trends created the US financial crisis. The first was the expansion in securitization of mortgages as higher capital buffers for regulated banks provided incentives for them to sell such loans to shadow banks with lower capital buffers. The second was the accompanying expansion of the lightly regulated shadow banking system centered on the investment banking groups, including their broker-dealer operations that borrowed wholesale funds from those with large pools of cash and lent them out to other parts of shadow banking such as hedge funds. Finally, the widening of repo collateral to mortgage-backed securities and foreign bonds led to an increase in private label securitizations based on questionable mortgages and an increase in dollar borrowing and purchases of such US assets by foreign banks.

The US regulatory revamp has largely negated these underlying weaknesses, although future changes in regulation could undermine this progress. Bringing the investment banks under Federal Reserve supervision has eliminated the differences between the capital buffers that provided the regulated banks with strong incentives to sell loans to investment banks and the rest of the shadow banking system. In addition, higher capital charges on securitized assets and repurchase agreements led to a dramatic decrease in the size of investment banks, including their loans to the rest of the shadow banking system. The result has been the effective elimination of private securitizations even in areas such as car and credit cards where private securitizations had always dominated. On the other hand, while differences in the size of capital buffers between US regulated and shadow banks have been eliminated, there remain significant gaps between the capital buffers required for Euro area and US banks. As discussed earlier, this risks the reappearance of trades based on this gap.

Recent moves by the Trump administration suggest a willingness to reduce the burden of financial regulation. While most of this effort is directed at small banks, these include moves to curtail the powers of the FSOC and of the Consumer Financial Protection Bureau and to weaken provisions such as the Volcker rule. In addition, talk about passing a twenty-first century Glass–Steagall law opens up the possibility that investment banks will again face less stringent regulation than regulated banks, potentially reopening the differences in regulation that helped drive the crisis.

On a wider level, the difference between the European regulators’ past and present focus on risk weights based on internal models and the US preference for including an unweighted leverage ratio leads to the question of the relative merits of the two approaches. Clearly, a risk-weighted measure would be more sensible if risk could be accurately measured, as bank capital buffers are held to offset potential losses that should vary with the type of loan. On the other hand, several studies have found that the leverage ratio based on total assets was a better predictor of bank failures than risk-weighted ratios in both the US and Euro area over the crisis, suggesting that an unweighted measure of assets is safer despite its apparent bluntness.10 The evidence about which banks were hit hardest over the crisis, however, may reflect the nature of this particular crisis, which involved a prolonged period of market turmoil. Such turmoil would naturally have its largest effect on banks with large investment banking operations. In addition, leverage ratios create their own distortions, particularly by lowering the incentive to keep low-risk loans on the books, implying that either fewer safe loans will be made or that banks will sell them to institutions under less onerous capital charges.

Ultimately, it is difficult to determine whether a generic risk-based capital standard based on standardized risk weights is better or worse than an unweighted one. This suggests aiming for a regime in which both matter by making sure that capital buffers are adequate using both the leverage ratio and the risk-weighted approach. Most basically, however, it is difficult to see why regulators continue to let banks use individual internal risk models to calculate risk-based capital charges rather than using standardized risk weights determined by supervisors. These models were introduced as a method of reducing the role of bank regulation on the assumption that market discipline would create the needed constraints on bank behavior. The crisis amply demonstrated that this approach was a mistake, and that using proprietary models to calculate capital buffers can create powerful incentives for a regulatory race to the bottom in search of short-term competitive advantage.

* * *

Charting the Post-Crisis Changes in the Financial System

It is useful to take a snapshot of the impact to date of financial reforms of the US and Euro area banking systems. While financial reforms remain a work in progress, with parts of Basel 3 only fully implemented in 2019, market analysts are already comparing bank performance with the eventual standards, which suggests more adjustment may have been made in practice than on paper. In addition, the data indicates that there have been limited changes in both systems between 2014 and 2016, suggesting that they are no longer experiencing radical structural shifts. At the same time, it needs to be recognized that the macroeconomic situation remains highly abnormal, with negative policy rates in the Euro area, some periphery banking systems still facing major problems from nonperforming loans, and banks continuing to hold large reserves at the central bank. Even in the United States, where economic recovery and bank cleanup are more advanced, interest rates remain low and reserves high.

There is also a question about the appropriate benchmark to compare the current situation with the past. Since the Euro area and US financial systems had become unsustainably large by 2008–09, the simple fact that they have subsequently shrunk says little about how far the system has been healed. This snapshot assumes using 2002 as a year in which the two systems were still relatively stable. This is the year before the SEC decision to widen the collateral eligible for repos. The first part of this book argued that the banking systems in the Euro area and the United States were relatively stable before this decision, and that it was after 2002 that the financial boom that led to major domestic and international financial imbalances gained steam. Consistent with the policy discussion earlier in this chapter, the results suggest that the Euro area banking system is much further away from being repaired than the US system.

