Traders- Risk, Decisions and Management

70+ DVD’s FOR SALE & EXCHANGE www. traders-software. com www. forex-warez. com www. trading-software-collection. com www. tradestation-download-free. com Contacts [email  protected] com [email  protected] ru Skype: andreybbrv TRADERS This page intentionally left blank TRADERS Risks, Decisions, and Management in Financial Markets Mark Fenton-O’Creevy Nigel Nicholson Emma Soane Paul Willman 1 Great Clarendon Street, Oxford ox2 6dp Oxford University Press is a department of the University of Oxford.It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan South Korea Poland Portugal Singapore Switzerland Thailand Turkey Ukraine Vietnam Oxford is a r egistered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc. New York  © Oxford University Press 2005 The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2005 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization.Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above. You must not circulate this book in any other binding or cover and you must impose this same condition on any acquirer. British Library Cataloguing in Publication Data Data available Library of Congress Catalo ging in Publication Data Data available ISBN 0–19–926948–3 3 5 7 9 10 8 6 4 2 Typeset by Newgen Imaging Systems (P) Ltd. , Chennai, India Printed in Great Britain on acid-free paper by Biddles Ltd. King’s Lynn, Norfolk Acknowledgements We gratefully acknowledge the help of the investment banks which cooperated in this research and provided ? nancial support, and the Economic and Social Research Council which provided funding as part of the Risk and Human Behaviour Programme (grant number L211252056). We are especially grateful to the traders and managers who gave us their time and shared their understanding. This page intentionally left blank Contents List of Figures List of Tables viii ix 1INTRODUCTION Traders, Markets, and Social Science 1 10 2 THE GROWTH OF FINANCIAL MARKETS AND THE ROLE OF TRADERS 3 ECONOMIC, PSYCHOLOGICAL, AND SOCIAL EXPLANATIONS OF MARKET BEHAVIOUR 4 TRADERS AND THEIR THEORIES 5 A FRAMEWORK FOR UNDERSTANDING TRADER PSYCHOLOGY 6 RISK TAKERS Pro? ling Traders 28 51 74 110 145 178 197 212 221 237 7 8 9 10 BECOMING A TRADER MANAGING TRADERS CONCLUSIONS APPENDIX The Study References Index List of Figures 2. 1 2. 2 2. 3 3. 1 3. 2 4. 1 4. 2 5. 1 5. 2 6. 1 6. 2 6. 3 6. 4Post-war UK equity market growth Post-war US equity market growth Global growth in OTC derivatives Expected utility theory Prospect theory The relationship between risk and return Idealized trader risk pro? les RAT Screenshot Distribution of traders’ illusion of control scores A model of individual risk behaviour Comparisons of personality scores by occupational group Risk propensity, risks taken—now and past Comparisons of risk propensity scores by occupational group 7. 1 Career mobility to date 7. 2 Likelihood of a career change in the next 5 years 8. Introducing incentive and monitoring effects to prospect theory description of risk behaviour 14 15 16 40 41 55 63 104 106 117 132 136 138 174 175 194 List of Tables 6. 1 Risk taking index 6. 2 Personality facets—signi? cant differences between occupational groups 6. 3 Relationships between RTI and Big Five personality factors 6. 4 Relationships between RTI and Big Five personality subscales 6. 5 Regression on total remuneration 8. 1 Controls and incentives associated with framing effects—empirical ? ndings 10. A1 Investment bank sample pro? le 10.A2 Personality and risk propensity sample pro? le 10. A3 Frequencies of self-ratings of performance 131 133 138 140 143 193 214 215 218 This page intentionally left blank Chapter 1 INTRODUCTION Traders, Markets, and Social Science I grew up in a small town in Florida and none of this stuff really exists like stocks and bonds and things like that. No one I ever knew growing up did this sort of thing and to me it all seems like a fantasy world sometimes and it’s very abstract. You know, I explain to my mother what I do and I can’t, you can’t put it into words, it just doesn’t make any sense. You can read also Portfolio Management QuizzesI am so removed from the daily life of the average person that I think at some point this has got to come to an end. Whether I really believe that or not I don’t know but in my head I kind of think this is all fantasy land and one day I’m going to wake up and I’m going to say I had the most amazing dream, I’ve been working on some place called Wall Street, that paid me lots of money and I just sat around and looked at computers all day and put these pieces together and everything worked out and it was all a lot of fun. So in my mind that’s kind of what I think.Derivatives Trader, ? rm B We live in a world that is shaped by ? nancial markets and we are all profoundly affected by their operation. Our employment prospects, Introduction our ? nancial security, our pensions, the stability of political systems and nature of the society we live in are all greatly in? uenced by the operation of these markets. Th e role and importance of international ?nancial markets and the traders who inhabit them has grown dramatically in the past few decades. The level of ? nancial ? ows in these markets can rise to quite staggering levels.For example, in the day before the setting of entry exchange rates to the Euro, trades in currencies entering the Euro totalled about ten times World gross domestic product (GDP). At any one time, outstanding derivatives contracts have a total value of around four times World GDP. Professional traders ? gure prominently in media accounts of the workings of ? nancial markets and the economy. Television news bulletins on the economy or stock market frequently include interviews with senior traders, or footage of a trading ? oor. Stories about ‘rogue’ traders are big news.The decisions of individual traders are often seen as having the potential to move markets and affect national economies. Yet, the role of the professional trader is largely absent from mai nstream ? nancial economic accounts of markets. Professional traders, we argue, inhabit a borderland in markets where some of the orthodox assumptions of ef? cient, instantaneously adjusting prices break-down. They are often well placed to exploit market imperfections, by virtue of lower transaction costs, access to privileged information, critical mass, or proprietary knowledge and models.However, at the same time, they work in a fast-moving landscape of noise, rumour, unreliable information, and uncertainty. Thus, it is often dif? cult to tell whether an opportunity is real or illusory. This is a book about professional traders in this noisy borderland: what they do, the kind of people they are, how they perceive the world they inhabit, how they make decisions and take risks. This is also a book about how traders are managed and the institutions they inhabit: ? rms, markets, cultures, and theories of how the world works. Our approach to writing this book is explicitly interdiscipl inary.We draw on psychology, sociology, and economics in order to illuminate the work of traders and their world. Our focus is traders and the ? rms they work in. It is not the purpose of this book to mount an extensive critique of the dominant rational–economic account of ? nancial markets, nor 2 Introduction is ‘markets’ our central focus. We are concerned principally with understanding the world of the professional trader. However, we do believe our work is relevant to an understanding of ? nancial markets. First, in order to understand the role and work of the trader, it is important to understand that the neoclassical paradigm of ef? ient markets and rational pricing breaks down at the margins and that professional traders both bene? t from and contribute to this departure from orthodox ? nancial economic theory. Second, the ef? cient markets paradigm rests on the assumption that in the absence of uniformly rational investors, there is a suf? cient group of rational investors who are able to drive out pricing anomalies through arbitrage. 