1
The Objectives of International Financial Regulation
At the beginning of the twenty-ļ¬rst century ļ¬nancial markets are more international than ever before. Capital markets were highly integrated before the First World War, with massive ļ¬ows of funds from developed to developing countries, but the degree of integration fell sharply during the next ļ¬fty years and capital movements were often highly controlled. Now, however, āglobalized capital markets are back, but with a difference: capital transactions seem to be mostly a richārich affair, a process of diversiļ¬cation ļ¬nance rather than development ļ¬nanceā.1 It is not true to say, as some do, that we live in a borderless world, but ļ¬nance certainly ļ¬ows more easily across borders than do goods or services.
The channels of ļ¬nancial intermediation have also changed. While only twenty years ago most business ļ¬owed through the balance sheets of banks or insurance companies, or through a limited range of investment funds usually dealing in products traded on regulated markets, the explosive increase in wealth held privately (partly as a result of greater dispersion of income) has led to the creation of a wide range of other investment vehicles, of which hedge funds and private equity are the most prominent, funded by high net worth investors and organized on an informal, largely unregulated basis.
New instruments have emerged which make it possible to transfer risk of all kinds on a far larger scale and in more complex ways, not solely through standardized exchange-traded derivatives, but through an almost inļ¬nite range of bespoke, over-the-counter arrangements: CDOs, synthetic CDOs and the like. In some cases banks hold them in off-balance sheet vehicles. While these instruments make it possible to lay off risk over a vastly greater range of risk bearers, which probably increases the systemās resilience, they also mean that when risks crystallize they may well have an impact in hitherto unfamiliar places, anywhere in the globe. They may make it easier to ride through small crises, but large ones will have many more dimensions of which we currently have no knowledge. This matters, because the last ten years or so during which these markets have evolved have also been remarkably benign in ļ¬nancial terms, but characterized by ready availability of credit on an unprecedented scale and consequently in some sectors by unprecedented levels of debt. Because of the structural changes which have taken place, the ways in which lenders and borrowers will react in the face of any major shock or prolonged downturn will test the ļ¬nancial architecture in ways for which the existing arrangements may be unprepared.
There are other new features of global capital markets. The most important is the growing dominance of a small number of huge institutions. A handful of ābulge bracketā investment banks dominate the major markets in corporate and sovereign debt and equity, most of them headquartered in the United States. Some commercial banks like Citigroup and HSBC have built signiļ¬cant market shares in many countriesā domestic markets. In a few large countries, such as Poland and Mexico, the majority of domestic banking is undertaken by subsidiaries of overseas institutions. The emergence of hostile cross-border bank takeovers, previously unknown, will accelerate that trend. Now even national stock exchanges, once seen as symbols of national virility, like the ļ¬ag carrier airline, are owned by foreign interests. Euronext, which includes the national exchanges of France, Belgium, the Netherlands and Portugal, as well as the UK futures exchange, has been bought by the New York Stock Exchange.
A small number of marketplaces, notably New York and London, increasingly dominate transactions in both cash and derivatives markets. Technology has allowed them to take on additional business from anywhere in the globe at very low cost. That, combined with the search for speed and liquidity, and a kind of āwinner takes allā phenomenon, is driving further geographical concentration.
But as concentration in the ļ¬nancial industry has grown, the global economy itself has become multipolar. Economic activity is no longer dominated by the United States and Europe, but spread much more broadly, including across markets once described as emerging. Thus there have been fundamental changes in both ļ¬nancial architecture and in the real economy, but no alignment between the two. It is not the purpose of this book to question the welfare beneļ¬ts that may or may not accrue from these developments, or to argue the case for or against free capital movement or ļ¬oating exchange rates. Our focus is on the challenges these developments pose for ļ¬nancial regulation, and on whether the global system of ļ¬nancial regulation, if it can be described as a system, is adequate to handle the consequences of these growing inter-dependencies. By ļ¬nancial regulation we principally mean the processes of authorizing, regulating and supervising ļ¬nancial institutions themselves, and the traded markets within which they operate. We comment only in passing on the macroeconomic dimensions of ļ¬nancial market oversight and on the implications for markets and economies of different tax regimes. And we say relatively little about the interaction between ļ¬nancial regulation and the rest of the legal system ā a subject which could justify a book in itself ā though we comment on some important links, such as with insolvency regimes.
Even with those important exclusions, the ļ¬eld is broad, and the diversity remarkable. Financial regulation encompasses a wide range of activities, from setting accounting standards, through bank capital requirements to insider dealer legislation, controls on money laundering and rules on investor protection. We traditionally think of three principal sub-sectors of ļ¬nance: banking, securities and insurance, but as our analysis will show, these sectors are increasingly interlinked, and the boundaries between them increasingly blurred.