Euro Area Banking Repair Remains Work In Progress

The Euro area banking system has shrunk since 2009 but remains much larger in proportion to the economy than it was in 2002 (Figure 46). Since peaking at well over 300 percent of output in 2009, the Euro area banking system has shrunk about one-third of the way back to its 2002 size. Within this overall pattern there is a significant difference between the core and the periphery. Assets in the Euro area core banks have shrunk about half way back to their 2008 values while in the periphery the equivalent shrinkage is one-fifth (although the core banks remain considerably larger in absolute terms and more dependent on investment banking). This suggests that major further Euro area bank restructuring is needed, particularly in the periphery where banks have limited the shrinking of their assets by increasing their holdings of zero risk-weighted government debt which has further linked the financial fortunes of banks with the government.11 This is also consistent with other assessments that suggest that the Euro area banking system remains overbanked with low profitability.12

Figure 46.Euro area core bank assets have shrunk more than those in the periphery.

Source: ECB Banking Statistics.

The compression in the size of the Euro area core banks largely reflects a pullback from foreign ventures in the United States and the periphery (Figure 47). Assets of Euro area core banks in the periphery, the United Kingdom, and the United States have approximately halved since 2008 as a ratio of output to below their 2002 levels. The pullback from the periphery may help explain some of the difference in the size of the contraction in banking systems between the two halves of the Euro area. The shrinkage of the UK/US operations reinforces the evidence of a significant pullback from investment banking. In additions, commercial loans have risen in importance to around their 2002 level when the data are adjusted for continuing large holdings of central bank reserves.

Figure 47.Euro area core banks have pulled back from the periphery and the US/UK.

Source: BIS banking statistics and EU national accounts.

The twelve Euro area mega-banks have shrunk as a proportion of the overall banking system, but remain more important than in 2002 (Figure 48). Most of the mega-banks survived the crisis with varying types of support, the exception being the French/Belgian bank Dexia which has been broken up. The pullback has been more marked for the eight banks headquartered in the Euro area core such as Deutsche Bank of Germany and BNP Paribas of France, plausibly associated with the retreat from investment banking that was always a much more important part of their business. By contrast, assets of the more commercially orientated periphery mega-banks have shrunk by much less. Indeed, the assets of the two Spanish mega-banks, Santander and BBVA, have expanded as a ratio of output.

Figure 48.Mega-bank assets have shrunk but remain a large component of the Euro area banking system.

Source: Bloomberg and Euro Area Banking Statistics.

Tighter regulation has improved capital buffers for the Euro area mega-banks. Figure 49 updates the relationship between the risk-weighted Basel capital ratio (on the horizontal axis) and the unweighted ratio of equity to assets (on the vertical one) earlier reported in Figure 25 in Chapter 3. Euro area bank capital buffers have improved compared to 2008 (and 2002). Risk-weighted capital ratios have leapt and the tail of banks with exceptionally low ratios of equity to assets has been eliminated. In addition, the clear negative relationship between the two measures of bank soundness seen in 2008 has disappeared as banks with large investment operations have been forced to raise more capital, but the positive relationship seen in 1996 has not been restored. In addition, as discussed earlier, capital buffers will likely thin as asset prices recover.

Figure 49.Euro area mega-banks have strengthened capital buffers.

Source: Bloomberg.

US Reforms Seem More Complete

Tougher post-crisis regulation has changed the structure of the US financial system more fundamentally than the European one. Assets of regulated banks have expanded slightly compared to output while those of investment banks have fallen by almost a half (Figure 50).13 This reflects the leveling of the regulatory playing field between the regulated and investment banks compared to the much looser regulation of investment banks before the crisis. As a result, despite the expansion in regulated banking since 2008 as a ratio to output, the size of the overall dual banking system has shrunk back close to its 2002 value. This suggests that most of the excesses in US banking have been redressed.

Figure 50.US commercial bank assets rose after the crisis, investment bank assets contracted rapidly.

Source: US Flow of Funds.

Note: Investment bank assets are calculated as twice the size of broker-dealers.