1 Professional traders in investment banks seem good candidates to play this role. Hence, the evidence that we present on the ways in which traders can deviate signi? antly from rational– economic norms of behaviour may be fruitful in helping to explain market phenomena. 1. 1 Our Work and How It Informs the Book This book is based on a study of traders in ? nancial instruments in four large investment banks operating in the City of London. Over the course of 1997 and 1998, we carried out interviews with 118 traders and trader managers in four large City of London investment banks and collected qualitative and quantitative data on their roles, behaviour, performance, and psychological pro? les. We carried out followup interviews in 2002. We use detailed quotations from the interviews throughout the book. Where we use these quotes they are presented verbatim. We had three main concerns. First, we came to the study with a strong interest in decisionmaking and risk. While all business is concerned to some extent with risk, investment banks and ? nancial traders are almost unique in the extent to which their work is founded on the management of risk and the extent to which they must make decisions about risk. Second, in the vast literature on ? nancial markets relatively little attention has been paid to the role of ? ance professionals in these markets and we wanted to redress this. 3 Introduction Third, we observed that the large literature on markets and the (somewhat slimmer) literature on traders are marked by very different approaches and paradigms in three branches of the social sciences: economics, sociology, and cognitive and social psychology. We wanted to bring together the insights of these different disciplines. Throughout the book we draw both on the data we gathered in this study and on the insights of prior research and literature in ? nancial economics, psychol ogy, and the sociology of markets.We turn now to those literatures. 1. 2 Traders in the Social Science Literature Neoclassical Financial Economics Financial economics is a relatively young discipline. The origins of modern (neoclassical) ? nancial economics are often located in the early 1950s in the work by Markowitz (1952) on portfolio theory. During this period, ? nance moved from a concern with describing the activities of actors in ? nancial markets to the construction of parsimonious models of markets founded on assumptions of rational investor behaviour. The central organizing idea of neoclassical ? nancial economics is the ef? ient markets hypothesis, which holds that price changes are essentially a random walk. All new information relevant to prices is incorporated into prices instantaneously (Fama, 1970). This central proposition and much of the theory which springs from it is founded on the idea that any asset which is not ‘rationally priced’ provides opportu nities for pro? t, which will be instantly taken up and cause prices to converge to the ‘rational’ level (i. e. arbitrage). This assumption is both illustrated and lampooned in the ? nance joke about two ef? cient market theorists who pass a $50 bill lying in the street.They leave it untouched and congratulate each other on realizing that if it presented an opportunity for pro? t someone else would have picked it up already. Even the strongest proponents of the ef? cient markets hypothesis do not claim that it represents a good description of the behaviour of individuals in markets. Rather it is claimed to be a good enough description, which should be judged on its predictions rather than its assumptions. 4 Introduction Fama (1970), who set out an early comprehensive account of the ef? cient markets paradigm, has more recently suggested that: Like all models, market ef? iency (the hypothesis that prices fully re? ect available information) is a faulty description of pri ce formation. Following the standard scienti? c rule, however, market ef? ciency can only be replaced by a better speci? c model of price formation, itself potentially rejectable by empirical tests. (Fama, 1998: 284) The ? nance professional is largely absent from orthodox ? nancial economic accounts of markets. The assumption of ef? cient markets, with no privileged information held by any investor, leaves little room for an account of how professional investors might make better than market returns.However, more recently, there has been an increasing interest within ? nancial economics in explaining empirically observed departures from the predictions of the ef? cient markets hypothesis and rational–economic pricing theories. Many of these fall in the emerging ? eld of behavioural ? nance. What has allowed consideration of the role different types of investor might play in markets is the growing recognition that perfectly ef? cient markets are not an automatic consequence o f the existence of arbitragers: an idea that has been captured eloquently by Lee (2001: 284).I submit that moving from the mechanics of arbitrage to the [ef? cient markets hypothesis] involves an enormous leap of faith. It is akin to believing that the ocean is ? at, simply because we have observed the forces of gravity at work on a glass of water. No one questions the effect of gravity, or the fact that water is always seeking its own level. But it is a stretch to infer from this observation that oceans should look like millponds on a still summer night. If oceans were ? at, how do we explain predictable patterns, such as tides and currents? How can we account for the existence of waves, and of surfers?More to the point, if we are in the business of training surfers, does it make sense to begin by assuming that waves, in theory, do not exist? A more measured, and more descriptive, statement is that the ocean is constantly trying to become ? at. In reality, market prices are buffete d by a continuous ? ow of information, or rumours and innuendos disguised as information. Individuals reacting to these signals, or pseudo-signals, cannot fully calibrate the extent to which their own signal is already 5 Introduction re? ected in price. Prices move as they trade on the basis of their imperfect informational endowments.Eventually, through trial and error, the aggregation process is completed and prices adjust to fully reveal the impact of a particular signal. But by that time, many new signals have arrived, causing new turbulence. As a result, the ocean is in a constant state of restlessness. The market is in a continuous state of adjustment. Lee argues that the relationship between inef? cient pricing and arbitragers may be like predator–prey dynamics. In equilibrium there must be both predator and prey. Similarly, in equilibrium there will be both arbitragers and arbitrage opportunities in the market place.There is another important way in which ? nancial ma rkets are widely accepted as departing from the ef? cient markets paradigm. Investors trade much more often than the theory suggests they should. More recent ? nancial economics accounts often distinguish two types of investors: ‘noise traders’ and ‘smart traders’ (a recent example is Daniel, Hirshleifer, and Teoh, 2002). Noise trading is trading on the basis of information that is either irrelevant to price or has already been discounted by the market. ‘Smart’ traders are those who act rationally, trading only on the basis of genuinely new and relevant information.This distinction is sometimes taken to map on to the difference between naive investors and trained professional investors (e. g. Ross, 1999; Shapira and Venezia, 2001). Behavioural Finance There has been increasing interest within the ? eld of ? nancial economics in using what is known about persistent biases in human cognition to explain departures of market behaviour from the pred ictions of ef? cient markets theory. Collectively known as behavioural ? nance, these models and empirical studies generally seek to explain market behaviour that departs from the predictions of orthodox ? ancial economics by reference to systematic cognitive bias among investors or important subgroups of investors. 