The major questions we will seek to answer are:
- how well suited is the system of ļ¬nancial regulation to todayās capital markets?
- has it kept pace with the massive growth in cross-border activity and the changed patterns of intermediation?
- are changes needed to strengthen our defences against both ļ¬nancial instability and market abuse?
Many would argue that the answers to these questions are clear, and that the system is obviously inadequate. The collapse of Long Term Capital Management in 1998 and the Asian ļ¬nancial crisis at the end of the 1990s crystallized concerns about whether the regulatory system, pieced together in an ad-hoc manner over the previous two decades, was able to address the challenges of globalization. Some argued then for the creation of a world ļ¬nancial authority with wide ranging powers to handle cross-border regulatory issues.2 After some debate these calls were rejected by the G7 Finance Ministers, in favour of more modest changes, notably the establishment of the Financial Stability Forum as a co-ordinating mechanism between existing structures, and an increased focus by the IMF and the World Bank on the quality of ļ¬nancial regulation in member countries.
Since then further modest improvements have been made, but recent market developments have once again generated questions about their adequacy. As cross-border stock exchanges are created, how will they be overseen? Can the regulators work together effectively to supervise a consolidated system of exchanges? Have hedge funds and, more recently, private equity funds created threats to ļ¬nancial stability and to the integrity of traded markets which the system is not designed to address? How can the rapid growth of Islamic ļ¬nance, with its rejection of the traditional concept of interest, be accommodated in a system designed well before it began to emerge as a signiļ¬cant market phenomenon? Speciļ¬cally in the European Union, there are those who argue that a single integrated ļ¬nancial market, especially those parts of it with a single currency, necessarily requires a creation of a single regulator.
How powerful are these arguments? Is the regulation of the global ļ¬nancial system still ļ¬t for purpose, if it ever was? If it is under strain, what changes might realistically be made to improve its robustness?
It is the aim of this book to suggest answers to these questions. Before doing so, however, it is necessary to explain why we seek to regulate ļ¬nancial markets and ļ¬nancial institutions. What are the regulators trying to achieve, and what expectations can we realistically have of them?
Why regulate ļ¬nancial markets?
One might expect to ļ¬nd a simple answer to the question, but in fact this is heavily contested intellectual territory. The basic economic rationale is straight forward. There can be externalities generated by ļ¬nancial market activity, which are not easily capable of being addressed by private sector actors. But the prime deļ¬nition of those externalities, and the nature of the interventions they justify is the subject of constant debate. Even if we exclude the extremes of the argument ā those who argue for rigid state control of the ļ¬nancial sector, and those who prefer no regulatory interference whatsoever ā there are many differences of opinion on the degree of regulation. When the British system of ļ¬nancial regulation was overhauled in the Financial Services and Markets Act 2000, there were lengthy debates about the intensity of intervention that should be allowed, and indeed about the borderline between statutory and self-regulation in markets. Similarly, since the passage of the Sarbanesā Oxley Act in the United States, which greatly extended the reach of regulation in the accounting and auditing ļ¬eld, there has been an intense debate about whether regulation has, in some sense, āgone too farā and should be reined back. For the ļ¬rst time there are powerful voices in the US, including Treasury Secretary Hank Paulson, arguing that regulation is too detailed and intrusive, with damaging consequences for Americaās capital markets.3 In his memoir, āThe Age of Turbulenceā, Alan Greenspan recommends that āRegulation approved in a crisis must subsequently be ļ¬ne-tunedā, identifying the Sarbanesā Oxley Act as ātodayās prime candidate for revisionā.4
The ebb and ļ¬ow of this debate is familiar. After every ļ¬nancial crisis there are calls for regulation to be ratcheted up. A few years later a reaction begins to emerge, as the costs become evident, and the beneļ¬ts are less so. At present, the pendulum is swinging violently. Regulated ļ¬rms and markets think they are over-controlled, while many politicians, by contrast, think there is too much scope to avoid regulation, and too much evidence of investor detriment and unjustiļ¬ed enrichment on the part of ļ¬nancial sector professionals. The recent sub-prime mortgage crisis in the United States, which spread to other markets, was a test case for these arguments.
There is no doubt that regulation imposes high costs on ļ¬nancial institutions and markets, costs which are ultimately passed through to the end user. It is also clear that āexcessiveā regulation can damage the functioning of ļ¬nancial markets and reduce their economic utility. In any system of regulation, there is a balance to be struck between safety and soundness on the one hand and risk taking on the other. The incidence of bank and insurance company failure might be signiļ¬cantly reduced by tighter capital requirements, but the returns available to depositors and policy holders will be correspondingly reduced. The terms on which investments can be offered to the public can be restricted, but the opportunity to diversify into more proļ¬table assets is also correspondingly constrained. In recent years hedge funds have, as a class, outperformed regulated collective investment schemes, yet retail investors have typically not been able to access them directly.
