Financial regulation.

Barrell, Ray ; Davies, E. Phillip

Introduction

The financial crisis that engulfed the world in 2007 and 2008 has led to a wave of re-regulation and discussion of further regulation that has culminated in the proposals from the Basel Committee as well as those in the Vickers Committee report on Banking Regulation and Financial Crises. This issue of the Review contains a number of papers on Banking Regulation, covering many aspects of the debate, and we can put that debate in perspective through these papers and also by discussing our work on the relationship between bank size and risk taking, which is reported in Barrell et al. (2011). We addressed the causes of the crisis in the October 2008 Review, and began to look at the costs and benefits of bank regulation in Barrell et al. (2009). In that paper we argued that we needed to know the causes of crises and whether the regulators could do anything to affect them before we discussed new regulations. It is now generally agreed that increasing core capital reduces the probability of a crisis occurring, and most changes in regulation that are being discussed see this as the core of their toolkit. The work by the Institute macro team in Barrell et al. (2009) and in Barrell, Davis, Karim and Liadze (2010) was the first to demonstrate that there was a statistically important role for capital in defending against the probability of a crisis occurring, and our findings were widely used in the policy community in the debate over reform.

Although banks and financial markets are very complex structures they serve very simple purposes, taking in the assets of persons (1) in the economy and pooling them to lend money to other persons, and in the process providing a significant part of the means of payment for goods and services. The process of lending is a risk), one, and banks have assets (loans) in excess of their liabilities (deposits) in order to be able to absorb risk and ensure an adequate distance from default. The difference between their assets and their liabilities is their loss absorbing equity base, which is not the same as their equity market value, as this will include goodwill and other valuation factors that cannot be used to cover losses on their loans and other assets. If the level of capital is too low, and hence the distance to default is too small or the losses too large, then banks collapse. Once one bank fails it is possible that its loans will have to be called early, and this may cause an implosion of credit in the economy. Banking crises of this sort: are to be avoided as they impose significant social costs. It was common in the early part of this century to claim that economics had made so much progress that banking crises had been abolished. (2) In particular, a combination of inflation targetting with a simple rule, fiscal inactivity and efficient financial markets was thought to be sufficient to ensure stability. Basing policy on deductive reasoning rather than on evidence has become common in economics, as Goodhart (2009) discusses, but it is to be hoped that the failures of such reasoning are obvious after the events of the past four years.

Market regulation

Financial markets are like any others, they exist within a framework of legislation and regulation. There are three functions for regulation, the protection of the direct interest of the consumer through product regulation, the protection of the consumer against monopoly power through structural regulation and the indirect protection of the consumer through regulations designed to reduce spillovers and contagion from damaging events such as bank failure. The first two may be described as micro prudential and the latter as macro prudential. Bank regulation has a number of layers of responsibility as well, with the overall framework being set by the Bank for International Settlements in Basel with additional country specific additions to those regulations. In the European Union the most significant parts of the regulatory framework are agreed across the whole European Economic Area, and in particular the basic structures of capital and competition legislation are shared. In all countries liquidity regulation and direct consumer protection are a national responsibility, but (branch based) banking activity can be undertaken outside the home country with little restriction. (3)

There were clearly many flaws in the regulatory structure before the crisis in 2007, with perhaps the most severe being a reliance on the market to regulate capital adequacy and especially liquidity. The UK had a very lax attitude to liquidity regulation, and relied on the existence of market or wholesale liquidity to provide for the needs of individual banks. Indeed as the liquidity crisis struck in the summer of 2007 it would appear to be the case that the UK banking system was holding less liquidity in aggregate than the floor of 3 per cent each bank was required to hold. The non-systemic approach to regulation followed by the regulators meant that the authorities believed banks, if they faced liquidity problems, could turn to the wholesale market. It dried up at just the point it was needed. Although this was a common experience, for instance in the 1970s secondary banking crisis in the UK, the scale of reliance on the wholesale market was new.

