8409 Prelims pi-xiv 16/4/02 3:21 PM Page i

Managing Risk
in Alternative
Investment Strategies
8409 Prelims pi-xiv 16/4/02 3:21 PM Page ii

In an increasingly competitive world, we believe it’s quality

of thinking that will give you the edge – an idea that opens
new doors, a technique that solves a problem, or an insight
that simply makes sense of it all. The more you
know, the smarter and faster you can go.
That’s why we work with the best minds in business
and finance to bring cutting-edge thinking and best
learning practice to a global market.
Under a range of leading imprints, including
Financial Times Prentice Hall, we create world-class
print publications and electronic products bringing our
readers knowledge, skills and understanding which can
be applied whether studying or at work.
To find out more about our business publications, or tell us
about the books you’d like to find, you can visit us at
www.business-minds.com
For other Pearson Education publications, visit
www.pearsoned-ema.com
8409 Prelims pi-xiv 16/4/02 3:21 PM Page iii

Managing Risk
in Alternative
Investment Strategies
Successful Investing in Hedge Funds
and Managed Futures
DR LARS JAEGER

London ■ New York ■ Toronto ■ Sydney ■ Tokyo ■ Singapore

Hong Kong ■ Cape Town ■ New Delhi ■ Madrid
Paris ■ Amsterdam ■ Munich ■ Milan ■ Stockholm
8409 Prelims pi-xiv 16/4/02 3:21 PM Page iv

PEARSON EDUCATION LIMITED

Head Office:
Edinburgh Gate
Harlow CM20 2JE
Tel: +44 (0)1279 623623
Fax: +44 (0)1279 431059

London Office:
128 Long Acre
London WC2E 9AN
Tel: +44 (0)20 7447 2000
Fax: +44 (0)20 7240 5771
Website: www.financialminds.com

First published in Great Britain 2002

© Pearson Education Limited 2002

The right of Dr Lars Jaeger to be identified as Author

of this Work has been asserted by him in accordance
with the Copyright, Designs and Patents Act 1988.

ISBN 0 273 656988

British Library Cataloguing in Publication Data

A CIP catalogue record for this book can be obtained from the British Library.

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, electronic, mechanical, photocopying, recording, or otherwise without either the
prior written permission of the Publishers or a licence permitting restricted copying in the United Kingdom
issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1P 0LP. This book may not
be lent, resold, hired out or otherwise disposed of by way of trade in any form of binding or cover other than
that in which it is published, without the prior consent of the Publishers.

This publication is designed to provide accurate and authoritative information in regard to the subject matter
covered. It is sold with the understanding that neither the author nor the publisher is engaged in rendering legal,
investing, or any other professional service. If legal advice or other expert assistance is required, the service of a
competent professional person should be sought.

The publisher and contributors make no representation, express or implied, with regard to the accuracy of the
information contained in this book and cannot accept any responsibility or liability for any errors or omissions
that it may contain.

10 9 8 7 6 5 4 3 2 1

Typeset by Pantek Arts Ltd, Maidstone, Kent

Printed and bound in Great Britain by Bookcraft Ltd, Midsomer Norton

The Publishers’ policy is to use paper manufactured from sustainable forests.

8409 Prelims pi-xiv 16/4/02 3:21 PM Page v

To my wife Julie
8409 Prelims pi-xiv 16/4/02 3:21 PM Page vi
8409 Prelims pi-xiv 16/4/02 3:21 PM Page vii

About the author

Lars Jaeger is a Partner of Partners Group, one of the largest European alternative
asset managers, based in Zug, Switzerland. He was a Managing Director and
Co-founder of saisGroup, a Swiss-based specialist firm for multi-manager
Alternative Investment Strategies (AIS) portfolios, which merged with Partners
Group in late 2001. He is responsible for quantitative analysis and risk management
for the Hedge fund portfolios managed at Partners Group.
Lars holds a PhD degree in theoretical physics from the Max-Planck Institute
for Physics of Complex Systems in Dresden, Germany (1997). He studied physics
and philosophy at the University of Bonn and Ecole Polytechnique in Paris. He
worked as a researcher in different areas of theoretical physics (quantum field
theory, atomic physics, and chaos theory).
Lars started his financial career at Olsen and Associates in Zurich as a quantita-
tive researcher, where he designed econometric and mathematical models for
financial markets (systematic trading models, portfolio and risk management).
He subsequently moved to Credit Suisse Asset Management, where he was
responsible for risk management and quantitative analysis of Hedge fund and
Managed Futures strategies.
Lars is author of numerous research publications in various leading scientific
journals and has been a regular speaker at diverse seminars and workshops.

.... vii ....

8409 Prelims pi-xiv 16/4/02 3:21 PM Page viii
8409 Prelims pi-xiv 16/4/02 3:21 PM Page ix

Contents

Preface xiii

1 Introduction 1
Changing investor demand 3
LTCM: What can go wrong for Hedge fund investors 5
Why effective risk management is crucial to realizing the
benefits of AIS 6
A new investment paradigm 7
The AIS investment approach and integrated risk management for
multi-manager portfolios (‘fund of funds’) 8
Is transparency achievable in AIS investments? 9
Liquidity of AIS investments 13
The challenges of AIS risk management 14
Notes 15

2 The Universe of Alternative Investment Strategies (AIS) 17

Definition of the AIS Universe 17
Development of AIS 20
Understanding the sources of AIS returns 24
Economic functions and Risk premiums 27
The structure of AIS multi-manager funds 31
Notes 34

3 Alternative Investment Strategies:

Hedge funds and Managed Futures 38
Hedge fund strategies 38

.... ix ....
8409 Prelims pi-xiv 16/4/02 3:21 PM Page x

Contents

Relative Value – Convertible Arbitrage 40

Relative Value – Fixed Income Arbitrage 48
Relative Value – Equity Market Neutral 53
Event Driven – Risk Arbitrage 60
Event Driven – Distressed Securities 69
Event Driven – Convertible Debenture Arbitrage (‘Regulation D’) 73
Opportunistic – Global Macro 78
Opportunistic – Long/Short Equity 82
Opportunistic – Equity Market Timing 87
Opportunistic – Short Selling 91
Managed Futures and Currency strategies 95
Futures Active – Systematic Technical 97
Futures Active – Discretionary 100
Futures Passive 102
Notes 106

4 Empirical Properties of Alternative Investment Strategies 111

Challenges of AIS performance measurement 112
Risk and return 113
Comparison with equities and bonds 119
Unconditional correlation properties 120
Conditional correlation properties 123
Behaviour in extreme market situations 126
Benefits of AIS in a traditional portfolio 129
Summary of empirical properties 130
Appendix: Data providers for AIS 131
Notes 134

5 Risk in Alternative Investment Strategies 136

Risk factors in AIS investments 136
Risk from an investor’s point of view 143
Risk factors of the different AIS sectors 145

.... x ....
8409 Prelims pi-xiv 16/4/02 3:21 PM Page xi

Contents

Risk management principles for the different AIS sectors 152

Portfolio diversification 164
Notes 168

6 Tools and Principles of Modern Financial Risk Management 171

The emergence of quantitative tools for risk management 172
Pre-VaR risk management: Traditional risk measures and
‘modern’ portfolio theory 173
Value-at-Risk (VaR) 174
VaR: Back testing and model verification 180
The problems of VaR 180
Variations of VaR 183
Stress testing and scenario analysis 184
Extreme Value Theory 185
Liquidity Risk 188
Credit Risk 189
Other risks 191
Enterprise-wide risk management 194
Active risk management 195
Risk analysis systems in practice 196
Notes 198

7 Active Risk Management in Multi-Manager AIS Portfolios 205

The AIS industry’s current best practices for risk management 205
Multi-manager portfolios 208
The issue of transparency in AIS 211
Liquidity of AIS 220
Managed accounts 222
Fund Redemption Policies for multi-manager
investment vehicles 223
Leverage 225
Fee structure and other key administrative considerations 227

.... xi ....
8409 Prelims pi-xiv 16/4/02 3:21 PM Page xii

Contents

The investment and risk management process for AIS

multi-manager portfolios 229
Strategy sector selection (top down approach) 232
Manager evaluation (bottom up approach) 237
Investment monitoring: ‘Post-investment’ risk management 246
Active risk management 254
Notes 257

8 Conclusion and Outlook 262

The AIS investment process, risk management, transparency
and liquidity 262
Regulating AIS? 266
The Hedge fund bubble? 266
Outlook for AIS 269
Notes 271

Glossary 272

Bibliography 279

Index 289

.... xii ....

8409 Prelims pi-xiv 16/4/02 3:21 PM Page xiii

Preface

Alternative Investment Strategies (AIS) – Hedge funds and Managed Futures –

have grown rapidly over the last few years and are on track to become a trillion-
dollar industry in the next years. So it is not surprising that there is a surge in
interest in AIS from a broad array of investors – institutional as well as private.
Until recently Hedge funds were considered the ‘cowboys of financial markets’ or
‘courtesans of capitalism’ (a title of a recent book by Peter Temple) and were of inter-
est only to the superrich. But sophisticated investors now understand that, if properly
included in a global portfolio, AIS can serve as a valuable diversifier. Approached
carelessly, however, they can easily create an investment disaster. Current and poten-
tial AIS investors are thus demanding improved risk management so that the benefits
of AIS can be realized while exposure to risk remains at acceptable levels.
Recent developments have given way to a new generation of AIS managers
who are better and more professional risk managers and thus have more credibil-
ity in the eyes of a much broader class of investors. Yet few even in the
professional investment community are prepared to meet the two key challenges
of managing AIS risk: complexity and rapid change. Alternative investment strate-
gies are much more complex and varied than traditional asset classes (equities and
bonds). To add to the challenge, AIS strategies and overall risk management prac-
tices are changing even as investment professionals struggle to master them.
Despite its challenges, active AIS risk management adds tremendous value to
the investment and asset allocation process of AIS investors and managers. This
book will provide the reader with the knowledge needed to reap these rewards. It
is not a ‘cookbook’ though, and does not provide fixed recipes for how to invest
in Hedge funds or manage their risks. Proper AIS investment is both an art and a
science. While the science refers to the increasingly quantitative approaches taken
by many AIS managers, the ‘art’ of AIS investing is the understanding of the com-
plexity of AIS and the experience necessary to appropriately allocate assets among
different strategy sectors and managers. This book aims at giving the reader an

.... xiii ....

8409 Prelims pi-xiv 16/4/02 3:21 PM Page xiv

Preface

understanding of this complexity and guides him in his assessment of AIS risks
and the appropriate ways of managing those risks.
With the increasing popularity of Hedge funds, the literature about AIS has
grown considerably during the last few years. The interested reader has a choice
among many different views and approaches. But despite its immense importance
to the AIS investment process, the topic of risk management has not yet been cov-
ered in sufficient detail. There is a variety of (in some cases excellent) articles –
often collected in multi-authored books – that provide insight into particular facets
of the topic. But the industry lacks coherent and comprehensive coverage of AIS
risk management in one publication. This is what motivated me to write this book.

Acknowledgements
This book represents untold hours of effort by many people other than myself and I
would like to thank everybody who helped me to complete this work. The first
person I owe gratitude to is my dear wife, Julie, who provided love, understanding
and support throughout many hours of writing. She also provided invaluable feed-
back and criticism while reading the manuscript and without her the book would
not have taken its present form. I would also like to acknowledge Michael
Jacquemai and Pietro Cittadini who were my partners and co-founders of saisGroup
and are now my partners at Partners Group. They were the joint architects of many
of the ideas presented in this book and who also provided valuable feedback on the
manuscript. My thanks also go to Renato Amrein at Partners Group for valuable
comments and proofreading of the manuscript. I also acknowledge the comments
from the following individuals: Susanne Classen (Dr. Hehn Associates), Bill Feingold
(Clinton Group), Michael Manning (Stratton Advisors), Patrik Säfvenblad (RPM),
Jeffrey Pease (Business Objects), Peter Rice (Ecofin Investment Consulting), Daniel
Rizzuto (Graham Capital Management), Adam Segal and Robert Rice (DLR
Advisors), Anthony Todd (Aspect Capital), and David White (JE Matthew).
Last, but not least, I thank Financial Times Prentice Hall for their enthusiastic
support of this book and for their assistance in editing and reviewing the manu-
script. Despite the extensive support I received, there will be mistakes,
misrepresentation and omissions in the book, for which I take full responsibility.
Lars Jaeger
April 2002

.... xiv ....

8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 1

CHAPTER 1

Introduction

Investing in Alternative Investment Strategies (AIS), i.e. Hedge funds and Managed
Futures, has become a multi-billion dollar industry and recent years have seen
unparalleled capital inflows into AIS. The attractive risk–reward characteristics of
AIS funds as well as their low correlation to traditional asset classes have led to
widespread interest in AIS investing. It is estimated that there are currently more
than half a trillion dollars invested with about 5,000 Hedge fund and Managed
Futures programs worldwide, the largest part of which originates in the USA and
Europe. The AIS industry is enjoying a 15–20% annual asset growth rate and it is
expected that AIS investing will develop towards a trillion dollar industry in just a
few years. Hedge funds and Managed Futures managers have become important
players in world financial markets, accounting for a good part of the daily trading
volume in numerous financial instruments.
Despite these very positive trends, in order for AIS to achieve its full potential,
the industry must address growing investor concerns about the diverse risks of AIS
investments as well as the lack of investment transparency, low liquidity and long
redemption periods which are generally characteristic of Hedge fund and
Managed Futures investments. The trend in investors’ attitude from accepting
(‘trust me’) to requesting (‘show me’) is clearly observable. While for years
investors followed a ‘black box’ approach to AIS investing, a number of factors
are leading to a shift away from this type of approach. Increased interest from

.... 1 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 2

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■■■■■■■■■■■■ institutional investors in AIS has led to new

demands for disclosure due to the fiduciary
Increased interest from
responsibilities associated with investing
institutional investors in
clients’ money. Further, several widely publi-
AIS has led to new cized Hedge fund failures during the market
demands for disclosure crisis of 1998 (e.g. LTCM) and periodic
due to the fiduciary reports of other Hedge fund ‘blow ups’ and
responsibilities fraud (e.g. ‘Manhattan Hedge Fund’ in the
associated with spring of 2000) have added to the concerns of
investing clients’ money all AIS investors about the risks of such invest-
ments. Finally, rapid developments in the
■■■■■■■■■■■■ financial industry in the area of financial risk
management have made risk analysis for even
complex AIS portfolios feasible on a real-time basis and therefore have increased
expectations with respect to the management of risk.
Discussion about how to address concerns regarding investment risk, low
liquidity and insufficient transparency in AIS is in its early stages. In this book I will
elaborate on what I refer to as the ‘transparency paradigm’ in which full disclosure
by AIS managers, detailed understanding of sources of returns and risks on the part
of AIS investors and active risk management by the AIS portfolio managers allow
investors to reap the benefits of AIS investing while eliminating undesired risks. I
will argue that such an approach is not only feasible but also essential for properly
controlling the risks of AIS and satisfying investor expectations.
I will further provide the elements and tools necessary for effective risk manage-
ment for AIS. An understanding of AIS risk issues necessitates thorough knowledge
of the underlying investment strategies. Chapter 2 opens with a presentation of the
evolution of Hedge funds and Managed Futures and gives a general characterization
of AIS, before I provide, in Chapter 3, a more detailed description of the various
AIS sectors. Here, my focus is on characteristics, sources of return and dominant
risk factors of the individual strategies. It should be clear that I cannot claim com-
plete knowledge about every single strategy, so my description is at risk of appearing
incomplete or selective to some experts (e.g. Hedge fund managers), but it aims at
providing the reader with a balanced view of all necessary aspects of the AIS sectors.