Against that background, it is important to have a clear understanding of the rationale for regulation, against which proposals for an expansion (or, less , contraction) of the regime can be assessed. In principle, regulatory intervention should only be justiļ¬ed where the beneļ¬ts clearly exceed the costs imposed. But cost beneļ¬t assessments of regulation are still in their infancy, in spite of much effort by the UK FSA and others. The costs are usually easier to quantify than the beneļ¬ts, but the beneļ¬ts appeal more to politicians. (A useful expanded discussion of the economic rationale for ļ¬nancial regulation can be found in Llewellyn.)5
Prudential standards
There are two principal strands to the rationale for regulating some ļ¬nancial markets, businesses and trans actions.
The ļ¬rst relates to the problem of systemic risk. There is persuasive evidence that a stable ļ¬nancial system provides a favourable environment for efļ¬cient resource allocation and therefore promotes economic growth. However, experience shows that, left to themselves, ļ¬nancial systems are prone to bouts of instability and contagion. A World Bank study shows that there were 112 systemic banking crises in 93 countries between the late 1970s and the end of the twentieth century.6Another study by Eichengreen and Bordo argues that ārelative to the pre-1914 era of ļ¬nancial globalization, crises are twice as prevalent todayā.7And the incidence of ļ¬nancial crisis has tended to rise as ļ¬nancial markets have become more liberalized and more international.
The cost of crisis does not fall only on the banks themselves, or their shareholders and depositors. In the last three decades of the last century there were 10 countries, from Mexico to Israel, where the ļ¬scal cost of bailing out the banking system was more than 10 per cent of GDP, a cost which was borne by taxpayers. So if we can ļ¬nd ways of reducing the incidence of systemic crisis without excessively constraining the functionality of the markets, there is a powerful case for adopting them.
The traditional systemic risk argument for the prudential supervision of banks starts from the premise that banks, through their role in maturity transformation and the provision of liquidity, occupy a special position in the ļ¬nancial system. They sit at the centre of the payments network and the failure of one bank can bring about a domino effect on others. The potential externalities of such a failure cannot easily be internalized. Walter Bagehotās Lombard Street8remains the classic exposition of this argument. So there may be a case in certain circumstances for rescuing a failing institution in the interests of minimizing the costs which may fall on others, rather than in the interests of the bankās own depositors and shareholders.
There is often a very difļ¬cult decision to make about whether an individual bank failure is likely to be systemic or not. The Bank of Englandās view in relation to Barings, for example, was that it would not be, and it therefore declined to extend support. In the event Barings was bought by another bank (ING) for one pound sterling. Most people now think that the Bank of Englandās judgement was correct. There is also a difļ¬cult question as to whether this systemic argument now applies to other, non-banking institutions. The case of Long Term Capital Management, whose failure in 1998 did seem to threaten systemic consequences, is suggestive in this context. In the event, the New York Federal Reserve did not provide ļ¬nancial support, though it convened a group of investment banks which did so. (The then President of the Fed, Bill McDonough, always maintains that the cost to the Fed was only a plate of doughnuts.)
Nonetheless, it is clear that the failure of a large investment bank, or a large insurance company, would have widespread ramiļ¬cations for the ļ¬nancial system as a whole. The British Memorandum of Understanding between the Bank of England, the Treasury and the Financial Services Authority9creates a framework in which the potential systemic consequences of non-bank failures can be assessed, but it is factually the case that no bail out of a non-bank has yet been seen to be justiļ¬ed in the UK under the new regime, and it is hard to think of examples elsewhere, either.
But this amounts to a justiļ¬cation for a lender of last resort function, typically held by the central bank, to supply liquidity and conceivably solvency support, and not necessarily for the whole apparatus of prudential supervision carried out by a regulatory authority, whether a central bank or some other agency. It would be possible to operate a completely hands-off approach from year to year, meeting banks only when they run into serious trouble. So why do more, why supervise on a continuing basis?
Here there are two main arguments. The ļ¬rst is about externalities. In running their businesses, we may presume that bank managers and shareholders do take account of the risk of loss to themselves if their bank fails, in terms of lost jobs, lost reputation and lost shareholder value (which should never be underpinned by the lender of last resort). The Bank of England has made it explicit that in the UK the price of public support will typically be that shareholders will lose their investment and top management their jobs. But banks do not necessarily take account of the potential external cost to the economy of their failure. So they will tend to take greater risks than they would do if there was a market for this risk. Supervision aims to counteract potentially excessive risk taking by requiring banks to hold larger reserves than they might otherwise do, and to conduct their business with more careful attention to ...