The UK and the US, along with many countries in Europe, felt that there was little need for macro prudential regulation, as is discussed by Erlend Nier in this Review, and as a result many of the barriers that would have prevented crises, and especially crisis contagion, were not in place. Light touch regulation of the assets and activities of banks was supposed to stimulate growth and increase incomes, and in the decade to 2007 the financial sector increased in size in many countries, as Barrell, Holland and Liadze (2010) discuss. However this increase in size was at least in part the consequence of rent seeking activity based on the construction of complex products. It was also accompanied by an expansion of lending to people with limited ability to repay loans, especially in the US. Banking regulation could have been designed to take account of these issues, but it was not. The most important part of the regulatory framework, provisions for capital adequacy, was particularly weak.

The Sub Prime crisis demonstrated that the overall level of capital proved to be too low to protect a number of institutions against the losses they incurred. Up to 50 per cent of capital could be held in the form of subordinated debt (Tier 2), which does not give the level of protection to banks that equity does. Indeed, in the crisis market participants focused on common equity only as a measure of banks' robustness. The regulatory structure lacked a measure relating the total assets of a bank to its equity capital, (4) and relied on inadequate and pro-cyclical measurement of risk. These together gave incentives for disintermediation which led to banks' effective exposures being undercapitalised. The Basel II framework lacked any international agreement on liquidity, whereas liquidity risk was the key component of the crisis, particularly up to the Autumn of 2008. No reference is made to the 'too big to fail' problem, which means that large banks have incentives to take excessive risks at public expense. Financial innovations were permitted to spread, and their credit ratings were assumed to be accurate, despite the fact that they had not been tested in a downturn. Any new framework has to address these problems whilst ensuring that the changes in regulation do not cause a sharp contraction in activity.

The new regulations, which are basically complete, will raise common equity from the previous minimum of 1 per cent of risk-weighted assets to at least 4.5 per cent, and Tier 1 as a whole to 6 per cent. A conservation buffer of 2.5 per cent of risk-weighted assets must also be built up with common equity, and if this is exhausted in a crisis then the bank will be wound up. The table below sets out details of the new capital structure, and the maximum proportion of Tier 2 is to be substantially reduced from 4 per cent to 2 per cent of risk-weighted assets. A minimum ratio of capital to total (unadjusted) assets of 3 per cent must be held. This should substantially reduce the risk that banks will undertake regulatory arbitrage and hence boost the ratio of their assets to their capital (4) without changing measured risk-weighted capital ratios. There is provision for a countercyclical capital buffer of up to 2.5 per cent of risk-weighted assets, which is to he imposed at the discretion of the regulators. The regulation of subsidiaries and capital market activities has been substantially tightened, including the introduction of stress-related benchmarks for trading book capital and counterparty credit risk. Two new regulations for liquidity risk are being introduced: first, a liquidity coverage ratio enforcing sufficient liquid assets to offset net cash outflows during a 30-day period of stress; second, a net stable funding ratio which seeks to ensure a degree of maturity matching over a one-year horizon, including allowance for off-balance sheet commitments. Although there is no proposal to harmonise emergency liquidity or capital assistance, the problems that emerged when Lehman Brothers collapsed may be sufficient to have tougher regulations not to act alone in future. Basel III does penalise size to some degree via the proposal for higher capital for systemic institutions.

There remain a number of shortcomings of the new proposals, and these are not directly addressed by the recommendations of the Independent Commission on Banking. For example, although the capital adequacy proposals are clearly an improvement, and we agree with a staggered introduction to avoid renewed credit rationing (Barrell et al., 2009), there remain concerns. Notably, the level of capital under Basel III is insufficient to reduce banking crisis risk to acceptable levels. Work at NIESR (Barrell, Davis, Karim and Liadze, 2010a) suggests that 7 percentage points more capital on a non-risk adjusted basis would be needed to reduce crisis risk to 1 per cent in all OECD countries and at all times. The ratio of unweighted to weighted assets comes between countries, but for the UK we would need to double capital requirements to meet this target. It is probably the case that Tier 2 capital is not only inadequate as a buffer but also gives adverse incentives for risk taking, both ex post and ex ante as Barrell et al. (2011) show. A greater proportion of Tier 2 for a given level of total capital is shown to lead to a higher rate of charge-offs and provisions in a 14 country 15 year panel of over 700 hanks in OECD countries. It would have been better to have excluded Tier 2 from capital entirely rather than only to reduce it. The proposal for bank-size-related capital charges is welcome but will not deal with the incentives to take risks in large institutions owing, we believe, to the implicit insurance provided by being 'too big to fail' (Barrell et al., 2011). It would be better to complement higher capital with a form of profit taxation linked to asset risk and leverage, as in risk-based deposit insurance premia--or even break-up of large institutions.