.... 2 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 3

1 ■ Introduction

Chapter 4 contains a presentation of empirical properties of AIS strategies, includ-

ing their diversification benefits, which are particularly important from the perspective
of a portfolio manager. A description of the most important general AIS risk factors
and a comparison and quantification of the specific risks of each strategy follows in
Chapter 5. In Chapter 6, I outline the principles of risk analysis in financial markets,
with the greatest focus on quantitative risk tools (Value-at-Risk, stress testing, scenario
analysis, leverage control etc.). While the chapter is merely an overview, references to
the literature are given for the reader interested in more details. Chapter 6 also dis-
cusses risk service providers and risk managing tools available to the AIS manager from
third parties today. Finally, Chapter 7 describes the principles of managing risk in an
AIS portfolio. It discusses how risk management can be integrated into the AIS asset
allocation process, describes approaches to sector allocation and manager evaluation
and outlines appropriate methods of portfolio monitoring and active risk control.
The book aims to provide a wide range of financial professionals, including Hedge
fund and Managed Futures managers, fund of funds managers (AIS allocators), bro-
kers, administrators, custodians and private and institutional investors with an
understanding of AIS risks and risk management. But the book is also well suited for
other types of professional involved with addressing the challenges of AIS risk such as
regulatory agencies, consultants, legal authorities, financial journalists and students.
Despite the broader view taken on the subject I hope even AIS experts will benefit
from the discussion presented. As the book is addressed to a broad audience, I avoid
the use of mathematical formulas. The knowledge necessary for this book is a basic
understanding of equity, fixed income, foreign exchange and commodity markets
(including plain vanilla derivatives such as options and futures) and the core principles
of modern portfolio theory. The discussion of the quantitative risk analysis tools
(Chapter 6) might require newcomers to the field of financial risk management to do
some background reading (references are provided).

Changing investor demand

Traditionally, Hedge fund and Managed Futures investing has been dominated by
high net worth investors who were willing to bear the disadvantages of illiquid
and non-transparent investment strategies (‘black box investing’). Their focus was

.... 3 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 4

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■■■■■■■■■■■■ often rather short term, they were less con-

cerned about diversification and accepted high
This new class of
levels of volatility. Recently, institutional
investor has put the investors have become increasingly interested
issue of risk in AIS, drawn to their attractive risk–return
management at the top characteristics and low correlations to tradi-
of their priority list tional asset classes. However, this new class of
investors has different demands to those of
■■■■■■■■■■■■
high net worth individuals. They have a com-
parably long-term view, show higher levels of
risk aversion and emphasize the stability of investment returns. Institutional
investors have put the issue of risk management at the top of their priority list.
Prerequisites of risk management are transparency and investment liquidity.
Unfortunately, many of the investment vehicles for AIS available in the market today
present investors with numerous liquidity and transparency issues and, therefore, risk
management problems. Most managers supply too little information to investors.
Monthly returns, standard deviations, maximal drawdowns and, in most cases, a
monthly or quarterly letter to the investors, do not provide sufficient information
about investment risk.
As a result of these recent developments (growing demand from institutional
investors), the AIS industry is currently going through an institutionalization
process in which Hedge fund and Futures managers are increasingly faced with
demands for increased investment transparency, higher liquidity (i.e. shorter
redemption periods) and greater clarity in terms of portfolio composition,
strategy details, performance and fee attributions and leverage. Investors’ views
range from managing risk through diversification across many AIS managers such
that a limited number of ‘blow ups’ have a minimal effect on the portfolio, to a
fully transparent and actively risk controlled investment approach.
A note here on terminology: throughout the book, I will use the word ‘manager(s)’
to describe the individual(s) responsible for the development and execution of the
Hedge fund or Managed Futures trading strategy (the word ‘trading advisor’ is also
commonly used in the AIS industry). For a multi-strategy portfolio manager (AIS
fund of funds manager), I will often use the term ‘allocator’.

.... 4 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 5

1 ■ Introduction

Due to the technical complexity of AIS, which include spread strategies, lever-
age, short selling and investments in a variety of different asset classes and
instruments, risk management has become one of the most important elements
(and most difficult challenges) of the AIS investment process. Risk management is
key to achieving high future institutional asset inflows, for without effective risk
control, pension funds, endowments and other institutional investors will resist
increasing their allocations to AIS. Several recent surveys1 indicate that Hedge
fund managers themselves are growing more aware of the importance of risk
management practices.

LTCM: What can go wrong for Hedge fund investors

Despite the ongoing changes in investors’ expectations regarding transparency
and sophisticated risk management mentioned earlier, the ‘black box’ investment
paradigm remains surprisingly persistent within the AIS community. This includes
the management of multi-strategy portfolios (‘fund of funds’). Many investors and
AIS allocators are excluded, or exclude themselves, from knowledge about the
strategy details and the particular holdings in an AIS manager’s portfolio. This can
lead to severe risks for the investor, as illustrated by the story of LTCM.
In September 1998 the failure of the Hedge fund Long-Term Capital
Management (LTCM) is said nearly to have brought down the world financial
system. The losses LCTM incurred were so large that the Federal Reserve Bank
took the unprecedented step of initiating the bailout of a private investment
vehicle, as the fear spread that forced liquidation would cause global financial
turmoil. Something very fundamental had gone wrong.2
During earlier years the fund had made very handsome returns with its core
Fixed Income Arbitrage strategy, described as ‘Convergence Arbitrage’. The man-
agers at LTCM had placed a large amount of money in ‘convergence spread
trades’ involving European interest rates within the European Monetary System.
The most prominent example had been buying Italian Government Bond (BTP)
futures and selling short German Bund contracts. Some other smaller trading posi-
tions involved yield curve Relative Value spreads and Japanese Government Bond

.... 5 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 6

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

swap spreads. Their strategy was clearly defined and paid off handsomely. By the
end of 1997 the fund paid back a significant amount of money to investors (about
one-third of its asset base of several billion US dollars). The original core strategy
had clearly lost most of its edge; the yield spread between Italian and German 10-
year government bonds had narrowed from about 550bp in early 1993 to only
about 20bp by the end of 1997. The fund managers were looking for other
opportunities and correspondingly found themselves engaged in a wider spectrum
of strategies including Merger Arbitrage, Selling Short volatility, Mortgage-Backed
Securities Arbitrage etc. Furthermore, in order to continue generating the attrac-
tive returns of the past, the fund increased its leverage substantially (from about
19 at the end of 1997 to about 30 in early 1998, and 42 in the summer of 1998).
Neither the style drifts nor the increase in leverage had ever been communi-
cated to investors. By September 23, 1998 (the day of the bailout), the fund had
lost 92% of its asset year to date and the leverage had gone up to about 120. The
excessive leverage taken by the fund remained undetected until the fund had
already lost most of its capital. The managers of LTCM had clearly shifted its
investment practice in the course of the months before the disaster. Investors had
no knowledge and understanding of the strategy LTCM was following.

Why effective risk management is crucial

to realizing the benefits of AIS
For reasons of diversification, it is widely understood in today’s investment commu-
nity that, if properly included in a global portfolio, AIS can enhance the return and
reduce the risk of a global investment portfolio. Improperly used, however, AIS can
create an investment disaster. Evaluating the
■■■■■■■■■■■■ ‘risk dimension’ is critical for realizing the
Improperly used, AIS return and diversification benefits of AIS.
The challenges of AIS risk management are
can create an
twofold: complexity and rapid change. AIS are
investment disaster
much more complex and varied than tradi-
■■■■■■■■■■■■ tional asset classes and AIS risk management

.... 6 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 7

1 ■ Introduction

requires a thorough understanding of many different underlying strategies. Yet

these strategies are changing even as investment professionals and risk managers
struggle to understand them. To make matters worse, the overall risk management
practices of the investment community are also rapidly changing across all asset
classes. The ‘state of the art’ in financial risk management has developed dramati-
cally over past years, with new paradigms and ever more complex models
continuing to emerge.
While confusing to some investors these new tools create new opportunities to
monitor and ‘fine tune’ risks in AIS investments much more accurately than even a
few years ago. Active risk management can add tremendous value to the investment
and asset allocation process of AIS investors and managers. I believe that this book
will provide readers and investors with the knowledge needed to reap these rewards.

A new investment paradigm

Much negative coverage has been dedicated to Hedge funds by the media. This is
mainly due to a mixture of myth, misrepresentation and the large scale of a few
Hedge fund failures and their global implications. The fact is that a detailed
understanding of the various strategies, a thorough manager due diligence process
and systematic third party monitoring and risk management will eliminate much
of the risk that has led to past problems.
While the ‘black box approach’ still underlies much of AIS investing today, the
‘transparency paradigm’ that I will elaborate on throughout this book is, in con-
trast, characterized by:

■ A detailed understanding of individual managers’ strategies and their risks.

■ Transparency in respect of the activities of each individual manager in the
portfolio and frequent disclosure of the aggregated portfolio exposure to the
end investor.
■ High investment liquidity. Most, but not all, AIS managers trade instruments that
are traded on public exchanges that provide high (in many cases daily) liquidity.

.... 7 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 8

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Systematic and continuous monitoring of open positions and measurement of risk.

■ Active management of risk. Note the difference between measuring and
managing risk: risk management entails using the results of risk analysis to
allocate risk optimally among different assets/trading strategies.

A ‘managed account’ structure is the most effective means of achieving maximum

investment transparency for AIS (the concept of managed account is explained in
Chapter 7).

The AIS investment approach and integrated risk management

for multi-manager portfolios (‘fund of funds’)
An increasing number of fund of funds managers have emerged who specialize in find-
ing the most interesting and best performing managers and thus diversifying the
traditional ‘manager risks’ of AIS investing. A fund of funds approach, if properly exe-
cuted, can further provide the security created by continuous portfolio management
and monitoring of managers. The added value of a fund of funds is realized provided
the fund of funds manager fulfils some fundamental criteria in his investment approach
regarding strategy allocation, manager due diligence and monitoring capabilities:

1 Sector allocation (top down analysis): Allocation of capital to AIS sectors. The
goal is to invest in the right strategy sector at the right time and to achieve the
appropriate level of diversification. This requires a sound understanding of the
individual strategy sectors, their general risk factors and risk levels as well as
their correlation features in various market environments. Chapter 3 looks at
AIS sectors in depth and provides insight into their general sources of returns
and most important risk factors.
2 Manager evaluation (bottom up analysis): Detailed examination of the
individual trading managers’ strategies and a thorough manager due diligence
process. The investor should understand the strategic edge and competitive
advantage of individual trading managers in great detail. He should also have
a sound understanding of the firm’s structure and evaluate the integrity of key
personnel. Just looking at past returns of trading strategies is insufficient. One

.... 8 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 9

1 ■ Introduction

must understand the general economic reason why and under what
circumstances a strategy shows inherent returns to the investor. Chapter 7
provides a description of the manager due diligence process.
3 Continuous monitoring/risk assessment: P&L, exposure and risk evaluation of the
portfolio. A prerequisite for continuous monitoring and risk assessment is
transparency. This enables the allocator to identify potential style drifts quickly
(i.e. the manager follows a different strategy than formerly indicated) including
undesired market bets that do not match the desired risk profile of a strategy.
Leverage controls and risk limits can be implemented and enforced efficiently
(e.g. VaR, stress test and leverage limits) and undesired risks can thus be
eliminated in time. Ongoing analysis allows the investor better to understand the
core strategy’s behaviour in different market circumstances. The fund of funds
manager’s performance expectations for the strategy can be compared to its
actual P&L and risk profile at different times and action can be taken quickly if
necessary. The anticipation of market conditions that would cause the manager’s
edge to disappear allows the allocator to exit the strategy in time. Chapter 6
provides an overview of risk analysis tools available today and Chapter 7
describes the process of active AIS risk management in detail.

The first two elements represent ‘pre-investment risk management’, while the
third element represents ‘post-investment risk management’. I refer to these three
elements as the ‘three dimensions of active and integrated risk management for
AIS investments’. It is important to note that all three, sector allocation (point 1),
manager due diligence (point 2) and transparency (point 3) are essential to AIS
risk management; one cannot replace the other.

Is transparency achievable in AIS investments?

Despite growing investor awareness of the importance of transparency, there
remains a surprising degree of resistance to such transparency in the AIS industry.
Three main arguments are frequently used against transparency and frequent dis-
closure of trading positions. AIS managers often bring up the first two arguments
and many allocators (fund of funds managers) raise the third point:

.... 9 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 10

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

1 Confidential position information will reach the market place, potentially

causing the manager to: (a) lose his edge if more players adopt the same
approach; and (b) be actively traded against by certain market players.
2 Investors lack the skill to evaluate the massive amount of information
associated with disclosure of positions. This could lead to investors being
overwhelmed by information and/or feeling a false sense of comfort.
3 Requiring transparency will eliminate the opportunity to work with the best
managers within the universe of Hedge funds. It is argued that for a variety of
reasons, including point 1, the best managers will not disclose their positions.

With respect to the first argument, one must consider who actually poses a threat
to AIS managers. This threat comes mainly from the dealer community and pro-
prietary trading desks within large investment banks rather than from fund of
funds managers or individual investors. The prime brokers, most of whom have
large proprietary trading facilities in house, do request and receive full disclosure
of all positions (and ‘Chinese walls’ are sometimes less secure than is desirable).
There is thus no reason why investors should be excluded from the same level of
information. Once Hedge fund and Managed Futures managers know who their
investors are and what their intention with the disclosed information is, they can
set up confidentiality agreements related to such information. Thus the positions
can be kept confidential while still providing the necessary transparency to the
multi-manager fund or the investor directly.
The second argument neglects the increasing expertise and capabilities of AIS
fund of funds managers. If the allocator has a sufficient understanding of the under-
lying strategies, downloads with positions and transactions can be evaluated very
efficiently. With the advent of information technology, the compilation of large
quantities of data has become quite feasible for sophisticated investors and profes-
sional portfolio managers. A wide variety of tools and software packages for
sophisticated risk management is now available. Risk management experts within
the team of the multi-strategy fund of funds manager can deal with the complex job
of interpreting the disclosed information and therefore tremendously increase the
benefits of transparency.

