Inhabitants of a community often agree that "something must be done." Thus after the BCCI scandal, the various pensions scandals, the Barings scandal, the Morgan Grenfell scandal, the NatWest derivatives scandal-and minor scandals obscured by more distinguished ones-the City of London's consensus has been that something must be done about financial regulation.
The government not only said that something must be done, it launched the Financial Services Authority (FSA) and made clear that this will not much resemble past practice. Such activism erodes consensus, revealing that the City disagrees considerably about what ought to be done. Indeed, some of its inhabitants think that the right thing to do about regulation is to abolish it. A new financial regulatory reform bill is being debated and should be operational by the end of 1999. It may not have an easy passage.
The government's chief idea has been to release the Bank of England from political control, simultaneously relieving it of nearly all the regulatory authority for banking which it had accumulated since 1720, when it crushed the Sword Blade Bank. Most of this authority will pass to the new FSA-which may become even tougher than the Securities and Exchange Commission (SEC) which has terrorised Wall Street since 1934.
The Bank's famous gain from this is undivided responsibility for inflation control: important-but reminiscent of those contests where the second prize is two weeks in Middlesborough. Less famously, the Bank will have a role in preserving financial stability, through the provision of sovereign credit against systemic risk. In other words, it will be allowed to rescue concerns considered (unlike Barings) too big to go bust. The FSA will have powers, but the Bank will have money; and only money revives a sick bank.
Part of Westminster's reformist ambition is easily explained-as is similar action in the US, where the SEC is considering changes deeper than any since its foundation. Both governments are aware that information technology is altering the financial playing-field. Both wish to set down some new rules before referees are forgotten entirely.
Both governments also expect to privatise more of the responsibility for pensions, saving and social insurance. Whereas in the early days of the century most people distrusted paper money-let alone banks-Anglo-Saxons now extend a wary trust to the financial services industry. But for universal privatisation to be politically viable, trust must be universal. If this is to happen, predators such as Robert Maxwell must be eliminated. Also, the severe stock market volatility we have seen in recent weeks may need to be calmed.
why do we need reform?
It is a long time since anyone supposed that the old system-product of the 1986 Financial Services Act-might be adequate. Its main characteristic was acronyms to boggle even a military mind. Distinguishing the Securities and Investments Board (SIB), the Securities and Futures Authority (SFA) and the Serious Fraud Office (SFO) could be tricky enough; but what of, say, FISMOU's effect on the ability of FIMBRA to help LAUTRO and PIA in updating the FFIN on changes in their respective SORPs? In theory, an array of self-regulating organisations would provide independent expertise from all the tribes of the City (brokers, fund managers and so on) under the statutory leadership of the SIB. In practice, said Andrew Large in his final report as chairman of the SIB, the old system created not independence but "hostility." Serious wrongs were ignored while turf warfare raged among the acronyms. Large said: "The most striking example of delays occasioned by the two-tier system has been in tackling the past mis-selling of personal pensions."
It was the discovery that death had made more progress with this issue than the internal mechanisms of the pensions industry which turned the government against self-regulatory inertia. It wants, right now, an activist organisation with real powers. The Bank of England never really recommended itself for the job. It has never cared much for consumer issues-and the Bank is thought, perhaps unfairly, to have spent too much time explaining why Barings and BCCI were not its fault, and too little time explaining how scandals might have been averted. The best candidate for an upgrade was the SIB, which, at least under the Large regime, had expressed discontent with the status quo. Retiring, and passing command to Howard Davies, Large described an opportunity "to redesign our system for regulation and supervision of financial business in ways which can benefit customers and the industry and will provide continuing and justified confidence in the UK as a global financial services centre."
But it is this assumption-that what is good for British customers is good for a global financial centre-and the prospect of an organisation invigilating everything from mortgage finance to derivatives, which raises hackles in large parts of the City. As Anthony Hilton, City editor of the Evening Standard, puts it, imposts upon individuals by flaky pension salesmen are clearly wrong-but an international bank "legging over a major multinational in a complex derivatives trade" is not the same at all. Most experts attribute London's global standing to its tolerance of consenting financial adults who wish to perform sophisticated acts on each other, untroubled by investigators concerned with the suburban saver's values. Harold Rose of the London Business School fears the FSA "will turn out to take as its mission the protection of customers to such an extent as to weaken London."
Arguments from this quarter suggest that the FSA must be resisted. More congenial is the "Twin Peaks" plan, publicised by the Centre for the Study of Financial Innovation, which suggests one regulatory regime for wholesale products-traded between professionals-and another for retail offerings made to the public. The Twin Peaks plan rests on two propositions: that the City, as a global financial centre, makes an important, even indispensable, contribution to the economy; and that financial professionals are competent predators who can be left to look after themselves. Both claims may be true, but they deserve some critical attention.
how much is the city worth?