3 Behavioural ? nance draws heavily on work from behavioural decision-making, a branch of psychology concerned with modelling human decision-making processes. While, in the main, this literature does not distinguish between professional traders and other investors, there have been 6 Introduction some attempts to compare the susceptibility to biases of ? nance professionals to that of the wider population.For example, Shapira and Venezia (2001) found professional brokers less susceptible than independent investors to one common bias, the disposition effect (a bias towards selling stocks more readily to realize gains than to realize losses), although they were not immune t o the bias. In an experimental study Anderson and Sunder (1995) compared the behaviour of laboratory markets populated by experienced commodity and stock traders with the behaviour of markets populated by MBA student traders. They found the amount of trading experience to be an important determinant of how well market outcomes approximated (ef? ient market) equilibrium predictions. Student traders’ markets exhibited departures from rational prices founded in common cognitive biases while bias levels in markets with experienced traders were substantially lower. However, as we explore in Chapter 5, our own research offers evidence that professional traders are just as susceptible as other groups to some forms of bias, with important consequences for their behaviour and performance. Sociology of Markets Sociologists interested in markets have paid rather more attention to the role of professionals than have ? ancial economists. Unlike ? nancial economists who take markets to be naturally occurring, sociologists tend to stress the ‘social embeddedness’ of markets and the ways in which they are sustained as social institutions through active intervention and regulation. One important strand of work is concerned with the social networks that operate within markets and in particular the ways in which professionals within markets act through these social networks and exercise informal sanctions over participants departing from accepted norms of behaviour (e. g.Baker, 1984a; Abola? a, 1996). Research by ? nancial economists also demonstrates the signi? cant effect the detailed structure and organisation of markets4 can have on the ? ow of information, liquidity, and prices (e. g. Amihud, Mendelson, and Lauterback, 1997; Lipson, 2003). Others have been concerned with the nature and consequences of ? nancial economic theory. Traders, from this perspective, do not simply inhabit markets; they enact them. That is, the beliefs they hold 7 Introduction ab out the nature of markets affect those markets in non-trivial ways.MacKenzie (2002), for example, describes how the adoption of the Black–Scholes equation for option pricing by traders did not simply enable more effective pricing of options, but helped to bring about conditions that better ? tted the assumptions on which it was based. The close empirical ? t between the predictions of the equation and options prices was bought about, at least in part, by the use of the equation to identify arbitrage opportunities. The empirical ? t has deteriorated subsequently as beliefs have changed to incorporate, inter alia, changed beliefs about the likelihood of market crashes.We pick up this theme of the re? exive relationship between beliefs and markets in Chapter 4. 1. 3 Overview of Book Chapters 2 and 3 set the context for our study and exploration of the role of traders. Chapter 2, ‘The Growth of Financial Markets and The Role of Traders’, considers the growth of inter national ? nancial markets in a historical context and outlines the role investment banks and professional traders have come to play. In Chapter 3, ‘Economic, Psychological, and Social Explanations of Market Behaviour’, we take a more detailed look at differing economic, psychological, and social explanations of market behaviour.Chapter 4, ‘Traders and Their Theories’, considers the nature of traders’ knowledge and the interplay between their subscriptions to theories of the ‘way the world works’ founded in neoclassical ? nancial economics and their more particularist and idiosyncratic theories of ‘how to work the world’. Chapter 5, ‘A Framework for Understanding Trader Psychology’, starts by outlining a psychological model of the trader founded in a selfregulation framework. It draws on the qualitative and quantitative evidence that we have about trader decision-making and bias. It challenges the ? ancial econo mics dichotomy between rational and non-rational and explains the different rationalities that arise as a consequence of internal goal states. We also present evidence on the vulnerability of traders to control illusions and the consequences for their performance. 8 Introduction Chapter 6, ‘Risk Takers: Pro? ling Traders’ presents a new model of risk taking that shows how trader behaviour emerges from a web of circumstantial and individual causes. The remainder of the chapter explores these individual differences in greater depth, especially how personality impacts different kinds of risk taking and decision-making.The chapter explores what kinds of people traders are, focusing particularly on personality and risk propensity, but also drawing on what we know about their demographics and background. Chapter 7, ‘Becoming a Trader’, uses a career transitions framework and a model of social learning to frame trader development and entry into a community of trad ing practice. We examine the ways in which they both learn and construct knowledge about the process of trading. In Chapter 8, ‘Managing Traders’, we explore the ways in which traders are monitored and managed within investment banks.We highlight the fact that traders are often not ‘managed’ at all, so much as monitored. Our concluding chapter (Chapter 9) draws together the implications of our ? ndings for traders, their management and regulation, and for further research. Notes 1. Arbitrage: purchasing currencies, securities, or commodities in one market for resale in others in order to pro? t from price differences. The effect of arbitrage is to act as a mechanism to bring about convergence of prices in different locations and markets or between equivalent securities. . A more detailed account of the sample and methods is given in the appendix. 3. We give a more detailed treatment of behavioural ? nance arguments in Chapter 3. 4. Often referred to as the institutional microstructure. 