.... 10 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 11

1 ■ Introduction

Multi-strategy fund managers frequently raise the third argument, claiming

that the best Hedge fund managers (which are, in the view of many, the largest)
will not provide transparency or insight into
their trading. It is therefore argued that fund of ■■■■■■■■■■■■
funds managers requiring transparency (i.e. fre- AIS managers with
quent disclosure of positions) are left with
strong past
lower performing trading advisors. This state-
performance but
ment bears little truth. For the large majority
of strategies, good performance has nothing to refusing to provide
do with lack of transparency. In fact, many transparency do not
high-quality ‘first tier’ managers are today will- necessarily present a
ing to offer transparency if asked or required better investment
(for an investment) to do so. Contrariwise, AIS ■■■■■■■■■■■
managers with strong past performance but
refusing to provide transparency do not necessarily present a better investment to
the investor. Often, a non-transparent strategy corresponds to a manager who is
unable to illustrate his edge and therefore hides behind a ‘black box’ approach. In
other words, a manager who refuses to explain his edge may not have one!
Many investors regard AIS as an industry in which returns are generated by
mysterious means and judge successful Hedge fund managers only by their stellar
past returns. The incorrect belief that better performing managers must operate in
secret is linked to this persistent misperception about AIS. In fact, most managers
follow systematic investment strategies that are understandable if studied suffi-
ciently and that, not surprisingly, perform better in certain market environments
than in others. Past performance is an insufficient indicator of the future potential
of a strategy (or as the statement on most disclaimers for publicly offered funds
puts it: ‘Past returns are not indicative of future results’). Rather, transparency is
needed in order to allow for an adequate and ongoing assessment of risks and
rewards. Considering the additional risks of AIS investing when transparency and
investor control are absent, one must question whether institutional investors
fulfil their fiduciary responsibility when they invest in a ‘black box’.
Examples of non-transparent and, for certain periods of time, very successful strate-
gies are LTCM, Quantum Fund (G. Soros), Tiger (J. Robertson) and Niederhoffer, all

.... 11 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 12

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

of which failed spectacularly in the end. Even the most brilliant investors in the past
are not protected against losses and drawdowns, as many investors in Hedge funds
have learned, to their detriment, in the past. Investors must learn to look beyond past
return and instead look at how the returns have been achieved. Conditions for success
are constantly changing. Recent studies have shown that there is little convincing evi-
dence that winning funds repeat in a way that can be exploited,3 and have also shown
that small and young programs show their best performance in the first two years.4
The appropriate level of disclosure to investors and institutional asset managers
(fund of funds) is the subject of ongoing discussion within the industry. A working ses-
sion of the Investor Risk Committee (IRC) was held on the topic,‘What is the right
level of disclosure for alternative asset managers?’.5 One of the conclusions was that
risk monitoring and style drift monitoring were among the most important objectives
of disclosure (see Chapter 7 for a more detailed discussion of the IRC report).
Transparency can take a number of different forms, from regular conversations with
managers about their strategies (the weakest form) to obtaining a daily download of all
positions from a manager’s prime broker (the strongest form). There is much discus-
sion around the question of whether aggregated ‘risk information’ is sufficient for risk
management purposes versus requesting disclosure of all positions. The belief that AIS
risk can be adequately monitored without obtaining underlying positions is wide-
spread. I disagree with this view. Chapter 7 will provide a more detailed discussion of
this issue. One may argue that for some strategies disclosure of individual positions
may not be absolutely necessary.6 But in most circumstances detailed position informa-
tion is the only way to provide the information necessary for the risk-monitoring task.
The level of information that should be provided to the investor or fund of
funds manager also depends on the investment style of the individual manager as
well as the strategy sector in which he is operating. There are different degrees of
usefulness of transparency for the various strategy sectors. Model-based
Systematic strategies and Arbitrage strategies are easier to monitor and understand
on a daily basis than discretionary Long/Short Equity, Macro and Short Selling
strategies, where positions are more difficult to comprehend without further
manager-provided information. For Relative Value strategies (Fixed Income
Arbitrage, Convertible Arbitrage, Equity Market Neutral) and Event-Driven

.... 12 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 13

1 ■ Introduction

strategies (Merger Arbitrage, Distressed Securities, Convertible Debenture

Arbitrage) transparency can be very useful, as leverage, instrument liquidity and
potential style drift are important issues for these strategies.

Liquidity of AIS investments

Transparency is of most value when combined with the appropriate level of liquid-
ity. Risk management has to be proactive and the risk manager should be in a
position to take steps to remedy critical situations in timely fashion. Often, when a
crisis has arrived, it is too late to make adjustments. If, for his investment with a
single manager, the AIS allocator faces a redemption period that does not corre-
spond to the level of provided transparency, he is prevented from responding to
time-sensitive information as he cannot mandate immediate adjustments to invest-
ment positions.
Some AIS investors prefer high liquidity, i.e. short redemption periods, while
others are willing to assume liquidity risks, i.e. accept extended redemption periods,
in pursuit of attractive returns. As the AIS industry grows, and the spectrum of
investors becomes more heterogeneous, requests for liquid multi-manager products
are increasing. Currently, the industry offers three types of multi-manager AIS prod-
ucts to investors: open-ended funds with redemption periods ranging from one
month to six months are the most numerous. Second, structured notes are increas-
ingly offered, especially in Europe, for the purpose of circumventing the regulatory
requirements of listed funds. They are usually traded in a secondary market pro-
vided by the issuer. Finally, a number of closed-end investment vehicles wrapped as
investment companies and listed on exchanges have been set up.
All these structures result in liquidity problems for investors. Besides the long
redemption periods for open-ended funds and structures notes, additional factors
can lead to a significant increase in the time span between redemption and the
receipt of monies. Settlement problems do not exist for an exchange-traded
investment company (as the instrument can be traded on an exchange on a daily
basis), however, other problems render these products unsuitable for most
investors. Due to the lack of a broad market, these instruments are not traded

.... 13 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 14

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

very actively, i.e. their liquidity is extremely low. Larger sizes cannot be sold with-
out a severe negative price impact. This usually leads to a significant discount in
the trading value compared to the NAV. Since most AIS managers invest in highly
liquid instruments, one must ask why there cannot be an AIS investment fund
with daily liquidity based on NAV which is as easy to buy and sell as any tradi-
tional mutual (equity) fund.7

The challenges of AIS risk management

There is not, as yet, a ‘risk management standard’ in the AIS industry, but generally,
the management of AIS risks goes beyond quantitative methods and includes essen-
tial qualitative assessments. The ‘art and science’ of AIS risk management is
developing as this book is being written and the discussion about proper tools and
approaches is ongoing. Some types of risk (e.g. market risk, credit risk) are today
easier to quantify and manage than others (e.g. operational risk, model risk),8 but
available tools and models for risk management are subject to constant change. The
AIS industry itself has started to become an important target for risk management
tools as well as a pioneer in the development of
■■■■■■■■■■■■
new risk analysis standards.
AIS risk management Compared to traditional investments (bonds,
can only be successfully equities), the risk factors of Hedge funds and
performed if the Managed Futures strategies are quite complex.
risk manager has The spectrum of AIS is very broad and stretches
along a wide universe of investment strategies
a thorough
and asset classes. Strategies earn their returns in
understanding of
a variety of ways and are exposed to different
trading strategies
types and degrees of risk. AIS risk management
■■■■■■■■■■■■ can only be successfully performed if the risk
manager has a thorough understanding of trad-
ing strategies on the strategy sector level as well as on the level of the individual
manager. I will provide the reader with the building blocks necessary to understand
and assess sources of risk and return of the various strategies. I will further demon-

.... 14 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 15

1 ■ Introduction

strate why it is essential to move beyond risk monitoring to active risk control and
will show how to integrate risk management into the asset allocation process itself.
Risk technology has become computationally fast, efficient and considerably
less expensive than a decade ago. Similar to the current development towards
‘institutionalization’ of the investment process in the AIS industry, one can antici-
pate a trend towards standardization of AIS risk management tools. Next to a
detailed knowledge of the strategy sectors and careful manager evaluation, active
risk management will become an essential element of AIS investing.

Notes
1. Capital Market Risk Advisors (www.cmra.com/html/hedge_fund.html) also
published in the AIMA Newsletter, Feb 2002. HedgeMar (Dec. 2000), the Investor
Risk Committee and the Hennessee Hedge Fund Advisory Group (‘Transparency In
Any Form Is In Demand’, by S. L. Barreto, HedgeWorld.com News on Nov. 27, 2000,
under http://www.hedgeworld.com/news). Recent publications of industry surveys
include the Barra Strategic Consulting Group FOHF market survey (2001) and the
Goldmann Sachs and Frank Russell Alternative Investment survey (2001).
Furthermore, a group of five Hedge fund managers including Soros Management LLC
issued the report ‘Sound Practices for Hedge Fund Managers’ in February 2000 as a
response to the President’s Working Group report on financial markets after the
collapse of Long-Term Capital Markets (‘LCTM’) (see Chapter 6).
2. For a detailed discussion of the LTCM bankruptcy, see Ph. Jorion, ‘Risk
Management Lessons from Long-Term Capital Management’, downloadable from
http://www.gsm.uci.edu/~jorion/research.htm; also ‘Hedge Funds, Leverage, and
the Lessons of Long-Term Capital Management’, report of the President’s Working
Group on Financial Markets, April 1999, on http://risk.ifci.ch/146530.htm.
3. See, for example, the following articles: M. Peskin, M. Urias, S. Anjilvel and B.
Boudreau, ‘Why Hedge Funds Make Sense’, Quantative Strategies, Morgan Stanley
Dean Witter, November 2000; and ‘The Young Ones’ by Crossborder Capital,
Absolute Return Fund Research, April 2001.

.... 15 ....
8409 Chapter 1 p1-16 11/4/02 1:12 PM Page 16

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

4. See reference in footnote 3. A more general audience is addressed in the

following article: ‘The Big, the Bold, and the Nimble’, The Economist, Feb. 24,
2001, p.87.
5. ‘Hedge Fund Disclosure for Institutional Investors’, available on the IAFE web
page http://www.iafe.org. The IRC report from October 2000 was updated in
July 2001.
6. The IRC members state that full disclosure does not necessarily allow the
portfolio managers to fulfil their monitoring objectives. Reporting of summary
risk, return and position information can be sufficient, the report states.
7. Currently, I am aware of only one such fund, which started in October 2001
(Bank Hofmann/saisGroup – now Partners Group).
8. The current discussion concerning the amendment to the BIS rules for capital
charges reflects the dynamics of the risk management discussions in the financial
industry very illustratively.

.... 16 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 17

CHAPTER 2

The Universe of Alternative

Investment Strategies (AIS)

The spectrum of AIS is very broad, as it stretches along an extended and heteroge-
neous universe of investment strategies, asset classes and instruments. It is essential
for an AIS risk manager to have a detailed understanding of this wide range of
strategies and investment styles. In this chapter, I define the AIS universe and pro-
vide an overview of the characteristics and the historical development of AIS as an
asset class. I also introduce the fund of funds concept for AIS. Chapter 3 then
discusses in detail each AIS sector in terms of its investment and risk characteristics.
Chapter 5 presents a general description of AIS risks, a quantification of the various
risks and the most important risk management guidelines for each strategy sector.

Definition of the AIS Universe

Due to the complexity and heterogeneity within the industry there is an ongoing
debate about how to define the AIS Universe.1 Nonetheless, a certain general clas-
sification scheme has emerged, which I present in Figure 2.1.2 ‘Alternative
Investment Strategies’ (on the left-hand side of Figure 2.1) form a subset of the
global ‘Alternative Investment’ Universe, which consists of all investments beyond
traditional bond and equity investments. AIS are commonly referred to as ‘skill-
based’ or ‘absolute return strategies’. In contrast to traditional stock and bond

.... 17 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 18

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

investments, returns are often unrelated to developments in the broader financial

market (this does not apply for all strategies by far, however).
I have excluded Private Equity from the ‘absolute return’ (AIS) universe on the
grounds that its investment, liquidity and correlation characteristics set it apart from
the AIS Universe discussed here. Some do consider Private Equity as part of the AIS
Universe,3 and then distinguish Hedge funds and Managed Futures from Private
Equity by categorizing them as ‘liquid AIS’. I do not follow this classification, as the
degree of liquidity varies extremely even between the different Hedge fund and
Managed Futures strategies. Further, liquidity (or the time horizon of investment) is
not the only factor that sets Private Equity apart from Hedge funds and Managed
Futures. Another factor is the smaller degree of strategic flexibility for Private
Equity investments (there is no short selling and only little derivatives trading
involved; it is basically a ‘buy and hold’ strategy). Another interesting distinctive
feature is also the degree of ‘efficiency’ of the markets that Private Equity and
Hedge fund managers operate in. Private Equity markets (similarly to real estate)
are not information efficient, i.e. research and informational advantages lead to
above average returns. Most Hedge fund strategies, in contrast, operate in markets
with a generally higher degree of efficiency (equity, FX, fixed income, commodities,
see later discussion). The sources of AIS return are thus much more diverse.
AIS can be further divided into two main categories: Hedge funds and Managed
Futures. Hedge funds invest in a variety of different asset classes (including equity,
fixed income and foreign exchange) on both a directional and a non-directional
basis. Hedge funds take advantage of their great flexibility regarding asset classes,
trading styles, markets, leverage, short selling and liquidity. They may hold long
and short positions and many strategies employ leverage through borrowing, the
level of which varies greatly among different strategy sectors. Managed Futures
programs are investment entities that assume long and short positions in exchange-
traded derivatives, in particular Futures and Options on commodities and
‘financials’ (equity, fixed income and foreign exchange). Most Futures strategy
managers are registered with the National Futures Association (NFA) and the
Commodity Futures Trading Commission (CFTC) as ‘Commodity Trading
Advisors’ (CTA) and/or ‘Commodity Pool Operators’ (CPO).

.... 18 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 19

2 ■ The Universe of Alternative Investment Strategies

Alternative Investments
(AI)

AI AI ‘Traditional’
Strategies Assets AI

High Yield Bond

Securitized Lower Credit/
Hedge Funds Managed Futures Physical Assets Private Equity Quality Bonds
Products
Emerging
Early Stage: Seed Markets
Land CCABS
Relative value Opportunistic (Farmland/Timberland)
Financing
(Credit Cards)
(Idea and Research)
REIT
Early Stage: Start-up Real Estates
Real Estate CLO
Event Driven Passive (Retail/Commercial)
Financing
(Loans) Investment Trust
(Product Development)

Early Stage: 1st stage

Commodities CMO
Opportunistic (Gold, Silver)
Financing
(Mortgages)
(Initial sales/first prod.)

Expansion: 2nd stage

Art CBO
Financing
(Paintings, Handicraft) (Bonds/Credit)
(Establ. prod. line)

Expansion: 3rd stage

Coins, Stamps Financing Tax Liens
(Major expansion)

Expansion/Mezzanine
Factoring/
Wine Financing
Forfeiting
(Expected IPO)

Buyout (LBO, MBO, MBI) Cat-Bonds/Notes

Financing (Hurricane/earthquake)

Turnaround Insurance Linked

Financing (Airspace programs)

FIGURE 2.1 ■ Alternative Investment Strategies in the global universe of Alternative Investments

AIS funds are typically organized as limited partnerships or limited liability com-
panies and are often domiciled offshore for tax and regulatory reasons.4 Another
characteristic of AIS is the way the investment manager is compensated, which
mostly occurs on two levels: an annual management fee, plus an additional per-
formance-based fee. This serves the purpose of aligning the manager’s interest
with that of investors’. It is worthy of note that many managers allocate a signifi-
cant amount of their personal net worth to their own funds in an attempt to
demonstrate commitment both in the pursuit of returns and exposure to risk.