Reform in the UK

The Vickers Commission Report on UK banking has to fit both with the regulatory changes discussed above and the nature and structure of regulation within Europe. Much of the regulation of banking structure comes under European Commission directives, as does the implementation of capital adequacy rules. The UK can only put additional restrictions on banks, and this inevitably leads to an accusation that UK regulation is making the City uncompetitive and damaging the UK economy. (5)

The Basel regulations have not addressed the issue of bank structure and the scope of activities. Permitting the combination of investment banking and related wholesale activities with retail banking was widely perceived to have been a major flaw in the previous regulatory system, and the Independent Commission on Banking addresses this issue. They have recognised that, although there may not be great gains from complex bank scope, there are no great problems with it either. They suggest the construction of Chinese walls between the wholesale and retail entities within a bank group, ensuring that either can go bankrupt without bringing the other to its knees, whilst maintaining any economies of scope that are available. In particular it will remain possible to shift capital between entities within a group as long as the regulatory floors are maintained. This should make the banking system safer.

The Commission also suggests that the retail arms hold 3 percentage points more capital than the Basel floor of 7 per cent. As we have noted, the Basel floor may be too low, and hence this is wise. However, such regulations can only apply to subsidiaries and other entities that are effectively incorporated in the UK, as are UK banks. Extensive branch networks and associated internet banking, such as those set up by Icelandic hanks, cannot be covered by this suggestion under current EU regulations. Hence unregulated banks could be given a competitive advantage, and the economy may be no more stable. If the Europeans as a group agree to move in the same direction, and there is evidence that they will, then consumers' borrowing costs will rise by perhaps 30 basis points, raising borrowing costs on a standard mortgage from 5.0 to 5.3 per cent per annum. It is not clear that the new regulations will impinge on corporate borrowing costs, as these may remain within the wholesale arm with lower capital requirements. However, the definition of retail banking remains unclear, with commercial property lending normally classified as retail, despite its extremely risky nature. These issues will all have to be addressed in the final Commission report in September 2011.

Financial regulation in the wake of the crisis

This issue of the Review includes five articles on regulation and structural change in financial markets, looking back at the experience of the advanced economies in the run-up to the crisis and also suggesting ways that regulation and oversight might improve. The Review addresses the overall structure of regulation going forward, and also the crucial role that enhanced liquidity regulation will play. The latter is likely to reshape the international banking system, leading to a significant re-territorialisation of banking. This process may already be starting in Europe, with cross-border activity peaking in 2007. The possibility of contagion between banking systems is important in our understanding of the recent crisis, with the propagation of the crisis from the US being a worrying, and relatively new, feature of the events. This propagation was not inevitable, and it was the result of international regulatory failures. The massive increase in cross-border holdings of assets may have meant that risks were being shared, but at least in the US it may have meant that the quantum of risks in the system increases. Many international banking sector problems emanated from the US sub-prime market, where poor quality loans were bundled lip into securities that were supposedly high quality, and these securities were then sold on to foreigners. If such out-of-country sales had not been possible, perhaps US banks and regulators would have been more cautious about the production of such poor quality assets. Barrell, Davis, Karim and Liadze in this Review look at the determinants of crises within the OECD, and show that capital and liquidity form defences, whilst property price bubbles cause problems. They also show that the occurrence of a financial crisis in the US raises the probability of crises occurring elsewhere fourfold. The new structure of regulation has not yet addressed this problem properly and it perhaps can only be addressed by the coordination of bank resolution regimes, with clear living wills making the process of dealing with a bank failure much easier. The compartmentalisation of banks into wholesale and retail entities, as proposed by the Vickers Commission, should make such living wills easier to construct.