Britain's financial services sector is unquestionably important, making up about one fifth of all economic activity. This is similar in Sweden, France or any mature economy. The British peculiarity is that the City's work in managing the nation's savings is thought to be subordinate to its international role. A flood of assorted international finance traverses London, but we know surprisingly little about what this does for the overall added value of the economy. Britain has one of the world's better sets of economic statistics, but they do not provide a firm basis for a value-added analysis of the City's internationally traded wholesale services.
The best recent attempt, by Robert Brealey of the London Business School, was published in the City Research Project in 1993. He traced net revenues worth ?11.1 billion for 1991. As an overall measure of wholesale services sold to non-British users, this probably was a self-cancelling overestimate and underestimate. It included derivative exchanges, although these are mostly British-bought, and it excluded over-the-counter derivatives because there was no useful data.
The larger doubt about the figures relates to trading activities. Losses are often substantial. Brealey's, and other studies, suggest that, in equity markets, the bulk of dealer revenues are absorbed by losses, which become gains to a small number of well informed traders. Such effects, Brealey thought, "may reduce the aggregate net revenues to about ?8 billion. If intermediate inputs (support services such as computers) are of similar importance across banking, finance and insurance as a whole... then real value added by these (international wholesale) activities would be somewhat under ?4 billion."
What emerges is a rough idea of the relative magnitude of different activities. But still, against a value added figure for the whole economy of ?500 billion, the contribution of the City's international trade may not be as important as is often assumed. Bad behaviour may be acceptable in a goose that lays golden eggs-but shouldn't we check their weight?
Brealey thought that as a high proportion of the world's inter-bank business-a zero-sum game in aggregate-is done in London, the likelihood is that London gets paid for this service; but he could not find out how much. There is evidence to suggest that London's share of international fund management is highly rewarding; but again, accurate measurement is impossible in almost all market sectors. Certainly there are large rewards to individuals, but that says little about net benefits to the economy. "It is possible," says Brealey, pointing to cases such as Lloyd's, "that the growth of intra-market activity has simply added to the cost base and therefore weakened the market's competitive position." Brealey argues that the statistical shortcomings of the financial sector should be reviewed "by a committee under the aegis of the Bank of England." Little has happened. Might this be a task for the FSA?
Whatever the truth about the city's earnings, there is a seductively coherent theory, popular among finan-cial professionals, which claims that all financial regulation and supervision is now self-defeating. Andrew Smithers, who used to run Mercury Asset Management, thinks that the best thing for the regulatory structure is abolition.
Regulation and supervision, he says, exist only to prevent bank failure. But because the world is too unpredictable to allow that, the practical effect is a government undertaking that no significant bank will go bust. Such undertakings are ineluctable vectors of market distortion and moral hazard.
This claim has some force. Suppose a fund manager punting in biotechnology stocks decides to insure himself via a neat option from the investment-banking subsidiary of a big clearing bank. Do fund managers look at this option with scepticism if it appears to offer something for nothing? Do they reflect on whether the bank has fully assessed the downside? No, says Smithers. They just think that payout is safe-however absurd-because the government bails out clearing banks. The market's subtle mechanism is distorted in favour of a rational recklessness.
regulation and financial science
The assumption is that professional judgement cannot be improved on-and can only be degraded by official action. Like the Twin Peaks argument, it states that professionals know best, but must be restrained from using their skills on common people. Like other elites, financial professionals see themselves as epigoni of stability and judgement; it is the small investors who should be protected, for the greater good, against their own infirmities. Pundits fret over the 40 per cent of US households which now have stock market assets. They suggest that inexperience will drive many to Gadarene reaction when the great bull market comes to an end.
But it can be argued that financial d?b?cles owe more to expert than amateur credulity. In 1929 it was not ordinary US investors-there were hardly any-who devised leverage mechanisms such as the Goldman Sachs Trading Corporation, which magnified upward movement of the market and had horrid reciprocal effects when it fell. It was not the clerks and doormen who decided that it was wise for New York banks to extend huge loans to shaky Latin American autocracies. The pension-selling scandal in Britain began when government and industry conspired to ignore the obvious difficulty of turning commission-driven salesmen into objective advisers.
But whatever history says, there is a new reason why regulation theorists such as Michael Taylor, author of the Twin Peaks proposal, want to leave the wholesale battles to professional judgement. "Will regulators," asks Taylor, "know any better?" The defining technique of modern, science-based, finance (risk management via the use of derivatives) has become so complex, and its practitioners so expensive, that regulators may not be competent to monitor technical process. Market operators have achieved a monopoly of relevant knowledge.
But the history of other science-based industries (pharmaceuticals, aeronautics or nuclear engineering) does not suggest that activities should be lightly regulated simply because they are complex. Before leaving the scientific financiers to their own devices, we should try to find out what they are up to.
The-scarcely modest-claim of financial engineering is that in qualified hands, the derivative of an investment can be used to eliminate the risk of holding that investment. It is also claimed that the extension of this notion-via computer power-allows large and complex portfolios to be "hedged" against risk to any precisely desired extent.