.... 19 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 20

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Development of AIS
The origins of some strategies within the AIS Universe date from well before the
terms ‘Hedge funds’ and ‘Managed Futures’ entered into the investment vocabulary.
The large consolidation wave in the railway, oil and financial industries in the late
19th century created an attractive environment for speculation and arbitrage on merg-
ers. Convertible Arbitrage strategies were particularly attractive and performed well in
the years 1929–32 during the stock market crash. Short Selling and Distressed
Securities investing have existed since the late 19th century. However, these invest-
ment activities were largely pursued as isolated trading activities by individuals.5 The
systematic application of these strategies within an investment vehicle offered to third
party investors is a relatively modern phenomenon and emerged in the early 1950s.
Unfortunately, the name ‘Hedge fund’ is somewhat misleading. In fact, most
Hedge funds are leveraged rather than hedged. Further, most ‘Hedge funds’ are
technically not ‘funds’ but Limited Partnerships. The original understanding of a
Hedge fund was an equity investment strategy where managers reduced their expo-
sure to adverse downward movements in the broad market by combining long and
short positions in stocks. The investment manager bought stocks he believed to be
undervalued and then sold short other stocks he considered overvalued. Today this
strategy goes by the name ‘Long/Short Equity’. It was A. W. Jones who created the
first Hedge fund of this kind in 1949 when he combined the purchase of stocks with
Short Selling and leverage, two main elements of Hedge fund strategies today
(derivates, a third element, was not yet widely available). Furthermore, he charged a
performance fee to his investors, which enabled him to benefit directly from his
investment success. This became another common feature of AIS.
The first large Hedge fund boom started with an article about the Jones’ strat-
egy in Fortune magazine in the mid-1960s.6 While this boom died off quickly in
the bust years of the late 1960s and early 1970s, another Hedge fund strategy
emerged in the late 1960s, ‘Global Macro’. This strategy entails ‘taking sophisti-
cated bets’ on probable future price moves of specific instruments based on
particular macroeconomic views. The Global Macro strategy is connected with
two names that for many years were the symbols of Hedge fund investing: Julian
Robertson and George Soros. Both showed high returns over almost three decades

.... 20 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 21

2 ■ The Universe of Alternative Investment Strategies

and gained wide public attention through an article about Julian Robertson’s fund
in Institutional Investors magazine in the mid-1980s7 and through the British
pound opting out of the European Currency System in 1992, which, it is widely
believed, was caused by George Soros’ ‘Quantum’ Hedge fund (and which created
large profits for ‘Quantum’).8
Futures predated equity markets, but it was not until the late 1960s that the use
of Futures and other derivatives emerged within diversified trading strategies.
Managed Futures strategies were born around the same time as Hedge funds.
Richard Donchian created the first Futures-based investment program in the same
year (1949) as A. W. Jones launched his first Hedge fund. Today the distinction
between the two is somewhat blurred and some actually no longer distinguish
Futures from Hedge funds. In 1965, Dunn and Hagitt started trading commodity
Futures using technical trading systems (they also offered the first offshore commod-
ity pool in 1973). A first boom in Managed Futures investment programs occurred
with the introduction of financial Futures in 1972 and the increasing availability of
computing power in the 1970s. Most of the investment programs were based on
technical trading and charting systems. Managed Futures quickly came to be viewed
as an interesting alternative investment class with attractive returns uncorrelated to
returns in equities (which were rather modest in the 1970s). Traditionally,
Commodity Trading Advisor (‘CTA’) funds are distinguished from Hedge funds on
the simple notion that they are limited to trading primarily Futures contracts and
that they are registered with the CFTC and comply with its regulatory rules. But
nowadays, many CTAs also trade in OTC securities markets, while Hedge funds
also use Futures as essential risk management tools (some Hedge funds are or used
to be registered with the CFTC, e.g. Long-Term Capital Management – LTCM).
The AIS industry recovered strongly from problems related to the rapid increase of
interest rates in early 1994 and the crisis following the Russian Bond default and the
liquidation of LTCM in 1998. The time period after 1998 can be referred to as the
‘institutionalization phase’. The AIS industry’s growth in 1999–2002 was enormously
supported by falling equity markets and the ‘NASDAQ crash’. The AIS industry is
now so far developed that many investors consider it as an asset class itself. On the
demand side, due to the attractive risk–reward characteristic as well as their low cor-
relation to traditional investments, institutional investors have increasingly expressed

.... 21 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 22

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

interest in AIS.9 On the supply side, the AIS industry has been and will continue to be
a lure for the most intelligent talents in finance. The smartest finance professionals
and most promising investment ideas are attracted to an industry that offers greatest
flexibilities for the implementation of investment and hedging strategies together with
a very high level of monetary compensation. Most banks and other large finance insti-
tutions have begun to offer a diverse range of AIS investment structures,10 as they
come to view AIS as a new and increasingly profitable business segment.
The question, whether AIS constitutes an asset class in itself or whether Hedge
funds and Managed Futures only extend the range of investment strategies within
certain existing asset classes, is subject to debate and is mostly a matter of perspec-
tive. Investors increasingly consider AIS as a separate class in their asset allocation
process (sometimes together with Private Equity investments). On the other hand,
Hedge fund and Managed Futures programs do not trade any particular new assets
or instruments but rather execute certain investment strategies within a set of exist-
ing instruments and asset classes. They can be seen as the active counter-party to
passive (i.e. index-linked) investment strategies in a ‘core–satellite’ portfolio set-up.
Sceptical market participants and investors have recently compared the devel-
opment of Hedge funds with the technology bubble in the 1990s and predict that
the current AIS euphoria will similarly end in tears.11 In the most simple sense, a
investment bubble is a phenomenon that builds up when expectations skyrocket
and everyone does the same thing at the same time.12 The heterogeneity of AIS
clearly contrasts with historical bubbles such as the Dutch tulip mania in the 17th
century, the US equity markets in the late 1920s, the Japanese stock market in the
late 1980s or the internet hype in the late 1990s. The AIS industry covers a very
broad range of asset classes and favourable and unfavourable market environ-
ments deviate strongly across strategy sectors. Economic developments, political
events and changes in the market environment create and destroy different profit
opportunities for different strategies. The AIS industry in its entirety is sufficiently
well diversified to deal with extreme market circumstances.
The opposite view is that Hedge funds are a new asset class that has a legiti-
mate place in every investment portfolio. The main argument underlying this view
is that AIS have strong absolute returns and low correlations to traditional asset
classes. But this might turn out to be new wine in old wineskins. A few years ago

.... 22 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 23

2 ■ The Universe of Alternative Investment Strategies

investing in emerging markets was marketed as a new way of decreasing overall

portfolio risk. But experiences in the 1990s (Mexico, Thailand, Russia) have
aligned the hype with reality. The diversification benefits of AIS might also be
overestimated, as the AIS industry has had a long equity bias in recent (equity bull
market) years and it is debatable whether the AIS industry can decouple com-
pletely from global economic trends. Further, given the strong inflows into Hedge
funds, one has seriously to ask whether return expectations are decoupling from
reality. Lower absolute Hedge fund performance achieved in 2000 and 2001 may
help gradually to align expectations with reality. AIS investing will increasingly
require strong skills on the side of the allocator (fund of funds manager, direct
investor) in order to realize the benefits of Hedge funds and Managed Futures
while avoiding excessive risk.

Distressed Emerging Equity Managed Convertible Event Relative

Sec. Markets Non-Hedged Futures Arbitrage Driven Value
2.7% 3.2% 15.9% 5.3% 2.8% 10.9% 11.1%

Equity
Short Market Equity Market Equity
Selling Reg D Macro Neutral Timing Hedged (L/S)
0.1% 0.4% 12.9% 5.2% 0.7% 29%

FIGURE 2.2 ■ AIS asset allocation by strategy sector

Source: Hedge Fund Research

.... 23 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 24

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Figure 2.2 displays the distribution of how assets are approximately invested in
the different sectors (as of October 2001). Long/Short Equity is the dominant
sector with more than 40% of all AIS assets invested in this strategy sector (Equity
Hedged 29% plus Equity Non-Hedged 16%), followed in roughly equal size
(11–13%) by Event Driven, Relative Value and Global Macro strategies. Futures
strategies and Equity Market Neutral each take about 5%, Convertible Arbitrage,
Distressed Securities and Emerging Markets about 3% each. Convertible Arbitrage
and Equity Market Neutral are Relative Value strategies, but they are counted sep-
arately here. Other strategies like Short Selling, Regulation D and Equity Market
Timing fall below the 1% range. Note that these numbers depend on the classifi-
cation scheme chosen (in this case by Hedge Fund Research).

Understanding the sources of AIS returns

It is widely understood among investment professionals that Hedge funds and
Managed Futures generate investment returns that are significantly more attractive
than average equity and bond investments measured on a risk-adjusted basis using
commonly available quantitative risk assessment tools. This creates some confusion
and scepticism, as most investors believe that financial markets provide no ‘free
lunch’ to investors. It is important to realize that Hedge funds and Managed Futures
are in fact exposed to a variety of different risk factors and thus possess a corre-
sponding number of return sources. In many cases the main part of the answer lies
in the inability of conventional risk measures and theories to measure the diverse
risks of AIS and to describe the corresponding sources of return properly.
Conventional finance theory as described by the Capital Asset Pricing Model
(CAPM) and similar asset pricing models states that expected investment returns are
directly related to the amount of market risk taken (e.g. the risk of the broad equity
market falling). The excess return over a risk free investment is linked to the ‘beta’ of
the investment. Any excess return above and beyond the return for taking risk is
considered the result of particular manager skills (or pure luck), e.g. detecting mis-
priced securities, having superior and price relevant information, or correctly timing
the market. This return is called ‘alpha’. Consistent alpha generation appears to contra-
dict a well-established paradigm in financial market theory, which is the ‘efficiency

.... 24 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 25

2 ■ The Universe of Alternative Investment Strategies

market hypothesis’ (EMH). The EMH comes in various forms related to different
types of information available to investors: a weak form, a semi-strong form, and
strong form.13 The EMH (in its strongest form) states that there is no price relevant
information available to any investor that is not yet reflected in market prices. This
implies that investment managers will not consistently generate alpha. Most investment
and academic professionals do not hold the EMH in its strong form for true, but its
weak and semi-strong form has more numerous supporters. The market efficiency
battle between proponents of standard finance and their counter-parties (e.g. advocates
of behavioural finance) is waged over the interpretation of price anomalies and consis-
tent alpha generation in the main equity, foreign exchange and fixed income markets.14
But it is rather undisputed that the overall global spectrum of financial markets
presents varying degrees of efficiency. The major foreign exchange markets, G7
Government Bond markets as well as the large capitalization segment of the major
international equity markets are generally considered to be quite efficient, while real
estate and private equity markets generally display a much lower degree of efficiency,
i.e. superior information or skill in these markets pays off in above average returns.
Hedge fund managers operate in security markets with various efficiencies and are thus
in some middle position between public equity and bond portfolio managers on the
one side and private equity or real estate experts on the other side.
For a further understanding of the ‘battle of alpha’, it is important to assess the
basic assumptions of common asset pricing models. The CAPM is based on the
following (and some other) important assumptions:15

■ Investors choose investments according to a mean variance framework (i.e.

they measure reward by the mean return and risk by the variance – or
standard deviation – of returns). Investors are generally risk averse and have a
quadratic utility function with respect to risk.
■ All investors have the same forecast of expected return variances and
correlations. This leads all investors to hold the same risky market portfolio
(with varying weights relative to the risk-free part of their portfolios,
depending on their risk profiles). Consequently, there is only one source of
risk for which investors are rewarded, which is the ‘broad market risk’ (as
measured, for example, by an equity index).

.... 25 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 26

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Trading is frictionless, i.e. there are not transaction costs, taxes, etc. Investors
can (and will) sell short securities without any restrictions.

The first assumption becomes questionable when the probability distribution of

investment outcomes is skewed (i.e. non-symmetric) or leptohurtic (i.e. exhibits ‘fat
tails’). In this case, the conventional measure of risk, namely the standard deviation,
provides an imprecise basis for risk measurement. Some AIS – comparable to option
strategies – have non-symmetric return distributions. The fact that investors have a
preference for positively skewed outcomes and an aversion against negatively skewed
outcomes is not captured by a risk measure that weighs each part of the distribution
identically. For many AI strategies, the right amount of return for the given exposure
to risk cannot be unambiguously determined in the framework of CAPM.
The second point is plainly false, as investments, particularly those in AIS, are
subject to numerous other risks beyond broad market risk. The ‘Arbitrage Pricing
Theory’ (APT) provides a more general framework,16 which allows the inclusion
of more factors to explain asset returns. The crucial question of which particular
factors to include remains a matter of dispute.17 The numerous extensions of the
one-factor CAPM usually include additional fundamental factors such as firm size
or value factors (price-earnings ratio, price-book value, etc.) or macroeconomic
factors (such as unexpected inflation or credit spreads).
Considering only market risk leads to the incorrect conclusion that superior
returns are always the result of unique manager skills or superior access to infor-
mation. But the reality is that investors who are willing to assume risks beyond
market risk (e.g. liquidity risk, credit risk, event risk) can earn additional returns
unconditional on superior information or skill. The investor might incur losses
from bearing one of these risks during some periods, but over time the returns
should be sufficient to make the investment profitable. These returns are called
‘risk premiums’. Multi-factor models (like APT) allow capturing these other
dimensions of risk besides overall market risk and they include the corresponding
‘factor risk premiums’ in the modelling of asset prices.
The third restriction is at best only partly true. Transaction costs are indeed an
important factor for investors to consider. Further, many investors are constrained
in selling short securities. This can create inefficiencies in financial markets that

.... 26 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 27

2 ■ The Universe of Alternative Investment Strategies

Hedge funds try to take advantage of (see the discussion of ‘Long/Short Equity’
and ‘Short Selling’ in the next chapter).
Effective AIS risk management requires a sound understanding of how returns are
generated and what risks come with these return sources. The limitations of quantita-
tive measures of risk-adjusted returns have to be considered (e.g. as in the case of
skewed return distributions). As already mentioned, it is generally useful to distinguish
the following two (real) sources of returns (besides ‘luck’): risk premiums and man-
ager skill. These two sources are by no means mutually exclusive. It takes skill to
capture a premium effectively and to manage the related risk. By the same token,
some strategies are based largely on a manager’s skill in forecasting price moves or
detecting relative mispricings, which is often performed with the help of quantitative
valuation models. One can argue that these strategies earn a ‘complexity’ (risk) pre-
mium. Further, price anomalies and apparent Arbitrage opportunities are often related
to different risk premiums.18 Despite certain overlaps and ambiguities that come with
this classification scheme, distinguishing ‘risk premium strategies’ from ‘pure skill
strategies’ nevertheless provides a good framework for an analysis of the AIS Universe.

Economic functions and Risk premiums

Investors in financial markets fulfil different economic functions and assume certain
risks in expectation of a return that is higher than (risk free) money market interest
rates. One should understand these economic functions and their role in the AIS
return generation process. The most broadly known economic functions are:

■ Capital formation, providing companies with access to capital.

■ Risk transfer, providing commercial hedgers with the possibility of transferring
unwanted risk in futures markets.
■ Price transparency and efficiency, making financial markets more efficient.
■ Providing liquidity, creating a market for less liquid investments.