9 Chapter 2 THE GROWTH OF FINANCIAL MARKETS AND THE ROLE OF TRADERS Hardly a day passes without newspapers and television carrying a story about ? nancial markets and their impact on our lives. Even a casual perusal of these news stories makes it apparent that the activities of ? ancial institutions and markets have come to play a central role in our economic well-being and security: whether through their direct impact on individual investments and pensions or through their pervasive impact on the level of economic activity within nations and across the globe. The last decade of the twentieth century was marked by a series of international ? nancial crises. These underlined both the interdependence of national economies and ? nancial markets and the global scope of those markets. Financial crises in Latin America, the Asian Tiger economies, and Russia highlighted the speed at which capital can ? e Growth of Financial Markets countries in which investors have lost con? dence and the impotence of national governments to control such out? ows. The impact around the world of these crises on economies and ? nancial institutions demonstrated the highly interconnected nature of ? nancial markets. In the same period a number of ? nancial institutions suffered very signi? cant ? nancial losses as a consequence of the actions of single traders. One of the best publicized of these was Nick Leeson’s role in bringing about the collapse of Barings Brothers, in 1995.The collapse of Barings caused Alan Greenspan of the US Federal Reserve to comment that It is probably fair to say that the very ef? ciency of global ? nancial markets, engendered by the rapid proliferation of ? nancial products, also has the capability of transmitting mistakes at a far faster pace throughout the ? nancial system in ways that were unknown a generation ago . . . Certainly, the recent Barings Brothers episode shows that large losses can be created quite ef? cien tly. Today’s technology enables single individuals to initiate massive transactions with very rapid execution.Clearly, not only has the productivity of global ? nance increased markedly, but so, obviously, has the ability to generate losses at a previously inconceivable rate. Moreover, increasing global ? nancial ef? ciency, by creating the mechanisms for mistakes to ricochet throughout the global ? nancial system, has patently increased the potential for systemic risk. (Greenspan, 1995) While the behaviour of individual traders has at times seriously damaged the ? rms they work for, individual ? nancial institutions have also shown the capacity to endanger the stability and operation of ? nancial markets around the world.In 1998, the collapse of Long Term Capital Management, a hedge fund holding positions in ? nancial derivatives with a notional value of $1,250 billion seriously endangered the stability of the world’s ? nancial systems. How could a single trader bring down a bank? How could a single hedge fund threaten the stability of the world’s ? nancial systems? The answer lies in the way in which ‘derivatives’ allow for the multiplication of market risks (and returns). The very features that make derivatives1 so useful as a tool for managing risk provide for the possibility of massively increasing risks.In this chapter, we argue that the role of ? nancial markets, in both world and national economies, has increased dramatically. 11 Growth of Financial Markets The potential, and sometimes actual, impact of individual traders on ? rms, markets, and economies is enormous. In the following chapters we show that ? nancial markets are neither as rational nor as natural as ? nancial economists paint them and that we need to bring a wider range of social science theory to bear on understanding traders, their ? rms, and the markets they operate in.As we show below, the current globalization of ? nancial markets is not new but sim ply the latest of several cycles of international ? nancial integration over two millennia. In particular, the recent growth in international ? nancial markets could be seen as a return to levels of international ? nancial integration seen at the end of the nineteenth century and interrupted by a period, which included two world wars and the Great Depression. However, the depth and scale of these markets does seem to be different this time and the emergence of new forms of ? ancial instruments, derivatives, capable of massively multiplying possible risks and returns has led to a qualitative difference in the potential impact of individual actions on institutions, markets, and economies. 2. 1 A Brief History of Financial Markets International ? nancial markets are not a purely modern phenomenon. Basic forms of ? nancial exchange can be found throughout recorded history and international ? nancial systems are known to have existed two millennia ago. Historical evidence suggests that t here have been a series of cycles of international ? nancial integration (Lothian, 2002).In the three centuries following the collapse of the Roman Empire, currencies were very unstable and constantly debased. However, in the fourth-century AD, the Emperor Constantine introduced a stable gold coinage, the bezant (also known as the nomisa or solidus). This became widely used throughout the Mediterranean region. It was produced in Byzantium till the thirteenth century and kept more or less the same gold content through till the eleventh century. Until the introduction of the dinar in the Muslim world in the seventh century, it had no competitors as an international medium of exchange.While records are patchy, it is clear that the existence of a stable medium of international 12 Growth of Financial Markets exchange during the period between the fourth and eleventh centuries allowed quite sophisticated ? nancial transactions to take place (Lopez, 1986; Lothian, 2002). The thirteenth cen tury was another period of growth in international trade, both within Europe and between Europe and other parts of the world. Much of this was organized around regular international trade fairs (most notably at Champagne and Brie).This period was marked by the growth of an extensive and sophisticated banking system and by the development of ? nancial instruments such as bills of exchange (which acted jointly as a credit and foreign exchange transaction). It is clear from the records of the dominant northern Italian banks of the time that not only were there quite sophisticated foreign exchange markets, but also that arbitrage was a common activity (Lothian, 2002). During the fourteenth century the importance of these trade fairs and the Italian banks declined. By the ? fteenth century, Amsterdam was the more important centre of ? ancial activity. The sixteenth century saw the development, in Amsterdam, of negotiable ? nancial instruments such as discounting commercial paper and, by the seventeenth century, the development of perpetual bonds, futures contracts, selling short, and other such ? nancial instruments and techniques that would be easily recognized in modern ? nancial markets (Homer and Sylla, 1996; Lothian, 2002). By the start of the eighteenth century, the Amsterdam Exchange, the centre of Dutch trading, had become a world market in which a wide range of commodities and securities were traded.During this period, London took on increasing importance as a centre for international ? nancial trade. With the establishment of the Bank of England and the London Stock Exchange and the intervention of the Napoleonic wars, London came to eclipse Amsterdam as a ? nancial centre by the start of the nineteenth century. The nineteenth century saw a marked expansion of international trade and further development of ? nancial markets. The growth of the US economy drove much of this expansion. The New York Stock Exchange was established in 1817 and by the end of 188 6 it hit its ? st day on which more than a million shares were traded. By the late 1920s New York had overtaken London as a world ? nancial centre. However, the early twentieth century, a period that included two world wars and the Great Depression, saw the collapse of international 13 Growth of Financial Markets trade and the rise of national regulation and