Put simply, the derivative of an underlying investment is an option to trade in that same investment at some future time. Risk in the underlying manifests itself in a price change, and when the current price of a class of shares declines, the price of an option over similar shares (the derivative) will decline also.
When prices fall, profits are available to those who have "sold short"-those who have a contract to sell something they have not yet bought. The profit from selling a derivative short can compensate for loss of value in the underlying. (And vice versa, when prices rise). So if a reliable link can be found between the movements of the underlying and of its derivative, we have the possibility of "dynamic hedging," the risks of today's trading neatly damped out by continuous adjustments of the future position.
Differential equations are required to manage relationships of this kind. These were not available until the 17th century when Newton and Leibniz devised the calculus. But neither could have hacked it as a derivative trader. Their classical calculus handled continuously varying quantities-but quantities clear-cut at any moment. A share price is not clear-cut. It is blurred by volatility.
The final step into modern risk-management was taken in the early 1970s by Fischer Black and Myron S Scholes, who adapted to option trading the mathematics of stochastic calculus. Developed by physicists to handle the collective behaviour of individually volatile particles, this assumes that the extent of volatility ("variance") can be recorded and a mean established. Volatility is then treated as a random fluctuation-around the mean and within the variance-upon which the rules of normal probability can be applied, and the equations become classically docile.
There have been many additions to the work of Black and Scholes, but the basics of stochastic calculus remain unchanged-and as essential to modern finance as the structural equations stabilising the skyscrapers in which it dwells. The presence of a mathematically elegant theory has turned option-trading from a colourful sidestreet into a thoroughfare of finance; today a fund launched by sober Scottish financiers may start life as a package of derivatives, with actual investments coming along at leisure.
The application of stochastic calculus to finance is eased by computer software-some available on the internet-such as JP Morgan's RiskMetrics and CreditMetrics. It works. It enables traders to deal in volatile non-linear quantities and still have numbers to put into the books at the day's close. It enables the construction of value-at-risk models, which increase banking profits by economising in capital reserves. No orthodox calculus can offer such advantages.
It is not so clear whether this profitable trading technique says anything useful about risk-in spite of confident front-office remarks. The trouble is that stock markets are more volatile than stochastic calculus expects. For a period, prices will stay roughly within the path traced by mean and variance. But in a normal stochastic process, departures from the mean which exceed the historic variance are extremely rare. In financial markets, such "jumps" occur often.
How often? Well, that is the problem of prediction. If the New York and London stock exchanges were normal stochastic systems, their volatility would stay for several thousand years inside a band about 3 per cent each side of the trend line. Since the second world war there have been numerous occasions on which both exchanges moved much more sharply-notably, in 1987 and 1994. Considerable efforts have been made to apply "abnormal" probability distributions, but so far they make the mathematics unworkable.
No competent person argues that stochastic calculus is used in finance because it fully describes market behaviour. It is used because it is mathematically handy-an acceptable simplification. Yet scepticism is widespread, if not much advertised.
Risk managers are adept at rebuilding hedged positions after market fluctuations; and at subjecting their portfolios to simulated stress. But this is not a settled science, where professional judgements compete within well understood parameters. It is not, therefore, an area in which we can easily say which risks have "systemic" potential and which have not.
Those who think that regulation should be inversely related to complexity argue that "no single person is capable of matching the knowledge and understanding of the whole market." But some problems of high-tech finance are complex in the same way that a new piece of aeronautical engineering is complex. The Boeing 747 design relied on radical ideas about the strength of airframes; their validity was decided by scientific regulation before market opinion was invited.
The hands-off theory can be starkly Darwinian. Let a few jumbos, or banks, go down. That will weed out unsound practitioners and make folk pay more attention to aerodynamics and stochastic calculus. But this turns the marketplace into a laboratory-without making allowance for the powers which technology and distribution have brought forth.
strong but subtle regulation
I favour a strong, coherent regulatory body-although it should behave modestly until several areas of ignorance have been reduced. But there are many respectable arguments which support the Twin Peaks and other plans. In practice, the FSA will have to make case-by-case judgements about the boundary between systemic and non-systemic dangers. In a period of institutional fluidity and rapid technical change, definitions are likely to be outdated as soon as they are made.
Perhaps the strongest practical argument for the FSA-unwelcome as it might be to some influential tribes-is that it will have a better chance of paying for the expertise it requires. In banking and finance the gap between public regulator and private practitioner has always been wide. Today, it threatens to exclude financial authorities from any access to the techniques they attempt to regulate. Some consider this a desirable condition-it will certainly become a reality, unless the government applies some serious thought to the matter. In no other area touching its security-medicine, military communications or power supply-would a democratic state, however privatised, surrender its capacity to understand process. Should finance be any different because it has unique purchasing power? It now uses-like other industries before it-some of the tools of science, about which 300 years of history teach us certain lessons. One is that the effects of science are unpredictable; they are only reasonably safe when labelled Public Property and Handle With Care.