I believe that many AIS earn their returns by assuming risk in a risk averse financial
world, rather than from the identification of market inefficiencies. By taking these risks
the investor is compensated with an expected return, the risk premium. I therefore refer
to these strategies as ‘risk premium strategies’. Analogously to the earnings of an insur-

.... 27 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 28

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

ance company, which is rewarded with a premium for insuring its clients that is higher
than the expected average claim per insurance contract, premiums in financial markets
are positive expected returns that exceed the ‘risk free interest rate’19 in exchange for
accepting the possibility of a financial loss. Over time, risk premiums provide an inher-
ent and permanent positive expected return, the source of which does not disappear if
spotted by other investors (although it can fluctuate over time). The nature of its under-
lying risk premium is directly related to a strategy’s risk profile. The risks and premiums
vary among different strategies. It is important to understand the economic rationales
for the premiums of each individual strategy sector. They are discussed in detail in
Chapter 3. For ‘risk premium strategies’, manager skill primarily expresses itself
through premium identification, proper timing and the appropriate risk management.
The existence of premiums as inherent sources of returns is most apparent for
Relative Value and Arbitrage strategies (Fixed Income Arbitrage, Risk Arbitrage
and Convertible Arbitrage). Note that the word ‘Arbitrage’ does not refer here to
the strictest meaning of the word (which is ‘generating a profit without risk’). In
this context, by ‘Arbitrage’, I mean ‘buying relatively undervalued securities and
selling overvalued securities’. There is a risk involved here, specifically the risk
that the undervalued securities become even cheaper and the overvalued ones
more expensive. ‘Arbitrage’ strategies earn spreads (i.e. premiums) between
market prices of two or more strongly related instruments as compensation for
taking very particular risks such as company specific risk, FX risk, commodity
price risk, credit risk, duration risk, liquidity risk and deal risk. Risk (Merger)
Arbitrage returns, for example, are directly linked to the spread between the
market price of the target company and the price offered by the acquiring com-
pany and are earned in return for taking the risk that the deal does not go
through. The following list summarizes different risk premiums and identifies for
each one of them the particular strategies attempting to capture it:

■ equity Market risk premiums (Equity Market Timing, Long/Short Equity,

Convertible Arbitrage)
■ corporate event risk premiums (Risk Arbitrage, Long/Short Equity, Distressed
Securities, Regulation D)
■ risk transfer premiums (Futures strategies)

.... 28 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 29

2 ■ The Universe of Alternative Investment Strategies

■ complexity premiums or ‘efficiency’ premiums (Equity Market Neutral,

Statistical Arbitrage, Fixed Income Arbitrage)
■ liquidity premiums (Distressed Securities, Regulation D)
■ duration risk premiums (Fixed Income Arbitrage)
■ credit risk premiums (Fixed Income Arbitrage, Regulation D, Distressed
Securities, Convertible Arbitrage)
■ FX risk premiums (FX strategies, some Global Macro strategies).

Equity risk premiums are compensation for the ‘capital formation’ function that
investors fulfil through buying companies’ stocks and thereby giving companies
access to working capital. Investors take the risk of financial loss, for example, due
to an economic downturn, less favourable earnings developments or bankruptcy.
According to the Capital Asset Pricing Model (CAPM) and related models, equity
risk is twofold. First, broad market risk is related to the volatility of the broad
market or industry sector. Second, corporate specific risk is the idiosyncratic risk of
loss due to an adverse development, which affects the stock of a particular company
(note that, according to the CAPM, idiosyncratic risk does not earn a risk premium).
Many Futures strategies, especially trendfollowing strategies, are related to an eco-
nomic function that is very different from equity investments. Investors in Futures are
willing to expose themselves to the natural risks of commercial hedgers, thereby pro-
viding those hedgers with the possibility to transfer their undesired price risks. By
fulfilling this function of risk transfer speculators in Futures markets earn a correspon-
ding premium, which could alternatively be called ‘commodity hedging demand
premium’. More details are provided in Chapter 3 in the sections on Futures strategies.
Many Arbitrage strategies earn premiums for providing market efficiency and price
transparency. Their aim is to detect pricing inefficiencies through the application of
(mostly proprietary) valuation models to complex financial instruments. It can be
argued that their returns are based on a ‘complexity’ (or, alternatively, an ‘efficiency’)
premium for taking the risk of mismodelling the underlying financial instrument and its
complexity and thus suffering a loss. Further, investors who are willing to accept lower
liquidity in their investments earn a liquidity premium. Such liquidity risk often goes
together with credit risk. Credit and duration risk premiums are connected to investing
in fixed income instruments with lower credit quality and longer maturity respectively.

.... 29 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 30

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

What are the sources of returns for FX strategies? Currency markets (at least for
the major currencies) are considered the most liquid financial markets in the world,
and the inter-bank FX market is often seen as the market that operates closest to
complete efficiency. This high liquidity has to be generated by somebody with mon-
etary incentives, which means the speculators. It can be argued that the (very strong)
liquidity request of commercial participants in FX markets generates a return source
for FX traders, as supply and demand from commercial players alone barely ever
balance each other exactly. This argument is similar to the risk transfer premium in
commodity markets. The return of FX strategies is thus a premium paid by commer-
cial market players for the generation of liquidity and price continuity (see the
discussion on currency strategies in Chapter 3 for more details). Another premium
earned by certain FX strategies is related to the risk of a (stronger than expected)
currency devaluation and expresses itself as a positive (interest rate) carry (i.e. the
differential in interest rates) between two currencies. The corresponding strategy
consists of buying a high yield currency and selling one with a low yield.
One problem for the evaluation of risk premium-based strategies is that, while
they earn returns due to the assumptions of certain risks, empirical measures of
these risks might be calculated for a time period that does not include a relevant
‘risk event’. This can lead to a severe underestimation of a particular strategy’s risk.
Risk premiums as a source of return are less (if at all) obvious for some opportunis-
tic strategies like Global Macro, Short Selling and many Long/Short Equity strategies.
The returns of these strategies are rooted in the manager’s skill in forecasting price
developments, detecting pricing anomalies and acting quickly on anticipated market
moves. The underlying opportunities and market inefficiencies are usually temporary
and quickly disappear when spotted by other investors. The greatest potential for these
‘pure skill-based’ strategies is where information is not freely available. As I mentioned
before, the distinction between manager skill and risk premium as sources of return is
not always absolutely clear. Manager forecasting skill and a ‘complexity premium’ can
both be argued to be sources of returns for some opportunistic strategies.
Ideally, financial economists would prefer to develop a universe of fundamental
risk factors that can explain the time series behaviour of AIS returns. For traditional
investments much work has been dedicated to examining the components of active
equity and bond manager performance and numerous studies have directly assessed

.... 30 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 31

2 ■ The Universe of Alternative Investment Strategies

their sources of return.20 Similarly, more recent research indicates that certain factors
(return drivers) help to explain AIS performance patterns.21 Some go even further and
argue that ‘generic’ systematic trading programs (with no discretionary or skill-based
input) can replicate most strategies’ returns (and subsequently say that the high fees
for AIS managers are not justified).22 The current academic thinking on how to evalu-
ate AIS returns is to include ‘style factors’ (e.g. option-like payoffs) in the set of
performance factors rather than just explain AIS return based on other asset returns.23
The detection of specific performance drivers for AIS strategies with the help
of quantitative tools such as factor models has been subject to intense discussion
and research in the academic and the financial community in recent years. Hedge
funds are now often classified as either ‘long biased’, i.e. primarily influenced by
the direction of international bond and equity markets (‘return enhancers’), or
non-directional attempting to be less affected by the direction of the major finan-
cial markets (‘diversifiers’). But the statistical significance of AIS factors models
has to date been rather low. This is partly due to the short time series available for
research. Empirical measures for some AIS risk factors may not yet sufficiently
describe the significant losses after ‘big events’ which are part of many strategies’
risk profiles. Most AIS professionals agree that qualitative reasoning has to sup-
plement such quantitative analysis. Schneeweiss et al. in a recent publication
provide a good summary of the discussion on sources of AIS return.24

The structure of AIS multi-manager funds

Along with the rapid growth and expanded complexity of AIS in multi-manager
portfolios has come the increasing prevalence of pooling Hedge funds in multi-
manager portfolios managed by a specialist ‘fund of funds’ manager. This way
investors are relieved of the complex task of strategy sector allocation, manager
due diligence and investment monitoring. A ‘fund of funds’ is an investment
vehicle that mixes and matches Hedge funds and Managed Futures, spreading
investments across strategies, thus diversifying AIS sectors and managers. Selecting
the appropriate managers requires an experienced portfolio manager with
the necessary knowledge, infrastructure and relationships in the industry.

.... 31 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 32

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

A fund of funds can also be a more liquid investment than direct investments with
single Hedge funds or Managed Futures. Furthermore, if properly managed, it can
provide the security created by continuous monitoring of managers and portfolio
and risk management. Therefore, investing in a fund of funds is the preferable route
for new investors, institutional as well as private.25 But the added value of a fund of
funds is only realized if certain conditions are met. The fund of funds manager must
have a complete understanding of the individual strategy sectors and their risks and
correlation features, perform thorough due diligence on all managers and their
respective ‘edges’, implement a system for continuous monitoring of open positions,
and actively manage risk (see Chapter 7 for a more detailed discussion).
The structure of AIS investment products such as funds of funds can be quite
complex and innovations change the industry continuously. Managing a multi-
manager product requires expertise not only on the level of instruments and
markets, but also in respect of legal and regulatory issues. The set-up of a multi-
manager AIS investment vehicle involves a variety of different parties (see the
discussion in Chapter 7 for more details):

1 Management company or fund of funds investment manager: The investment

manager selects and monitors the individual trading managers and determines
the legal and administrative structure of the investment vehicle. He usually
receives a management as well as a performance fee.
2 Investment advisors/managers: The managers are hired by the investment
manager and are responsible for the actual investments, either in his own fund
or through a managed account.
3 Administrator: The administrator is responsible for registration and issuance
of the shares, performing the necessary legal and tax tasks, calculation of NAV
and performance fees, account administration, handling of subscriptions and
redemption and the organization of audits.
4 Prime broker: Prime brokers provide for the global settlement, clearing and
execution of trades done by the managers. They also provide credit lines for
financing leverage and Short Selling capabilities. Furthermore, prime brokers
often give the investment manager access to value added services like research
reports and risk management tools.

.... 32 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 33

2 ■ The Universe of Alternative Investment Strategies

5 Execution brokers: The execution broker arranges the execution of trades,

either over the counter or on an exchange. Most managers use several
executing brokers in order to get best prices. A good part of the trades is
usually executed through the prime broker.
6 Custodian: The custodian is the connecting piece between the broker and
the administrator. He tracks the receipts and payment obligations for subscriptions
and redemptions, monitors the brokers and records the different transactions.

The number and variety of multi-manager AIS investment products is growing

rapidly.26 They come in a variety of structures, the details of which depend on the
chosen domicile, promotional issues (sales restrictions), required investment flexibility
and tax issues. Many funds of funds are set up offshore, as offshore centres provide
investors with tax benefits and relieve the managers of regulations on the use of deriv-
atives, short sales and leverage. US fund managers primarily choose the nearby
Caribbean islands (Bahamas, BVI, Cayman Islands, Bermuda, Netherlands Antilles),
while Europeans select Luxembourg, Ireland, the Channel Islands (Jersey, Guernsey)
and Switzerland (where regulations are usually stricter than in Caribbean countries).27
The actual location of the AIS portfolio manager usually differs from the fund domi-
cile and is usually in the proximity of main financial centres (New York, London,
Geneva, Zurich, Hong Kong, Tokyo, Frankfurt etc.). Today four general structures of
AIS fund of funds investment vehicles can be distinguished: open-ended funds, closed-
end funds, structured notes and managed accounts (see also Chapter 7 for a
discussion of the different redemption policies of these fund of funds products).
Open-ended funds of funds with redemption periods ranging from one to six
months are usually structured as limited liability companies in the USA and the
Caribbean offshore centres, trust structures in the UK (including the Channel Islands)
or some form of collective investment contract scheme in continental Europe (e.g.
SICAV in France and Luxembourg). Low liquidity (long redemption period) and lim-
ited investment transparency are the main disadvantages of this structure.
Structured notes are designed to fulfil specific investor needs such as capital pro-
tection, regulatory specifications, legal constraints and tax protection.28 They have
become a common structure for multi-manager AIS products in Europe, often offered
by investment banks in cooperation with fund of funds specialists. In Germany, for

.... 33 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 34

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

example, index-linked bonds, so called ‘Zertifikate’, are currently the most common
AIS investment structures for tax and regulatory reasons (most AIS vehicles offered to
German investors are based on some self-created ‘index’ made up of the selected man-
agers). The customization to specific needs is the advantage of structured notes, but
their disadvantage is that they often come with higher fees. A common form is a prin-
cipal protected note which guarantees the 100% payback of the investment at
maturity. Capital guaranteed notes are usually structured with a zero coupon bond or
an option.29 For most structured notes the issuer provides for some liquidity in a sec-
ondary market.
Other AIS products take the form of a closed-end investment vehicle, which
often enjoys more regulatory flexibility than an open-ended fund. These products
are wrapped as investment companies, which issue non-redeemable shares. After an
initial offering period the investment company is closed for subscription and
redemption and the shares are traded on an exchange. The problem of these invest-
ment structures is that trading is often extremely low in volume and liquidity. This
can lead to significant discounts of the market price to the net asset value (NAV).
Investors and allocators can structure their investments with individual man-
agers through managed accounts, where the money is held in an account with an
independent custodian or broker. Managed accounts structures offer the most
flexible and transparent investment structure. But they require a significant mini-
mum investment with each manager ranging from around $2 million to about $25
million. In a multi-manager investment set-up, a ‘fund of managed accounts’
requires a special legal framework to ensure that the individual managed accounts
carry no liability from the other accounts.

Notes
1. A good discussion of the characteristics of AIS strategies (including their
historical return properties) and an AIS classification scheme is presented by
A. Ineichen in ‘In Search for Alpha’, October 2000 (updated and extended version
‘The Search for Alpha Continues’, September 2001).
2. Some authors present classification schemes which deviate in certain nuances.
I will try to indicate other schemes, wherever it is appropriate.

.... 34 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 35

2 ■ The Universe of Alternative Investment Strategies

3. For example, A. Ineichen in ‘In Search for Alpha’, October 2000 and ‘The
Search for Alpha Continues’, September 2001.
4. See the article ‘A Primer on Hedge Funds’ by S. Fung and D. Hsieh for more
details of AIS legal and regulatory issues.
5. See the book by Edwin Lefevre, Reminiscences of a Stock Operator (1923) for
more details.
6. C. Loomis, ‘The Jones Nobody Keeps Up With’, Fortune, April 1966, p.237.
7. J. Rohrer, ‘The Red Hot World of Julian Robertson’, Institutional Investors,
May 1986, p.86.
8. An interesting history of Hedge funds is presented in T. Caldwell,
‘Introduction: The Model for Superior Performance’, in J. Lederman and R. Klein,
Hedge Funds: Investment and Portfolio Strategies for the Institutional Investor,
Irwin Professional Publishing, New York, 1995.
9. Three surveys by Watson Wyatt/INDOCAM illustrate the increasing institutional
demand for AIS products: ‘Alternative Investment Review Relating to the Continental
European (respectively United States and United Kingdom) Marketplace’, Fall 2000.
See also report by Golin/Harris Ludgate, ‘The Future Role of Hedge Funds in
European Institutional Asset Management 2001’. PricewaterhouseCoopers performed
a survey among private banks in European with respect to the status quo and their
expectations of AIS and their importance, ‘European Private Banking / Wealth
Management Survey 2000/2001’. Other surveys with similar results were performed
by Goldman Sachs/Frank Russell (2001 Alternative Investing Survey), Deutsche
Bank’s equity prime services unit (see http://www.hedgeworld.com/news/read_news.
cgi?story=peop606.html&section=peop), and the Barra Strategic Consulting Group
FOHF (market survey, 2001).
10. An illustration of Hedge fund activities in the USA and Europe is presented in
‘The Hedge Fund “Industry” and Absolute Return Funds’, Goldmann, Sachs & Co.
and Financial Risk Management Ltd., The Journal of Alternative Investments,
Spring 1999 and in ‘Starting a Hedge Fund – a US Perspective’ and ‘Starting a
Hedge Fund – a European Perspective’, both published by ISI publications.
11. See, for example, ‘Hedge Funds – The Latest Bubble?’, The Economist,
September 1, 2001, ‘The $500 billion Hedge Fund Folly’, Forbes Magazine, August
6, 2001, ‘The Hedge Fund Bubble’, Financial Times, July 9, 2001.