controls on international ? ows of capital, which effectively unwound the integration of international ? nancial markets. Rajan and Zingales (2003) show that on a range of indicators of ? nancial development including stock market capitalization as a proportion of GDP, world ? ancial markets did not regain their pre-war (1913) levels until the late 1980s. The second half of the twentieth century once again saw a very substantial increase in international ? nancial integration. As we have seen, there is historical evidence that the current period of globalization of ? nancial markets is not a new phenomenon. Rather there have bee n cycles of high international integration of markets interspersed with periods of low integration throughout the last two millennia. However, it is also clear that with each new cycle the nature and depth of those markets has been changing. Changes in the sophistication of ? ancial instruments and technologies, and changes in communications and information technologies have all been important factors in? uencing the scale and complexity of ? nancial markets. The period since the 1970s has seen a very substantial increase in the size of ? nancial markets. Figure 2. 12 shows the increase in annual 2500 Value of annual turnover (? billion) 2000 30 1500 25 20 1000 15 10 5 0 1965 0 1970 1975 1980 1985 Year 1990 1995 2000 40 Number of bargains (million) 35 Value Reported trades 500 Fig. 2. 1 Post-war UK equity market growth—UK equity turnover 1965–2002 Source: London Stock Exchange. 4 Growth of Financial Markets Value of annual turnover ($ billion) 12000 10000 8000 6000 400 0 2000 0 1967 Value Reported trades 600 500 400 300 200 100 0 2002 Number of bargains (million) 1972 1977 1982 1987 Year 1992 1997 Fig. 2. 2 Post-war US equity market growth—New York Stock Exchange equity turnover 1967–2002 Source: New York Stock Exchange. value of shares traded on the London Stock Exchange between 1965 and 2002. Figure 2. 2 shows the change in annual number of shares traded on the New York Stock Exchange between 1960 and 2002 and the annual value of shares traded from 1985.Both markets show exponential growth over the period, but the real story over the last decade is the growth in derivatives trading. By 2002, outstanding over-the-counter derivatives3 (OTC) contracts had a notional value of $128 trillion, around four times greater than total world GDP. Figure 2. 3 shows the growth in number of active contracts between 1992 and 2002. Much of the recent concern about systemic risks in markets has centred on the role of derivatives. All ? nancial invest ments carry risk. However, there is a difference of degree with derivative trading.They involve contracts which are contingent on the price of underlying assets and because of the way in which trades are regulated, derivatives4 enable investors to speculate on the price of an asset while only depositing a small proportion of the underlying asset price (margin requirements) (Zhang, 1995). In other words, the ? nancial risk borne in an options trade may be many times the money actually deposited to make the trade. Financial ? rms which do not have sophisticated control mechanisms to manage their exposure to derivatives risk may 15 Growth of Financial Markets 000 Gross market value ($ billion) Gross market value 6000 5000 4000 3000 2000 1000 0 92 93 94 95 96 97 98 99 00 01 19 19 19 19 19 19 19 19 20 20 20 02 160 000 Notional amounts 120 000 100 000 80 000 60 000 40 000 20 000 0 Notional amounts ($ billion) 140 000 Fig. 2. 3 Global growth in OTC derivatives—global value of outsta nding contracts Source: 2000–2, Bank for International Settlements; 1994–9, Swaps Monitor publications Inc. unwittingly ? nd themselves exposed to potential losses greater than the total ? rm assets. Such risks can emerge very rapidly in the course of trading and require analysis of the whole ? m’s current portfolio of trading assets in real time to identify potential overexposure to market risk. Of course, the leveraging effect of derivatives does not only affect market risk but also ampli? es risk in the other categories. For example, since derivatives typically have greater volatility than the underlying asset, even a short period in which a ? rm is unable to trade (say due to computer failure) could result in signi? cant risk exposure. The complexity of some derivatives may mean that managers are ill-equipped to understand the trades dealers are engaging in, increasing behavioural risk (Chorafas, 1995: 16).In evidence given to the US House of Representatives , George Soros, a highly successful ? nancial speculator, said of derivative instruments: There are many of them, and some of them are so esoteric, that the risks involved may not be properly understood by even the most sophisticated of investors. Some of these instruments appear to be 16 Growth of Financial Markets speci? cally designed to enable institutional investors to take gambles which they would otherwise not be permitted to take. For example, some bond funds have invested in synthetic bond issues that carry a 10 or 20-fold multiple of the risk within de? ed limits. And some other instruments offer exceptional returns because they carry the seeds of a total wipe out. (Soros, 1995: 312) 2. 2 The Role of Investment Banks in Financial Markets To understand the role of modern investment banks it is necessary to understand how world ? nancial markets have come to be dominated by an American model of ? nance. Much as Byzantium, Lombardy, Amsterdam, and London have been the dominan t centres of ? nancial innovation and power in previous eras, US ? nancial markets and institutions are today.The central feature of the US model that emerged in the post-war years was the decline of relationship banking and the increasing commoditization of ? nancial products and services. The roots of this system lie in the unintended consequences of anti-trust and banking legislation passed in the United States during the 1930s. The segregation of commercial and investment banking in the United States laid the foundation for the development of a strong investmentbanking sector. The fragmentation of the banking industry, imposed by legislation, created conditions in which ? ancial transactions were more readily managed through markets than within large banks. The elimination of ? xed commissions for broking ? nancial instruments in 1975 provided a further impetus for competition. More and more, ? rms seeking to raise ? nance looked to impersonal markets rather than relationships w ith banking institutions. Progressively more transparent and liquid markets in both corporate debt and equity and the corresponding increased competition in these markets served as a signi? cant stimulus to ? nancial innovation.As these markets developed it became apparent to market participants and to the government that effective market operation could only be maintained through active intervention and regulation. A series of waves of external and self-regulation, often in response to market crises, led to the development of regulations and supervisory arrangements designed to contain insider manipulation of markets and ensure free 17 Growth of Financial Markets ? ow of information. On the demand side, the expansion of institutional investment (insurance, pensions, and mutual funds) stimulated and was stimulated by the growth of these ? ancial markets. The slower growth of ? nancial markets and institutions in other parts of the world meant that, as other countries began to follow the United States in opening up competition, US ? nancial institutions were well placed to play a major role. In the wake of the major changes in market regulation in 1986, the long-established