One consequence of the subprime crisis has been to prompt an overall shift from mainly microprudential to more macroprudential regulation, given the lesson that a focus on depositor protection is not sufficient to mitigate systemic risk. In this context, in this Review Erland Nier from the IMF highlights that there has been a major effort to make macroprudential policy operational, with new committees and authorities being established as well as new powers being granted to existing institutions. In this context, he seeks to distinguish microprudential from macroprudential policy, considers the appropriate mandate and powers of new macroprudential bodies and finally considers what may be the best governance arrangements for macroprudential policies. The new regime will have to cope with the rapid structural change in financial sectors, which will also be provoked by the new regulations themselves. This requires substantial powers for the macroprudential authority over aspects such as information collection and rulemaking. It will also be necessary to ensure appropriate incentives for macroprudential action, when the costs are immediate and the benefits more long-term.

The financial system has always been procyclical, with easy availability of credit boosting growth in the upturn and credit crunches often aggravating the downturn, and this feature was present notably in the subprime crisis. One underlying factor is that provisions are based on immediate risk of loss so cushions are not built up in advance of recessions. Jesus Saurina from the Bank of Spain outlines experience of one of the first systematic macroprudential policies, which long predate the subprime crisis, namely the dynamic provisioning system applied in Spain since 2000, which builds up a buffer of provisioning in economic boom periods to be drawn on in recessions. This system, he argues, has markedly enhanced the robustness of Spanish banks and of the system as a whole. It is worth noting that the provisioning system was evidence based, with its foundation in observations of the behaviour of Spanish banks. Provision had to be built up when growth was strong, and not just when credit growth was strong. This system did not prevent problems emerging in the Spanish savings banks in 2009-10, but their mutual structure made their capital ratios opaque.

In the wake of a boom in cross-border financing up to 2007, a salient feature of the subprime crisis was the impact of cross-border losses, cross-border funding problems and failures of transnational institutions. These in turn prompted the widespread rescue measures that were put in place, including both recapitalisation and liquidity provision. Such issues continue to be highly relevant given the potential losses to EU banks from government defaults on the periphery, and ongoing ECB liquidity support for EU banks. Stephen Cecchetti, Dietrich Domanski and Goetz yon Peter from the Bank for International Settlements examine the range of liquidity measures in Basel III. They conclude that although tighter regulations overall will be beneficial to financial stability at low cost for the real economy, it is the new global liquidity regulations which will have the greatest impact on the pattern of global banking activity. Whereas capital is monitored on a consolidated basis, liquidity regulations will be imposed locally by the host supervisor in domestic currencies. This in turn will favour banks with decentralised multinational structures, as opposed to those currently managing liquidity centrally. Pressures for cross-border operations to be structured as subsidiaries rather than branches--and even for branches to hold their own liquidity--will intensify this change.

While Basel III liquidity rules may well spur a relocalisation of banking, they are not the only pressures moving in that direction. In the EU there was of course a widespread opposite move to more cross-border activity in the period up to 2007, stimulated by the Single Market. However, the rescue operations undertaken widely across Europe--themselves necessitated by poor EU-wide regulation and a worry that some institutions were 'too big to fail'--raise important issues as to whether integration and growth in bank size will continue. This is not least the case given the recapitalisations were often limited to restoring activity in the home country rather than the wider international operations of the banks concerned. Ray Barrell, Tatiana Fic, John Fitzgerald, All Orazgani and Rachel Whitworth from NIESR and ESRI outline econometric work that shows that larger banks have lower net interest margins, owing to economies of scale in portfolio pooling although, beyond a certain very large size level, monitoring costs may rise to such a point that there is some upward pressure on margins. The minimum cost point is reached for large banks with around 4 per cent of the European market. There are banks above this size in most economies, but the structure of European regulation means that they are normally much more important in their home market than in Europe as a whole. If banks relocalise and become smaller, then margins will rise which will raise the cost of borrowing for firms, and this will reduce sustainable output. It will also raise borrowing costs for consumers, reducing the ratio of borrowing to income. This may be desirable, but it will be unpopular. Using NiGEM simulations they show that the reversal of the Single Market in Financial Services would reduce sustainable output in Europe by 0.5 to 1.0 per cent, and the impact would be greater in smaller countries and those where firms are most dependent on bank finance.