.... 35 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 36

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

12. For a more detailed discussion on bubbles in the financial world, see the
paper by R. Shiller, ‘Human Behavior and the Efficiency of the Financial System’,
and the book by H. Shefren, Beyond Greed and Fear: Understanding Behavioural
Finance, Harvard Press, Boston (1999).
13. These are discussed in detail by E. Fama in a seminal paper: ‘Efficient Capital
Markets: II’, published in the Journal of Finance, December 1991.
14. Standard finance proponents argue that market efficiency is not testable
because such tests must be jointly accompanied by a test of an asset pricing model.
15. Most finance books cover the CAPM including its fundamental assumptions
in great detail, e.g. F. Reilly and K. Brown, Investment Analysis and Portfolio
Management, The Dryden Press, 1997.
16. The APT was introduced by S. Ross in the early 1970s and first published in
1976: ‘The Arbitrage Theory of Capital Asset Pricing’ in Journal of Economic
Theory, December 1976.
17. See Chapter 10 of Investment Analysis and Portfolio Management by F. Reilly
and K. Brown for a good discussion of recent research results.
18. One example is the fact that stocks with high book to price value (BV/PV) have
outperformed other stocks significantly in past years. It is still disputed whether this
is a pricing anomaly or a risk premium. An argument raised by E. Fama and K.
French in their papers ‘The Cross-Section of Expected Stock Returns’ and ‘Size and
Book-to-Market Factors in Earnings and Returns’ is that investors pursuing a strategy
of buying high BV/PV stocks provide ‘recession insurance’ for other investors.
19. The risk free rate of return has nothing to do with a risk premium. In
economic terms, the risk free rate of return is the compensation to the investor for
not persuing current consumption in exchange for higher future consumption.
20. See the seminal paper by W. Sharpe, ‘Asset Allocation: Management Style and
Performance Measurement’ and the paper by E. Fama and K. French ‘Multifactor
Explanations of Asset Pricing Anomalies’.
21. See the following articles for a further discussion: T. Schneeweiss and R. Spurgin,
‘Multifactor Analysis of Hedge Funds, Managed Futures, and Mutual Fund Returns
and Risk Characteristics’; B. Liang, ‘On the Performance of Hedge Funds’; W. Fung
and D. Hsieh, ‘Empirical Characteristics of Dynamic Trading Strategies: The Case of
Hedge Funds’; W. Fung and D. Hsieh, ‘Benchmarks of Hedge Fund Performance:
Information Content and Measurement Biases’; W. Fung and D. Hsieh, ‘The Risk in

.... 36 ....
8409 Chapter 2 p17-37 11/4/02 1:12 PM Page 37

2 ■ The Universe of Alternative Investment Strategies

Hedge Fund Strategies: Theory and Evidence from Trend-Followers’; V. Agarwal and
N. Naik, ‘Performance Evaluation of Hedge Funds with Option-based and Buy-and-
Hold Strategies’; F. Edwards and M. Caglayan, ‘Hedge Fund Performance and
Manager Skill’; See also the summary discussion of different articles on this subject by
A. Ineichen in ‘The Search for Alpha Continues’, pp.93–95. An interesting study on
return profiles of Hedge funds is also the following: G. Amin and H. Kat, ‘Hedge
Fund Performance: 1990–2000: Do the Money Machines Really Add Value?’
22. See the article ‘Hedge Funds Placed Under the Microscope’ by G. Polyn in Risk
Magazine, August 2001, and references therein (especially the study by the International
Security Market Association). Academic research has also focused on the direct
replication of the AIS with generic trading models, see e.g. the discussions by T.
Schneeweiss and R. Spurgin in ‘Trading Factors and Location Factors in Hedge Fund
Return Estimation’, T. Schneeweiss and H. Kazemi, in ‘The Creation of Alternative
Tracking Portfolios for Hedge Fund Strategies’, as well as the paper by W. Fung and D.
Hsieh, ‘The Risk in Hedge Fund Strategies: Theory and Evidence from Trend-Followers’.
23. A promising approach of a ‘style factor model’ is presented by A. Weismann
and J. Abernathy in the article ‘The Dangers of Historical Hedge Fund Data’, in
Risk Bucketing: A New Approach to Investing, 2000.
24. ‘Understanding Hedge Fund Performance: Research Results and Rules of
Thumb for the Institutional Investor’, Lehman Brothers Publication, Dec. 2001.
25. For a detailed discussion of the characteristics and value added of fund of
funds, see also the contribution by A. Ineichen, ‘The Search for Alpha Continues –
Do Fund of Hedge Fund Managers Add Value?’
26. A good (but somewhat outdated) overview of the different structures is
presented in Hedge Funds and Managed Futures by P. Cottier, p.55.
27. A discussion of the different legislations for Hedge funds is presented in ‘The
Capital Guide to Alternative Investments’, ISI publications, 2001.
28. See the following articles for more details: P. Astleford, W. Yonge and R.
Edwards et al. and I. Cullen in ‘The Capital Guide to Alternative Investments’, ISI
Publications, 2001; P. Watterson, ‘Offering Private Investment Funds in the Capital
Markets’, Alternative Investment Quarterly, October 2001, p.43.
29. For an interesting discussion of structured AIS products, see the article ‘French
Fight over Hedge Fund Products’ by N. Dunbar in Risk Magazine, February 2001.

.... 37 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 38

CHAPTER 3

Alternative Investment Strategies:

Hedge funds and Managed Futures

In this chapter each AIS sector is discussed in terms of its investment strategy,
sources of return and risk characteristics. In order to acquaint the reader better
with the strategies, realistic examples are provided.

Hedge fund strategies

Figure 3.1 expands on Figure 2.1 and displays an overview of today’s Hedge fund
and Managed Futures strategies. For Hedge funds, I distinguish the sub-categories
Relative Value, Event Driven and Opportunistic.1 The most important differenti-
ating factor between these three categories is the degree of sensitivity to the
direction of financial markets. While the returns of Opportunistic strategies often
depend on the (equity) market direction, Relative Value strategies earn returns
that are largely independent from it. Event-Driven strategies fall in some medium
range in terms of their directional market sensitivity.
Most of the strategies in the array of Hedge fund styles emerged in the last 15 to
20 years. Relative Value strategies (Convertible Arbitrage, Fixed Income Arbitrage,
Equity Market Neutral) seek out different specific risk premiums and arbitrage
opportunities in equity and fixed income markets. Event-Driven strategies
(Distressed Securities, Merger Arbitrage and Convertible Debenture Arbitrage) in
contrast capitalize on the occurrence of special events that impact the value of
certain securities. Opportunistic strategies cover a heterogonous third sector.

.... 38 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 39

3 ■ AIS: Hedge funds and Managed Futures

Relative Value strategies usually have very little net exposure to broad equity and
bond market, while Event-Driven strategies have some exposure to specific equity
and fixed income risk factors. Opportunistic strategies (Long/Short Equity,2 Macro,
Short Sellers, Equity Market Timing) are broadly influenced by the returns in broad
equity, fixed income and foreign exchange markets.

Alternative Investments Strategies

(AIS)

Hedge Funds Managed Futures

Relative Value Opportunistic (Active)

Discretionary–Active
Convertible Arbitrage
(Fundamental/Spread)

Systematic–Technical
Fixed Income Arbitrage
(Trendfollower)

Equity Market Neutral Passive Index

Event Driven Systematic–Long/Short

(sGFI, MLM)

Commodity Long Only

Risk Arbitrage
(GSCI, CRB)

Distressed Securities

Convertible Debenture
Arbitrage–Reg D

Opportunistic

Macro

Long/Short Equities

Equity Market Timer

Short Sellers

FIGURE 3.1 ■ AI strategies

Source: Partners Group

.... 39 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 40

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Relative Value – Convertible Arbitrage

The strategy
Convertible securities are equivalent to holding both a bond and an option on the
stock of the issuer.3 The pricing relationship between the convertible bond and
the stock can be complex and market pricing is often not completely efficient for
this combination (especially if, as is often the case, the bond is in addition callable
and/or putable). Because the option part(s) of the bond is (are) not traded sepa-
rately, pricing anomalies occur regularly, particularly in volatile periods. With the
help of quantitative valuation models accounting for the bond’s specific character-
istics, Convertible Arbitrage managers pursue Arbitrage opportunities by
identifying these pricing disparities. The strategy consists of taking long positions
in ‘cheap’ convertible securities and hedge their equity part by selling short the
corresponding stock. Long positions in convertible securities include convertible
bonds, convertible preferred stock and warrants. When the convertible ultimately
achieves its ‘fair value’ relative to the stock price, the position is sold and profits
are realized. Besides equity risk, managers often hedge other risk factors such as
credit and interest rate risk using a variety of hedging techniques.
Convertible Arbitrage managers usually start with a universe of convertible
securities and screen candidates with a bottom-up valuation analysis. In order to
identify ‘cheap’ securities managers compare the convertible’s market price with
its calculated expected value based on interest rate levels, the market price and
volatility of the underlying equity, the bond floor, the expected equity premium,
the conversion price and ratio, the conversion premium and other variables
related to the issue (credit quality, call schedule etc.).4 While most Convertible
Arbitrage managers target a market (beta) neutral position, some take a funda-
mental view on the issuer and build this opinion into a net short or net long
position. The equity hedge ratio (delta) generally ranges from 0.3 to 1.0. Most
Convertible Arbitrage managers employ leverage ranging from 1:1 to 6:1. The
relatively high risk-adjusted returns of Convertible Arbitrage strategies in the past
are partly attributable to positive beta positions in the bull equity markets of the
1980s and 1990s.

.... 40 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 41

3 ■ AIS: Hedge funds and Managed Futures

Different variations of Convertible Arbitrage strategies are:

■ buying convertible bonds and selling short the underlying stock/stock option
■ buying convertible bonds and shorting index Futures/index options
■ searching for price inefficiencies in complicated convertibles with numerous
conversion characteristics and callable and put-able schemes
■ focusing on low credit quality or distressed convertibles in combination with
the stock.

The concept of Convertible Arbitrage strategies is threefold:

1 earning static returns

2 being long stock price volatility
3 exploiting price inefficiencies.

Profits can be generated in all the following ways:

Static return:
(a) Convertibles pay interest or preferred dividends (current yield).
(b) The short equity portion generates positive cash flow (interest income
from short rebate minus dividends).
The highest level of static return is earned with in-the-money convertibles
with low credit ratings. These pay high coupons and require a high equity
hedge, i.e. they earn high short rebates (assuming a constant bond floor and,
in particular, constant credit rating).
Volatility trading:
The strategy is typically long volatility. As the stock price volatility changes, the
convertible’s value will change due to the commensurate change in the inherent
option value (conversion premium). The higher the volatility of the underlying
stock, the more opportunity for the manager to realize trading profits. The
optimal trade involves healthy companies (from a credit perspective) with
overvalued stocks. Being delta hedged and long gamma, the strategy benefits
from an increase in stock price volatility or a decrease in stock price.

.... 41 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 42

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Fair value:
Utilizing quantitative pricing models, the manager identifies mispriced
securities. These ‘cheap’ positions are acquired and hedged. When ‘fair
value’ is ultimately reached, these positions are sold and profits are realized.

Figure 3.2 illustrates the way in which Convertible Arbitrage returns are achieved.
An important distinction between different Convertible Arbitrage managers is
their varied focus on the ‘moneyness’ of the convertible, i.e. how much the option
is in or out of the money. Some managers focus on deep-in-the-money convert-
ibles, while others trade mainly out-of-the-money issues (or even ‘busted
convertibles’, i.e. deeply out-of-the-money issues). Depending on the moneyness of
the option, credit quality of the convertible securities can be an important factor
to consider. Together with prevailing interest rates, the credit quality determines
the bond floor of the convertible security. The average grade of the bonds in a typ-
ical Convertible Arbitrage portfolio is BB to BBB with individual ratings ranging
from AAA to CCC. However, as the default risk of the company is somewhat
hedged by shorting the underlying common stock, especially for in-the-money

20 Total Return

15

Contribution
10 Current from Leverage
Short Interest Yield
5 Rebate Position Trading

+ + + =
0

Dividend
–5 Payment Cost of Leverage

FIGURE 3.2 ■ Return sources of Convertible Arbitrage

Source: Discussion with Clinton Group

.... 42 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 43

3 ■ AIS: Hedge funds and Managed Futures

convertibles, the overall risk of the combined position is considerably better than
the agency rating of the non-hedged bond indicates. Typically, lower credit rating
implies the necessity for a higher equity hedge.
Convertible Arbitrage is more than solely identifying ‘cheap volatility’. The man-
agement of tail risk is a key component of a successful strategy implementation. In
addition to hedges for protection against movements in stock prices, most managers
also seek to hedge interest rate exposure. Furthermore, as a somewhat liquid credit
derivatives market is developing, credit risk swaps are becoming an important hedg-
ing tool for Convertible Arbitrage managers.5 The level of sophistication in hedging
techniques for Convertible Arbitrage Strategies is developing rapidly.

EXAMPLE
The US XYZ Co. issues a 5-year, 7% convertible bond due December 1,
2002 and the manager buys $1.5m face value of the bond. The bond sells at
a price of $120 1/2 with a current yield of 5.81% and a conversion premium
of approximately 17.5%. The holding period of this position is assumed to
be 21 days. At the same time, 87,500 shares of XYZ Co. common stock are
sold short at $14 per share creating additional interest income from the
rebate of 5%. The static returns and assumed trading returns on this position
are listed below.
The following shows the different parts of the trade and the returns:

Activity Security 1/1/2001 1/21/2001 Gain

Buy Bond ($1,807,500)

Sell Bond 1,812,495
$ 4,995
Short Stock $1,225,000
Repurchase Stock (1,214,062) $10,938

.... 43 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 44

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

number of days $ 6,125

(
Coupon interest at 7% accrual basis: –––––––––––––––
360
)
Short stock interest rebate at 5% $ 3,573
Total gain $25,631

Annualized rate of return 24.6%

Sources of return
Convertible Arbitrage strategies have a variety of return sources. An important eco-
nomic source of return of Convertible Arbitrage strategies is the risk premium related
to taking ‘long volatility’ positions. Being able to take a position cheaper than ‘fair
value’ is rewarding this risk of being long volatility. Furthermore, depending on their
specific implementation, Convertible Arbitrage strategies earn various premiums for
taking credit, duration, liquidity and negative convexity risk (if the bond is callable).
The relationship between convertible bonds and their underlying stock is rather com-
plex and the market of this combination is often not completely efficient. The ability
to detect mispriced securities is a function of the manager’s skill in valuing convertible
bonds correctly when the market does not. Most pricing inefficiencies of convertibles
are due to ‘volatility mispricings’. Convertible Arbitrage is the search for underpriced
volatility. It can be argued that the model risk, i.e. the risk of modelling complex secu-
rities incorrectly, corresponds to a premium the manager earns. Besides capturing
cheap volatility, a further important source of returns is identifying undervalued cred-
its and benefiting as the market recognizes the undervaluation and corrects it.