London merchant banks were swept away by the US-based investment banks and non-US owned European investment banks have increasingly adopted US approaches. The principal competitive advantage of American ? rms lay in their expertise in managing risk (Steinherr, 2000: 49).Investment banks manage risk in four main ways: they absorb risk for clients, they act as intermediaries for the diversi? cation of risk, they advise on the management of risk and they engage in proprietary trading—taking risk on their own account in the pursuit of returns (Casserley, 1991). Absorbing Risk Investment banks absorb risk for clients in a number of different ways. For example, when they act on behalf of a client they absorb credit risk (the risk the client will default on payment for the transaction and th ey are unable to unwind the transaction at a favourable price).They underwrite issues of securities (e. g. commercial paper5 to cover shortterm ? nancing needs), guaranteeing to buy from the client at a ? xed price should the security fail to achieve its expected price in the open market. They also play an important risk absorption role in trading markets. In some of these the bank will act as a market-maker,6 providing liquidity in a particular ? nancial instrument. The bank ? xes prices at which it will buy or sell a ? nancial instrument and stands ready to buy or sell at those prices even if there is no party to pass the transaction on to immediately.In return for the spread between these prices the bank absorbs the risk of the market moving against them. Risk Intermediation In other cases the bank will act as an intermediary for the diversi? cation of clients’ risk. This may be by acting as an intermediary in trading 18 Growth of Financial Markets markets or by putting to gether complex OTC deals that rely on aggregating (or disaggregating) ? nancial instruments provided by third parties. The banks bene? t from this intermediation work in two principal ways. First, they charge commission and second, they have access via their customers to information about order ? ws in the markets in which they operate. Such ? ow information provides opportunities to exploit temporary market imperfections and pro? t through trading on their own account. Risk Advice The risk advice role overlays risk absorption and risk intermediation. For example, the bank may play an important advisory role related to underwriting activities or in putting together a complex OTC deal. The role of the bank in providing risk advice to clients rests not just on technical skills and experience in managing risk, but also in a (sometimes) greater overview of the markets in which they operate.An important issue here is the tension between the bank’s desire to make pro? ts on its own account and to earn a return through providing effective advice and services to customers. This tension is re? ected to some extent in tensions which emerge in most banks between trading and sales desks. As we will see later in the book, banks vary in the priority they give to serving customer needs versus seeking opportunities for returns through trading on their own account. 7 Proprietary Trading In providing services to customers, investment banks build up information on order ? ws, they develop expertise in valuing particular securities or in economic fundamentals in particular sectors or countries, they build proprietary models of price behaviour and they build up data on historic behaviour of prices and relationships between them. This can place them in a better position to judge risks and returns than other market participants and opens up the possibility of earning good returns on their own account. This activity typically takes two forms: short-term (often intra-day) trade s designed to exploit knowledge of temporary price ? ctuations linked to ? ows of orders in the market and longerterm trades, often based on arbitrage (exploiting pricing inconsistencies between different securities, markets, or time periods). 19 Growth of Financial Markets 2. 3 The Role Played by Traders The work of traders can be divided into three broad categories: trading on behalf of customers, market-making, and proprietary trading. 8 Traders acting on behalf of customers take the least risk on behalf of the bank, while proprietary trading potentially involves the greatest risk.However, in practice, the three spheres of activity often overlap. For example, a trading desk acting on behalf of clients may also have authority to take intra-day positions to bene? t from short-term price movements in the markets they operate in. Alternatively, in some circumstances, while not strictly acting as a market-maker, they may stand ready to create liquidity for important clients by buying or selling to those clients when they cannot ? nd a counterparty for their trades. As one senior trader told us: We are paid to be on the wrong side of the market for our customers.If we have an institution that pays us thirty million dollars a year in commissions, we will, on occasion at their request, be a buyer for them when there are only sellers on the market or be a seller for them when there are only buyers. When they’re in a more normal market environment where there is plenty of liquidity and good two-way ? ow, they don’t necessarily need our capital. In fact they prefer not to use our capital because all that does then is create another buyer or another seller in the market with them.But when the market is heavily tilted in one direction than the other, even the market’s selling off, there are much more sellers than buyers or a very strong market where there are much more buyers than sellers. That’s when they need us to step in and serve as that intermediary to facilitate the execution of their order. 9 Alternatively, a trading desk operating as a market-maker may combine this with some proprietary trading. One trader described the activity of his desk: We have a P&L [pro? t and loss], budget of about $20m a year through plain vanilla market making with customers.However, we make about half the money in proprietary trading using the ? ow and information from customers—putting it on our book instead of putting it back into the market. For the ? rst half of this year we were number one for turnover in our niche with between 10% and 15% of the market. The 20 Growth of Financial Markets more that number increases, the better information we would have for proprietary trading, but we would probably start losing money from the market making function because prices would have to be so keen, so there is a balance.Equally, traders mostly engaged in proprietary trading will seek opportunities to generate customer business: I d o proprietary business and I’m supposed to be doing proprietary but I interface with the ? ow desk so I would be looking at customer business trying to generate customer business. My slant is proprietary but I’m always trying to emphasise customer business using my positions. 2. 