The private securitization market effectively disappeared after the regulatory playing field was leveled, underlining that the market was driven by pre-crisis differences in capital buffers between the regulated and investment banks rather than fundamentals (Figure 51). The eclipse of private label securitized assets can be seen most clearly in the market for consumer loans (credit card debt and car loans). One-quarter of this debt was sold to markets using private securitized assets in 2008 (a similar proportion to 2002), while by 2016 such securities were negligible. There are still some private mortgage-backed securities in the market, but this reflects the remnants of pre-crisis issuance rather than new activity, and the stock of assets continues to fall. The flip side to the disappearance of the private mortgage-backed securities has been a larger role for those issued by the GSEs. Around 60 percent of US mortgages are now guaranteed by the federal government through Fannie Mae and Freddie Mac, 10 percentage points higher than in 2002. This outsized federal role in the mortgage market is likely to likely to continue as reform of the GSEs has been put on the political back-burner.

Figure 51.Private securitization has dwindled.

Source: US Flow of Funds.

Tougher regulation of the investment banks can be seen in the enormous rise in their capital buffers. Figure 52 compares the average leverage ratio for the independent US investment banks versus the US commercial and universal banks. At its pre-crisis nadir in 2007, the investment banks had a leverage ratio of just over 3 percent, only one-third of those at regulated banks and modestly lower than in 2002. By 2016, with the independent investment banks having been converted into bank holding companies and hence under the same regulatory regime as other banks, the capital buffers for the two remaining independent investment banks had almost tripled to 8 percent of assets. There is still a gap with the rest of the banking sector, whose capital buffers had also increased modestly as a result of tougher regulations. However, the limited size of this gap suggests it could plausibly reflect difference in underlying risk due to business models.

Figure 52.Capital increased most for investment banks.

Source: SNL.

* * *

The Road Ahead

The post-crisis period has seen the largest rejigging of the North Atlantic and global financial system since 1988 when Basel 1 imposed (relatively) consistent capital charges across countries. In the process, capital buffers have been increased, particularly for investment banking, rules on liquidity have been introduced, and the use of short-term wholesale funds has been reduced. While legacies of the crisis, including disappointing output growth and a volatile trading environment, have delayed the return to financial normality, particularly in the Euro area, it is clear than the North Atlantic financial system is more stable than it was before 2008.

Many of the weaknesses within the Euro area and the United States banking systems have been tackled. Within the Euro area, the major banks are under the centralized supervision of the ECB, which has put an end to the supervisory competition that initiated a pre-crisis race to the bottom as national regulators supported the expansion of their own national champion banks at the expense of sound supervision. In the United States, the remaining investment banks are now under federal supervision and support rather than being allowed to operate with inappropriately thin capital buffers. While other, less systemic parts of the shadow banking system remain largely outside of the regulatory framework, they have shrunk as short-term funds from investment banks have become more expensive and as the supply of complex financial instruments driven by differences in regulation has withered. The financial market place has become simpler and the corresponding trading strategies less obscure.

Looking ahead, the Euro area and United States face different banking challenges. In the United States, the priority is to ensure that legitimate simplification of red tape faced by smaller banks does not morph into a rollback of needed tougher regulation of large banks. In particular, any new version of the Glass–Steagall Act separating banks into investment and commercial banks should not be used to lower the stringency of regulation on the former.

The Euro area needs to complete its planned regulatory revamp and shrink banking assets to eliminate the overbanking that is crimping profits and setting the stage for future banking problems. The major priorities are twofold. First, introduce a meaningful leverage ratio above the Basel minimum of 3 percent ratio on the mega-banks, as well as a floor that limits the abuse of risk weights through internal risk models, such as the US approach of forcing large banks to also calculate risk-weighted assets using the standardized approach. Currently, the capital buffers for the Euro area mega-banks are still calculated using internal models whose properties are not well understood or closely policed. As asset prices recover, there is a risk that capital buffers of the mega-banks will erode further as models interpret high asset values as meaning lower risk. Even now, the advantage from internal risk models have meant that the European banking system continues to be dominated by “too big to fail” mega-banks despite post-crisis talk of cutting down the size of banks by raising capital buffers on systemic banks. Ensuring that Euro area banks are under similarly tough regulation to those currently in place in the United States will also avoid the risk of restarting “North Atlantic financial drift” trades created by looser regulation in the Euro area rather than competitive advantage, and which allowed a US housing crash to create a Euro area banking crisis.

Second, the Euro area banking union needs to be completed by setting up a centralized Euro-area-wide system of bank support. At present there is a risk that the new misalignment of incentives in Euro area bank regulation will slowly erode the advantages of centralized ECB banking supervision. The pre-crisis misalignment was that individual national bank supervisors were in competition with each other and allowed banks to expand too fast. The new misalignment is that supervision is centralized but the costs of banking problems remain national. Unless bank deposit insurance and support is also centralized, there is a risk that the national supervisors, who are a crucial component in making centralized supervision work, will begin to start gaming the central supervisors.

    Other Resources Citing This Publication