Risk factors
Although the strategy might appear conceptually rather simple at first sight,
Convertible Arbitrage is actually quite complex and a variety of risk factors are
involved. The successful implementation requires experience and skill in manag-
ing these risks. The value of convertible securities depends on different
characteristics, mostly the bond floor, the credit rating, the premium over conver-
sion value (i.e. the ‘moneyness’ of the option component) and the stock volatility.

.... 44 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 45

3 ■ AIS: Hedge funds and Managed Futures

There are different focuses within Convertible Arbitrage strategies with respect to
moneyness and credit quality of the convertible security. These differences induce
various profiles with respect to credit, equity and interest rate risk. Equity risk is
highest for at-the-money Convertible Arbitrage positions (they have the highest
gamma), while in-the-money and out-of-the-money convertibles have lower
gamma exposure, i.e. their delta hedges are less sensitive to changing equity
prices. Note that this lower sensitivity to the stock price refers to the combination
of the long convertible/short stock position (the convertible security by itself, of
course, has highest equity risk when it is far in the money). Convertibles trading
close to their bond floor (i.e. the option component is far out of the money) are
most sensitive to changes in interest rates. Among these ‘busted’ convertibles,
given an equal credit quality, the bonds trading above par are more interest rate
sensitive than bonds trading below par (due to positive bond convexity).
Convertibles trading at discounts below 65% of par value face significant credit
risk in the form of high default probability (i.e. their ‘bond floor falling through’).
In this case it is unclear whether the strategy actually benefits from an increase in
volatility, as mostly credit issues determine the value of the convertible security.
An important input for the determination of the option’s fair value is the esti-
mate for the underlying stock’s volatility. Correspondingly, one of the key risk
factors for Convertible Arbitrage strategies is vega risk, which denotes the risk that
the volatility will decrease, rendering the option value of the convertible (conversion
premium) lower. In addition, the strategy is long gamma, i.e. the delta of the long
option position increases with a rising and decreases with a falling stock price.
Gamma risk is not as significant as vega risk, but both risks are difficult to hedge, as
maturity and strike price are in most cases not known. An incomplete delta hedge
(e.g. after a move in the stock price) exposes the strategy to delta risk, i.e. being
over- or under-hedged with respect to future moves in the stock price (equity risk).
Related to this is the risk of an increase in the dividend of the stock sold short,
which reduces the static returns and decreases the value of the option. It is finally
very important to consider any callable features of the bond (i.e. the risk of the bond
being called by the issuer) when evaluating the appropriate hedging strategy.
A long convertible bond position is (like any regular bond) a long duration and
long convexity position. The bond price moves inversely with interest rates.

.... 45 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 46

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Therefore, Convertible Arbitrage strategies face interest rate risk or duration risk
(plus negative convexity risk, if the bond is callable). With rising interest rates, the
straight debt value (the ‘bond floor’) decreases. Although stock returns tend to be
negatively correlated to interest rates, losses on the bond part are usually only
partly offset by the short equity position held against it. Another partial hedge is
given by the positive rho of long convertible position. The option value increases
with rising interest rates. Many managers use interest rate derivatives such as
interest rate Futures and options on interest rate Futures, interest rate swaps and
high yield total return swaps to hedge against duration risk. Option adjusted
spread (OAS) is a concept widely used in valuing bonds with embedded options.6
Convertible Arbitrage strategies are exposed to corporate event risk. While
some corporate event risk is hedged through the combination of long convertible
and short stock, particular events can still influence the overall position negatively,
e.g. through a credit event or a merger transaction involving the company.
Generally, stock takeovers often involve simultaneously a credit upgrade and a
reduction in volatility, so it is difficult to say a priori whether they will help or
hurt a given issue. Cash takeovers in contrast are generally punitive for convert-
ible arbitrageurs, as the underlying goes from a highly volatile asset (equity) to
cash (that by nature has no volatility). As for any non-G7 government bond, a
long convertible bond position is subject to credit risk, i.e. the risk of credit down-
grade, spread widening or straight default (bankruptcy). Changes in credit quality
affect the security’s bond floor. Part of the credit risk is hedged with the short
equity position. Degrading credit quality of the bonds usually comes along with a
decrease in the market value of the stock. This correlation is not always exact,
however. With very sharp drops in the equity value of the issuing company, the
value of the convertible bond becomes a pure function of the bond floor, which
might behave differently than the stock (and can fall faster than the stock in some
scenarios, indicating the presence of negative gamma/vega). The technology
market downturn in 2000/2001 provides numerous examples of busted convert-
ibles with degraded credit quality. Credit risk generally becomes most dominant
for convertible securities trading below 65 to 70% of par value. Credit swaps and
other credit derivatives can be used to ‘strip out’ the credit risk.

.... 46 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 47

3 ■ AIS: Hedge funds and Managed Futures

The market for convertible securities has its own valuation cycles driven by
supply and demand for convertibles that are more or less independent from stock
market valuations. The performance of Convertible Arbitrage strategies is largely
uncorrelated to equity market directions in periods of average volatility. Due to the
increasing option values, periods of higher volatility are usually favourable to the
strategy. Extreme stock market volatility often gives rise to ‘flight to quality’ situa-
tions, however, in which convertible security prices fall rapidly, liquidity dries up
and bid–ask spreads widen abnormally (as e.g. in the summer/fall 1998). This
exposes Convertible Arbitrage strategies to liquidity risk through:

■ Higher bid–ask spreads for the convertible bond.7 Most convertible bonds are
traded OTC with less regular price updates than exchange-traded instruments.
■ ‘Short squeezes’, a situation where there is insufficient supply of stocks to
deliver on the short position (the fact that the convertible can be converted
into stock makes the loss from this marginally less pronounced; conversion to
cover a short stock involves losing all conversion premium, which causes
major losses on all converts except busted converts on good credits and deep-
in-the-money issues with little or no call protection).
■ Higher margin haircuts.

An important factor in a manager’s ‘edge’ is the valuation model underlying the

pricing of the convertible securities. But every valuation model is exposed to
model risk, i.e. the risk that the model is mis-specified. Even a ‘correct’ model can
lead to problems in the hands of a manager who blindly follows it, because
models can, by their very nature, only capture parts of the market dynamics.
Models should be combined with experience.
To the degree Convertible Arbitrage managers are dealing with foreign issues,
their investments are exposed to foreign exchange risk, i.e. the risk of adverse
moves in the bond’s currency against the manager’s home currency. FX risk can
be hedged with Forwards, Futures and options. Other types of risks include legal
risks and regulatory risk. Details of the convertible’s prospectus can be important,
e.g. the treatment of special dividends or the treatment of merger situations. As
Convertible Arbitrage strategies involve stock shorting, the strategy is exposed to

.... 47 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 48

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

the specific risks of Short Selling (shorting risk) connected to the fact that the
short seller has to borrow stock (e.g. short squeezes, execution risk). (These risks
will be discussed in detail in the section about Short Selling strategies.)
Generally, markets with high and rising volatility combined with decreasing
interest rates are the most favourable for Convertible Arbitrage strategies. In these
circumstances being long volatility and long gamma pays off. In cases where the
manager keeps a positive overall delta, rising equity markets produce better
returns, while falling equity markets lead to losses. Generally, rising interest rates
and decreasing volatility create a negative environment for the strategy. Extreme
volatility combined with a flight to quality environment and a liquidity crisis can
be very damaging for the strategy.

Relative Value – Fixed Income Arbitrage

The strategy
Fixed Income Arbitrage is a generic description for a variety of arbitrage strategies
involving different fixed income instruments. The strategy generally seeks profits
by exploiting pricing inefficiencies between related securities and their derivatives.
Investment instruments include US government and agency debt, G-7 government
debt, sovereign emerging market debt, corporate debt, municipal debt and
mortage-backed securities. By investing in one or more securities and simultane-
ously taking an opposite position in another related instrument (e.g. T-bill versus
Eurodollar), the Fixed Income Arbitrage manager neutralizes exposure to interest
rate risk factors such as yield curve changes. These positions are referred to as
‘spread positions’. Different types of Fixed Income Arbitrage trades are yield curve
trades (spread positions including bonds with different maturities), corporate
versus treasury yield spreads, municipal versus treasury yield spreads, cash versus
Futures and on-the-run versus off-the-run bonds. The Fixed Income Arbitrage
Universe also includes a variety of strategies involving Asset-Backed Securities
(ABS). Managers differ quite substantially in terms of how and to what degree
interest rate risks, credit risks, foreign exchange risks and inter-market spread risks
are hedged. Some managers even take ‘directional bets’ with respect to yield curve
or credit spread changes, others employ spread trades only.

.... 48 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 49

3 ■ AIS: Hedge funds and Managed Futures

Different forms of pure, i.e. market neutral, Fixed Income Arbitrage are as follows:

■ Arbitrage between similar bonds (e.g. long underpriced 7-year duration US

T-bond and short overpriced 7.1 Y duration US T-bond). This often involves
cheapest-to-deliver bonds underlying a Futures contract on the bond or on-
the-run vs. off-the-run issues.
■ Butterflies (e.g. long cheap 7 year and 9 year, short expensive 8 year). Relative
mispricings along the yield curve often occur due to high institutional demand for
certain benchmark bonds. The strategy is also referred to as ‘yield curve arbitrage’.
Figure 3.3 displays an example of a profitable yield curve arbitrage situation.
■ Basis trading is an arbitrage between physical securities and their Futures (e.g.
short overpriced US 10-year note Future, long underpriced US 10-year note).
Bond Futures have a delivery option and a wild card option, which can lead
to pricing inefficiencies and provides for arbitrage opportunities.
■ TED spread, i.e. spread between Treasury bill Futures and Eurodollar Futures
(this can be seen as a credit spread trade: government debt versus AA rated
inter-bank debt).
■ Arbitrage between on-the-run and off-the-run bonds (e.g. short the latest
government issue of a 10-year note and long the second most recent issue).

Other, non-market neutral, strategies include the following:

■ Yield curve spread trading based on a forecast of the directional change of the
yield curve. An example could be going long the short end of the curve (up to
3-year maturities) and going short the long end (i.e. 10–15 years) anticipating
a steeper yield curve in the future.
■ Credit arbitrage or credit spread trading capturing a credit-pricing anomaly
and profiting from yield curve differentials for papers with different (but
generally closely related) credit qualities (e.g. short an AAA rated bond with a
spread to T-bonds of 50bp and long an AA rated bond with a spread of 80bp).
Sometimes a ‘credit barbell’ strategy is employed, a technique whereby
managers assume credit risk in short and intermediate maturities and use safe
government issues with long maturities.

.... 49 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 50

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

5.5

5
5.2
Long $50 9Y Bond
Yield

4.5 5.1
Long $50 7Y Bond

Yield
5
4
4.9
Short $100 8Y Bond
3.5
4.8
6 7 8 9 10
Years
3
0 5 10 15 20 25 30
Years

FIGURE 3.3 ■ Profitable Fixed Income Arbitrage situation

■ Asset-Backed Securities (ABS), e.g. credit card receivables, auto loans or mortage
debt, offer enhanced returns for investors, which assume exposure to the
embedded option features (prepayment, call option) and accept lower liquidity
and possible credit risk of the ABS. Examples for ABS arbitrage strategies are:
– Spread position ABS against T-bonds (e.g. long government secured Fannie
Mae or Freddie mortgage-backed securities (MBS) and short US Treasuries
with similar duration).
– Spread position of MBS against collateral mortgage obligations (CMO). An
example is selling certain tranches of a CMO and buying a plain pass-
through MBS.
– Arbitraging between different CMO classes. Two examples are: going long
interest-only tranches (IOs) and shorting principle-only tranches (POs);
shorting a Principal Amortization Class (PAC) tranche and going long a
support tranche.
The hedging of prepayment risk is rather complex but nevertheless
commonly employed within ABS Arbitrage strategies.

.... 50 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 51

3 ■ AIS: Hedge funds and Managed Futures

The spreads available for Arbitrage strategies in fixed income markets tend to be
small. In order to earn attractive returns, most Fixed Income Arbitrage strategies
must employ a high level of leverage, which may range from 10 to 25 (in some cases
even higher) times the asset base and is created through borrowing, repo (repur-
chase) transactions or the use of options, Futures or swaps. A creditworthy investor
with good dealing relationships might be able to transact $100 million notional
value while putting up less than $1 million collateral. The simpler strategies (e.g.
basis trades) are generally more highly leveraged than trades that are systematically
exposed to more risk and specific risk factors (e.g. ABS strategies).

Sources of return
Pricing inefficiencies and arbitrage opportunities in fixed income markets occur
for a variety of reasons including sudden market events, exogenous shocks to
supply or demand, investors having maturity preferences or restrictions in certain
fixed income investments (e.g. lower credit ratings), recent downgrade in credit
ratings, complex options/callable features connected to a bond, deliverable charac-
teristics for a Futures contract and complex cash flow properties. Most of these
‘pricing anomalies’ are related to certain risk premiums due to liquidity or credit
risk. Fixed Income Arbitrage managers are often long and short equal amounts of
securities with similar but not equal credit quality and liquidity. Thus the strategy
earns a premium (the ‘spread’) for holding less liquid or lower credit quality
instruments and hedges other risks (e.g. interest rates, duration) by selling short
securities with higher liquidity and credit ratings. In an abstract sense, the strategy
sells economic disaster insurance. Managers take positions that correspond to
‘short put positions’ on financial market turmoil.
The successful implementation of a Fixed Income Arbitrage strategy requires a
very high degree of sophistication as prices of fixed income instruments depend on a
large variety of factors with complex interactions: yield (spot, forward) curves, volatil-
ity curves, (credit) spread curves, expected cash flows, prepayment features (for ABS)
and option characteristics (e.g. call, put and prepayments schedules). A ‘complexity
premium’ can be determined as one source of the return of Fixed Income Arbitrage

.... 51 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 52

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

strategies, as they are paid to understand complex pricing relationships that others do
not follow and take the risks of mismodelling this complexity.