4 How do Traders Make Pro? ts? If, in ef? cient markets, price changes are essentially a random walk and all new information relevant to prices is incorporated into prices instantaneously (Fama, 1970), then how do traders make money? The ? st answer is that they charge commission for their intermediation and advisory role. By aggregating customer orders they can reduce transaction costs. However, as we will explore in Chapter 3, in practice, markets are not completely ef? cient and information asymmetries exist. Traders essentially earn economic rents10 by exploiting information advantages. These may come from a number of sources, including information on asset ? ows within markets (e. g. from having a large customer base); privileged information on the economic basis for an asset price; proprietary databases allowing more accurate calculation of probabilities (e. . historical asset volatility for pricing options); models of the relationship between prices and economic fundamentals; models for extracting the information inherent in historical price changes of an asset and other related assets; and effective understanding of the ‘sentiment’ and likely behaviour of other market actors. All of these information advantages are potentially short-lived. The very act of trading may reveal information to other parties. Others may emulate models. Others may access the same sources of information.New information may wipe out the utility of earlier information. At the same time markets are in practice very ‘noisy’. That is to say, there is a lot of trading going on that is not based on information 21 Growth of Financial Markets genuinely relevant to the und erlying value of an asset. Black (1986) noted in his presidential address to the American Finance Association that Traders can never be sure that they are trading on information rather than noise. What if the information they have is already re? ected in prices? Trading on that kind of information will be just like trading on noise.Traders can only earn above market returns, on average, over time, if they are genuinely trading on new and relevant information. However, on any individual trade it will be dif? cult to tell whether a positive outcome is the result of trading on information or of essentially unpredictable market movements (as a result of noise trading in the market, changes in sentiment, or new unexpected events). Similarly, for any individual trade it is dif? cult to determine whether a negative outcome is the result of trading on noise rather than information or the result of unforeseeable market movements.So it will often be the case that trading outcomes are not cont ingent on the trader’s strategy or information. Further, it will often be dif? cult to determine once an outcome is achieved whether the outcome was indeed contingent on a trader’s information and skill. While trading is a skilful activity, many trading outcomes are not contingent on skill. At the same time traders are highly motivated to establish causal relationships between information they hold and prices, since a signi? cant source of rent for any trader is the capacity to establish contingent relationships before others observe them.This problem of determining the links between behaviour and outcome for traders is one we will return to repeatedly in the book. While the detail of different trading strategies is not our principal focus, we describe some common trading approaches to set the stage for our later discussions. In order for traders to achieve better than average market returns, it is not suf? cient that markets are imperfect; it is also necessary they ha ve some competitive advantage relative to others who seek to exploit those imperfections.Within this fast-moving and uncertain world, traders adopt a variety of strategies to exploit the information and expertise to which they have access. These can be divided into four main categories: insider strategies, technical strategies, fundamental strategies, and ? ow strategies. 22 Growth of Financial Markets Insider Strategies Insider strategies involve achieving advantage by exploiting privileged access to information (Casserley, 1991). Of course, some such strategies are illegal. It is, for example, illegal to exploit privileged access to advanced knowledge of company earnings news or potential takeovers.However, most of these strategies are concerned with perfectly legitimate attempts to build an information advantage over rivals. The extent to which it is possible to achieve such information advantages varies signi? cantly from market to market. For example, in relatively undeveloped markets such as the ‘emerging markets’ there may be frequent and persistent information asymmetries. In these circumstances, traders who are able to establish good personal networks may build an advantage, which enables them to anticipate price movements. However, in mainstream equities markets, the speed and ef? iency of information dissemination may make such advantages dif? cult to achieve. Insider strategies can improve a trader’s ability to anticipate market movements. However, as we noted earlier, it is often dif? cult or impossible for a trader to determine whether they have a genuine information advantage or whether their information is simply noise, already discounted by the market. Technical Strategies If markets are perfectly ef? cient, then historic prices contain no information that can be used to infer future price movements. However, many traders claim to do just that.They seek to exploit market imperfections through the analysis of past price info rmation. One form of technical trade concerns using patterns in price data to identify likely turning points in price trends (charting). Traders seek to identify trends early, buy into those trends and exit before the trend breaks. Many traders consider these patterns and trends in market prices to be driven by underlying investor sentiment. While there is some evidence that supports the existence of exploitable patterns in market prices (e. g. Kwon and Kish, 2002), many ? ancial economists are sceptical of their existence. Fama (1970) dismissed technical analysis as a futile undertaking on the grounds that historical prices have no predictive validity. However, more recent arguments against technical 23 Growth of Financial Markets trading strategies take a weaker position: that while there is some predictability in market movements, exploiting these does not, on average, make returns in excess of transaction costs (e. g. Allen and Karjalainen, 1999). A second important technical st rategy requires the analysis of historical price relationships between different ? ancial instruments. Traders scan markets looking for discrepancies in pricing relative to these relationships on the assumption that they will move back to the historical pattern. Often the gains on technical trades will be small and over short time periods, thus these trades often depend on an ability to identify opportunities rapidly and frequently. This allows the trader to make large numbers of such trades each making a small pro? t. To bene? t from such trading strategies requires the ability to trade at low transaction costs, frequently, with considerable IT support.Many traders use technical strategies to supplement other approaches. For example, a trader having established a trade on the basis of customer ? ow information may use technical information on trend behaviour to determine the precise point at which to take pro? ts or cut losses. Others, while fundamentally sceptical about strategies relying on historical trend data, assume prices will be driven to some extent by investors using such models. For example, one trader told us: A lot of traders are chartists and a lot of people here don’t like you looking at charts, they don’t believe in them.However, I look at a chart if I am putting on a large position, or looking for something to trade because if there are people out there who use charts as a model to trade, this will affect how things trade in the markets whether I believe in it or not. Fundamental Strategies Technical strategies are purely concerned with anticipating trends and pay no attention to the underlying economic basis for evaluation of the security being traded. By contrast, fundamental strategies are concerned with the fundamental relationship between economic value of the underlying asset and market price.Traders following these strategies