Conclusion

Banking crises are often associated with property price bubbles, and they often look like repeated, but forgotten events caused by repeated but forgotten mistakes. The primary mistake is the Panglossian demeanour of much of the economics profession. Theory can always be marshalled to explain why we are in the best of possible worlds, even when we are clearly not. This approach was utilised to justify the UK housing market bubble and the excessive borrowing that it engendered. (6) It was also used to support the same development in the US and also to justify the extremely dangerous deregulation of low income high loan-to-value mortgage markets. Theory also told us market deregulation was wise in banking, and that banks would regulate themselves. Barrell et al. (2011) argue strongly that new regulations should be based on evidence, and that if there is a conflict between theory and evidence economists and regulators should perhaps consider that it is the theory that is at fault, and not reject the evidence, as has been common in academic economics.

In this Review we publish fully spelled out forecasts of the UK, the US, the Euro Area and other economies. Forecasts of this nature, based on data and models (but not over-theoretical DSGE ones) should be part of macroprudential regulation. We regularly warned about the risks of excessive borrowing and house price bubbles in the run-up to the 2007-9 crisis, and stressed the extreme cost of banking crises. In this Review we also publish a note on the determination of house prices in the UK. We suggest they remain 10 per cent overvalued in fundamental terms. We also suggest that tightening loan-to-income conditions for mortgages will reduce both borrowing and house prices further. As the FSA is implementing a tightening of loan-to-income terms, we can expect house prices to face further downward pressures. As with all bank regulation, somebody has to pay and somebody has to be affected. When consumers have been borrowing too much too cheaply and house prices have been too high, new regulations will have to reverse this. We may then save more for our retirements, rather than relying on being able to sell our houses to somebody else to finance them.

DOI: 10.1177/0027950111411368

REFERENCES

Barrell, R. and Davis, E.P. (2005), 'Policy design and macroeconomic stability in Europe', National Institute Economic Review, 191, pp. 94-105.

Barrell, R., Davis, E.P., Fic, T., Holland, D., Kirby, S. and Liadze, I. (2009), 'Optimal regulation of bank capital and liquidity: how to calibrate new international standards', FSA Occasional Paper No 38.

Barrell, R., Davis, E., Fic, T. and Karim, D. (2011), 'Is there a link from bank size to risk taking?', National Institute Discussion Paper no. 367.

Barrell, R., Davis. E.P., Karim, D. and Liadze, I. (2010), 'Bank regulation, property prices and early warning systems for banking crises in OECD countries', Journal of Banking and Finance, 34, pp. 2255-64.

Barrell, R., Davis, E., Karim, D. and Liadze, I. (2010a), 'Calibrating macroprudential policy', National Institute Discussion Paper no. 354.

Barrell, R., Holland. D. and Karim. D. (2010). Tighter financial regulation and its impact on global growth', National Institute Economic Review, 213, pp. F39-44.

Barrell, R., Holland, D. and Liadze, I. (2010), 'Accounting for UK economic performance 1973-2009', National Institute Discussion Paper no. 359, forthcoming in a European Commission book on the UK economy.

Goodhart, C. (2009). 'The continuing muddles of monetary theory: a steadfast refusal to face facts', Economica. 76, pp. 821--30. Vickers Report (2011), Interim Report--Consultation on Reform Options, Independent Commission on Banking.

NOTES

(1) The terms person, agent and actor can all be used interchangeably to represent the legal agents in an economy, be they firms, household or other entities.

(2) Indeed the Institute macro team were told in 2004 that banking crises in the OECD were no longer possible, and hence we should not have any funding to research their causes and consequences.

(3) The expansion of the Icelandic banking system into the UK and the Netherlands in particular was just one example of the poor regulatory structure that had been set up in Europe, with the host regulators unable to investigate the capital adequacy of the companies (or their probity, which may have been more important), Barrell and Davis (2005) discussed these problems at length.

(4) The literature uses the phrase 'leverage ratio' although the older English term 'gearing ratio' may be better.

(5) Many industry comments on the effects of regulation can be seen as special pleading, especially as the benefits of bank activity are largely captured by banks and bank owners.

(6) Barrell, Holland and Karim (2010) demonstrate that house prices 'cause' borrowing and not the other way around.

Calibration of the Capital Framework
Capital requirements and buffers (all numbers in per
cent)

 Common equity
 (after Tier I Total
 deductions) capital capital

Minimum 4.5 6.0 8.0
Conservation buffer 2.5
Minimum + conservation buffer 7.0 8.5 10.5
Countercyclical buffer range 0-2.5