essentially seek to use expertise and information in the accurate valuation of securities, on the assumption that market values will 24 Growth of Financial Markets converge to theoretical values. To the extent that traders can establish an advantage in valuation of securities, they may be able to earn pro? ts from identifying securities that are undervalued or overvalued by the market. One highly successful trader told us: I tend to take positions that depend a lot on central bank decisions e. g. nterest rates, so depend on macro economic position of the country, the judgement about how the Bank of England is going to behave and how the market is going to proceed. I try to put myself in Eddie George’s11 feet and try to understand. We have been building a model of Bank of England reactions to economic events. I have lunches with people who decide our interest rates and try to understand how they think . . . It all comes down to focus and completely immersing myself in an area. However, as with insider strategies it can be genuinely dif? cult for a trader to understand whe ther they have a genuine advantage in valuation.Further, as we will see in Chapter 3, trading on valuation advantage depends on the market converging to a value in a time scale over which you can ? nance a trade. Flow Strategies This strategy predicts prices as a function of demand and supply for securities in the market. Particularly for securities in which there is not much liquidity,12 large trades can shift prices signi? cantly. Where a bank has a large customer base in a particular niche, this can give them access to valuable market information, in particular, information on trading ? ows.These kinds of advantage are more readily achieved in OTC markets, which lack the transparency of trades organized through exchanges. However, in any given market niche, there will be a very limited number of ? rms that can capture suf? cient order ? ow information to give them a genuine advantage. Feldman and Stephenson (1988) studied the use of ? ow information in the US treasury bonds marke t. They suggest that through the use of informal information trading with customers, a ? rm with a 3–4 per cent share in trading may have a good sense of what is going on in 30 per cent or more of the market.However, they also show that medium sized players in these markets are often unable to exploit their customer relationships effectively. They argue that large players systematically 25 Growth of Financial Markets shut medium sized players out of information networks while providing good market information to smaller players who they mostly relate to as customers rather than competitors. As we have seen, ? nancial markets have a long history and have been through multiple cycles of global ? nancial integration over the last two millennia, but their development into domains of such immense complexity and global in? ence has occurred only within the last 50 years. The volume of trading and of traders has no historical precedent, nor has the complexity and variety of the inst ruments traded. Within this context, the activities of traders within investment banks are important not just to their customers, but also at the level of national and international economies. Naturally, these phenomena have attracted the attention of academics and commentators, from a variety of disciplines, who have, as we shall show in the next chapter, different and sometimes competing explanations of what in? uences and explains behaviour within global ? ancial markets. Notes 1. Derivatives are ? nancial products, which depend on or derive from other assets. 2. Values in all ? gures are nominal (non-in? ation-adjusted). 3. OTC derivatives are not traded in an exchange but are contracted directly between the two contracting parties. 4. Exchange requirements generally only require traders selling options to deposit a proportion of the potential claim. Further, speculation using derivatives is often highly leveraged (funded through borrowed funds). 5. Market traded short-term corp orate debt. 6. Market-makers stand ready to buy or sell an asset or class of assets.Typically a market-maker quotes a buy (bid) and sell (offer) price to a client before the client declares whether they wish to buy or sell. The spread between bid and offer both provides a return and some protection against market movements in the time taken for the marketmaker to readjust their holdings after a trade. 7. There are also important differences between the United States and the United Kingdom in how this tension is regulated. UK banks face fewer constraints on the relationship between customer business and proprietary trading. 26 Growth of Financial Markets 8. The types of ? ancial instruments dealt in by traders cut across these categories. Some traders specialize by a particular type of instrument (e. g. equities or bonds in a particular sector), others deal in a range of instruments related to a particular geographical region or sector. 9. See also Abola? a (1996) for a description o f such market stabilizing behaviour by market-makers. 10. Returns in excess of the market risk premium. 11. Eddie George was Governor of the Bank of England at the time of interview. 12. Liquidity: the availability of parties willing to buy or sell a security at any given time. 27 Chapter 3ECONOMIC, PSYCHOLOGICAL, AND SOCIAL EXPLANATIONS OF MARKET BEHAVIOUR For at least forty years psychologists have amassed evidence that economic man is very unlike a real man and that reason—for now, de? ned by the principles that underlie expected utility theory, Bayesian learning and rational expectations—is not an adequate basis for a descriptive theory of decision making. De Bondt, 1998 I am in fundamental disagreement with the prevailing wisdom. The generally accepted theory is that ? nancial markets tend towards equilibrium and, on the whole, discount the future correctly. I operate using a different theory, according to which ? ancial markets Market Behaviour cannot possibly di scount the future correctly because they do not merely discount the future; they help to shape it. Soros, 1995: 111 If we are to understand traders, we have to ? rst understand the markets they inhabit. Neoclassical economics has been extraordinarily successful in explaining most market behaviour in the aggregate. However, it has two principal weaknesses for our purposes. The ? rst concerns what it does not address and the second concerns some important failures at the margins. Neoclassical ? nancial economics treats markets as a given, or naturally arising.Investor preferences and risk appetites are treated as external to the model but predictably ordered and distributed. Markets are modelled as adjusting instantaneously with little attention to the detail of how such adjustments come about. While neoclassical ? nancial economic models effectively explain a great deal of market behaviour, there are some important failures at the margins. There is a wide range of anomalies which are dif? cult to explain within this paradigm. If markets instantaneously adjust and are perfectly ef? cient, then the only role for professional traders is as intermediaries who cannot earn above market returns, but ssentially earn commission as intermediaries. There is nothing to be earned by arbitrage activities or speculation. Indeed, it is not even clear within neoclassical accounts of markets that there is a role for intermediation. However, if we assume markets to be only nearly perfect and ‘sticky’, the trader’s role as someone with privileged expertise, tacit knowledge, and access to private information (within limits) makes more sense. Here, traders are the oil in the market machine; they are on