Risk factors
Fixed Income Arbitrage strategies are ‘short volatility’ strategies. Conceptually, they
sell financial crisis insurances and subject themselves to ‘sudden event risk’. The
risks of the strategy become apparent in market stress situations. For many years
LTCM successfully executed a highly quantitative Fixed Income Arbitrage strategy,
until it failed spectacularly in the weeks after the Russian default in August 1998.
Fixed Income Arbitrage is exposed to correlation breakdown risk, which is related to
the event risk of ‘flight to quality’. Correlation breakdown is the sudden change of
historical co-movement patterns between corresponding instruments which occurs
due to changes in government policy, sovereign default (e.g. Russia in August 1998)
or other economic shocks or dramatic events (e.g. the terrorist attacks on September
11, 2001), when interest rates move rapidly, credit spreads widen and liquidity dries
up (‘flight to quality’ scenario). These ‘flight to quality’ events usually happen when
many market participants want to liquidate positions at the same time (‘everybody
running for the door’). They occur relatively infrequently, but when they do occur
the value of Fixed Income Arbitrage portfolios can drop significantly on a mark to
market basis resulting in losses of several standard deviations (tail risk).
In volatile markets the strategy can easily become a captive of the extreme
leverage, where a single margin call on a position can destroy an entire portfolio.
Often fixed income arbitrageurs are short liquidity, i.e. they hold a long position
in a comparably illiquid security and an offsetting short position in a relatively
liquid asset, thereby earning a ‘liquidity premium’. Liquidity risk comes in two
forms: the inability to meet margin calls (funding risk) and the (temporary) in-
ability to unwind a position at normal bid/ask spreads (liquidation risk). Prime
brokers may withdraw financing at particularly difficult times, which might neces-
sitate the liquidation of the portfolio. This phenomenon should be taken into
consideration when examining liquidity risk. Again, the failure of LTCM is an
illustrative example of the liquidity problems a Fixed Income Arbitrage strategy
faces in situations of market distress.

.... 52 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 53

3 ■ AIS: Hedge funds and Managed Futures

The strategy is exposed to credit risk (spread risk, risk of change in rating,
default risk) of the underlying bonds, which is sometimes hedged with credit
derivatives. Fixed Income Arbitrage managers often hedge interest rate risk, but
some managers (e.g. yield curve arbitrage) execute a particular view on the future
shape of the yield curve. In such an approach the manager is therefore exposed to
yield curve risk, the risk of change in level, slope and curvature of the yield curve.
Yield curve risk mainly stems from duration risk and convexity risk. Some strate-
gies are contingent on low financing costs and are therefore affected by higher
financing costs due to rising interest rates. Strategies that involve ABS are particu-
larly exposed to prepayment risk (risk of not receiving cash flows at foreseen
times). Prepayment risk is directly linked to the level and direction of interest
rates and interest volatility.
Operational risks include model risk, execution risk and legal/tax risks. As
many trading managers employ very complex quantitative models to find pricing
discrepancies in different markets, they are exposed to model risk, i.e. the risk of
mis-specification of the valuation method or risk models employed. Certain
strategies take a large number of different positions simultaneously, which leads to
execution risk (bad fills, slippage, clumsy order execution by the broker). Changes
in tax laws or a financial or political change (tax or legal risk) can cause a stable
relationship between two instruments to break up.
Fixed Income Arbitrage strategies perform well during times when there is a
low likelihood of financial distress and in markets with constant or slowly chang-
ing volatility and correlation between different fixed income instruments. In times
of severe financial disorder (e.g. sovereign default, rapid increase of interest rates,
political change), the strategy is exposed to potentially large losses.

Relative Value – Equity Market Neutral

The strategy
Equity Market Neutral strategies usually involve being simultaneously long and
short matched stock positions taking advantage of relatively underpriced and
overpriced stocks. The strategy strives to generate positive returns in both bull

.... 53 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 54

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

and bear markets by selecting a large number of long and short positions with a
total portfolio net exposure of zero, which can be established in terms of size ($
neutral), beta (beta neutral), country (country neutral), currency (currency neu-
tral), industry (sector neutral), market capitalization (size neutral), style
(value/growth stock neutral) or a combination thereof. An interesting feature of
Equity Market Neutral strategies is the doubling of alpha. The manager targets
returns on the long side as well as on the short side. The strategy sector is seen as
‘alpha generator’ par excellence. Obviously, however, there is a double amount of
risk involved. Equity Market Neutral strategies can be classified into Statistical
Arbitrage trading and Neutral Long/Short Equity (also called ‘Fundamental
Arbitrage’). Statistical Arbitrage usually involves model-based short-term trading,
while Neutral Long/Short Equity strategies have longer holding periods, realizing
profit when inefficiencies disappear and prices return to ‘fair value’. A rather
straightforward Arbitrage opportunity an Equity Market Neutral manager would
exploit is a situation where a company’s stock is traded with different prices on
two exchanges.
Statistical Arbitrage strategies utilize quantitative and technical analysis to detect
profit opportunities in under- and overvalued stocks of related companies, often
operating in the same sector. Normally, a particular type of arbitrage opportunity is
hypothesized, formalized into a set of trading rules and then back-tested with histor-
ical data. This way the manager expects to identify a persistent and statistically
significant scheme to detect profit opportunities. Critics refer to this strategy as
‘black box investing’, as characteristics of the model are often not made transparent
to investors. One example is ‘mean reversion’ of price differences between two com-
panies. This describes the tendency of the price difference of two stocks to revert to
some ‘normal’ level after they have developed apart from each other, often follow-
ing an extreme move by one or both stocks. Figure 3.4 gives a graphical illustration
of the corresponding ‘buy the dips, sell the rally’ strategy: the price difference of
two stocks is expected to develop around a certain mean.

.... 54 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 55

3 ■ AIS: Hedge funds and Managed Futures

Sell level

1 std. dev.
Mean
1 std. dev.
Buy level

FIGURE 3.4 ■ Equity Market Neutral: buy low, sell high

There are generally three steps in Statistical Arbitrage strategies: initial screening
and ranking, stock specific selection and portfolio construction. Technical models
are used to rank a universe of stocks according to return expectations and risk in
order to determine attractive long positions in undervalued stocks and short posi-
tions in overvalued stocks, thereby neutralizing the exposure to general market
risk. The underlying quantitative models are often linear multi-factor models that
result from a regression analysis on past data and establish a predicting formula
linking the chosen factors to subsequent stock returns. The trading decisions usu-
ally result from relative ranking systems. For the portfolio construction, many
managers utilize sophisticated computer algorithms.8 Powerful optimizers are com-
mercially available (e.g. by BARRA), while some managers construct their own
proprietary tool set. Trading for a Statistical Arbitrage strategy is often short-term
oriented and portfolio turnover is typically high. Often, significant leverage is used
to enhance returns.
The neutral Long/Short Equity strategy mainly consists of building portfolios of
long positions in certain industries’ strongest companies and short positions in those
companies that show signs of weakness. Analysis here is mostly of a fundamental
nature and less quantitative than Statistical Arbitrage strategies. Some managers use
technical and price momentum indicators including moving averages, relative
strength and trading volumes as supporting decision tools. Fundamental factors used
in the analysis include earnings ratios (price/earnings, price/book value, price/cash

.... 55 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 56

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

flow), discounted cash flows, return on equity, operating margins, earnings growth
and other performance indicators. Portfolio turnover is comparably low and, typi-
cally, modest levels leverage is employed. The size of the universe the individual
stocks are selected from is usually more limited, as more discretionary analysis has
to be performed on the individual companies.

EXAMPLE
Consider the following pair trade: Based on a valuation model, Company A
in an industry sector is determined to have a fair value of $50 a share,
while it is trading at $48 at the moment and is expected to pay $1.32 divi-
dend next quarter. Company B, in the same sector with similar distribution
channels, suppliers and financial performance, is determined to have a fair
value of $60 a share but trades at $63 and is expected to pay $1.70 within
the next three months. The manager buys 13,125 shares of A for $48 and
sells short 10,000 shares of Company B for $63 (the short sale might
require additional margin which is included under financing costs). Three
months later he reverses the trade paying and receiving fair values.

Today
Buy Company A (630,000) (13,125*48)
Sell Company B 630,000 (10,000*63)
Transaction costs (1,200)

Three months later

Sell Company A 656,250 (13,125*50)
Sell Company B (600,000) (10,000*60)
Bank A dividend (long) 17,325 (13,125*1.32)
Bank B dividend (short) (17,000) (10,000*1.7)
Financing costs (4,725) (margin requirements for short: 50%
of 630,000 at 6% annualized rate)

.... 56 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 57

3 ■ AIS: Hedge funds and Managed Futures

Transaction costs (1,200)

–––––––
$ 49,450

$ 49,450 365
Annualized return is ––––––– × ––– = 31.83%
$ 630,000 90

Sources of return
Equity Market Neutral managers employ their skill to detect market inefficiencies.
The basic assumption is that anomalies in relative stock valuations may occur in
the short term but, in the long term, these anomalies correct themselves as the
market processes information. As Equity Market Neutral strategies often employ
complex models to detect pricing inefficiencies, it can be argued that they earn a
‘complexity premium’. This is particularly true for Statistical Arbitrage strategies,
less so for Long/Short Equity Market Neutral (‘Fundamental Arbitrage’) strategies.
The inefficiency is determined as the difference between market price and a theo-
retical ‘fair value’ price determined by a model. Good models earn consistent
returns as long as others cannot replicate them in the market. However, the strat-
egy bears the risk of using inappropriate or incorrect models or the markets
moving further against ‘fair value’ (as seen with internet company valuations in
1999 and early 2000).
According to the CAPM, specific (or idiosyncratic) risk is not rewarded by
extra return. The generation of alpha by Equity Market Neutral managers appears
to be a contradiction to the efficient market hypothesis. It has indeed been shown
by several studies that the semi-strong form of the EMH must at least partially be
reconsidered.9 Studies of market anomalies focusing on firm size (smaller compa-
nies outperforming larger ones), book to market value (investing in companies
with high book to price value yields significantly higher returns than those given
by the CAPM), and price to earnings ratios (low P/E ratio companies tend to out-
perform companies with high P/E values) have called the EMH into question.10
However, an interesting argument raised by Fama and French is that investors fol-
lowing investment strategies based on these statistical effects, e.g. a value versus

.... 57 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 58

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

growth strategy or a focus on small to mid-cap companies, earn risk premiums

related to particular risks not covered by the CAPM rather than contradicting the
EMH.11 Examples of these additional risks are small firm risk, recession risk and
company leverage risk. Small firms have a higher probability of bankruptcy. They
also tend to be followed by fewer analysts which causes the reliability of informa-
tion to be lower and the stock to be less liquid. Low book to value stocks tend to
be cyclical, i.e. they are particularly hurt by a recession. Rather than saying that a
strategy provides a positive alpha one might include additional risk factors in the
pricing models and note that Equity Market Neutral strategies’ returns result at
least partially from exposure to these risks.

Risk factors
Statistical Arbitrage strategies are ‘short volatility strategies’ and can, in some
sense, be seen as the ‘equity market comparable’ to Fixed Income Arbitrage strate-
gies. With the help of statistical models managers detect relative mispricings in
different stocks, which in ‘normal’ periods should align themselves to fair value.
The event risk of the strategy is similar to a ‘short option position’, which makes
money in most circumstances, but in rare cases takes a significant hit (some Equity
Market Neutral manager actually do write options). This happens when simulta-
neous event and volatility shocks lead to extreme deviations from fair value
models. This risk is referred to as tail risk.
Many Neutral Long/Short Equity strategies take particular equity pair posi-
tions hoping that the market value of the mispriced securities moves back to fair
value levels based on fundamental analysis. Ultimately, the success of Equity
Market Neutral strategies comes down to stock selection and the principal risk
factor of the strategy is specific stock risk or sector risk. Stock/sector risk includes
event risk and any risk that affects the value of a company or sector in an unfore-
seen way, e.g. a surprise announcement such as earnings revision, merger or
changes of management.
Equity Market Neutral strategies aim to generate returns which are statistically
uncorrelated to the overall equity market and therefore usually insulate their portfo-

.... 58 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 59

3 ■ AIS: Hedge funds and Managed Futures

lios from market risk factors such as beta, sector, country, style, size and sometimes
even interest rates and commodity prices. Most managers do not hedge against all
these risk factors and are therefore exposed to residual risk. Even if the manager tar-
gets neutrality against market risk, this does not mean that the strategy bears no risk
related to the broad market. The strategy sector is generally exposed to volatility
risk. Extreme volatility can lead to significant losses. Most dangerous for the strat-
egy is non-reverting volatility.
Model risk is the risk that valuation methods used by the manager are inappro-
priate for the basket of stocks traded. Changes in market structure, e.g. the
emergence of new industries such as the internet, can expose the strategy to the
risk that certain valuation paradigms become inapplicable. In the late 1990s the
market value of internet stocks exceeded the outcome of most model calculations
and looked largely overpriced. Managers who shorted these stocks based on their
valuation models suffered huge losses. The use of multi-factor models, which are
based on regressions on past data, introduces the risk of ‘over-fitting’. Over-fitted
models have high predictive value on past data but little forecasting ability for
future returns. Managers have to be in a constant process of model review and
must refine models dynamically in order to keep them accurate.
Equity Market Neutral strategies, in particular Statistical Arbitrage strategies,
usually make hundreds of complex offsetting trades and have a wide web of posi-
tions in many different instruments open at any moment. Thus, the strategy bears
operational risk in the form of execution risk (slippage, clumsy order execution by
the broker). A large number of positions have to be monitored and many trades
executed. This normally happens either through humans or automatically through
computer systems, both of which are prone to errors. As Equity Market Neutral
strategies involve short selling of stock, the strategy is exposed to the specific risks
of Short Selling (short risk) connected to the fact that the short seller has to
borrow stock (e.g. short squeezes, execution risk). These risks will be discussed in
detail in the section about Short Selling strategies.
Markets with extended trends and valuations moving out of line with historical
patterns create an environment where Equity Market Neutral strategies face prob-
lems. Periods where for an extended period of time valuations are extreme render

.... 59 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 60

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

the valuation models inappropriate and therefore cause problems to the strategy.
Inversely, volatile and choppy markets with erratic price moves but generally
showing ‘normal’ valuation levels create positive environments for Equity Market
Neutral strategies.

Event Driven – Risk Arbitrage

The strategy
Risk Arbitrage (also called Merger Arbitrage) involves investing in securities of
companies which are the subject of some form of extraordinary corporate transac-
tion, such as acquisition or merger proposals, exchange offers, cash tender offers
and leveraged buyouts. A merger generally involves the exchange of securities for
cash, other securities or a combination of cash and other securities. Typically, Risk
Arbitrage managers purchase the stock of a company being acquired or merging
with another company and, if feasible, sell short the stock of the acquiring com-
pany. Tender offers for a merger are usually made at a significant premium to the
pre-announcement share price. During the negotiations of a merger the target
company’s stock typically trades at a discount to the announced value of the
merger transaction, as there remains a residual risk that the merger will fail and
the stock price will drop back to the original pre-announcement level. Figure 3.5
illustrates the evolution of the two companies’ stock prices in a successful merger
situation. A profit is derived on realizing the differential (the ‘spread’) between
the price paid on entering the positions and the value ultimately received when
the deal is consummated. In mergers and acquisitions and other corporate transac-
tions the market is generally not completely information efficient. As information
is very specific and difficult to obtain (and correspondingly high in value), the sit-
uations offer opportunities to investors knowledgeable about the specific features
of a deal. The experienced manager usually translates the spread into an annual-
ized return, estimates the probability that the deal will be completed and then
gauges the risks of the particular deal against its return. He invests in deals that
offer sufficient compensation for the estimated risk of the transaction failing.

.... 60 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 61

3 ■ AIS: Hedge funds and Managed Futures

130

120

110

100

90

80

70

60

50

40

30.09.00

30.11.00
29.02.00

31.03.00

30.04.00

31.05.00

30.06.00

31.07.00

31.08.00

31.10.00

31.12.00
31.01.00
31.12.99

Acquirer Target

FIGURE 3.5 ■ Spread convergence in successful merger deals (illustration)

A successful Risk Arbitrage strategy relies strongly on fundamental research of the

involved companies and the issues surrounding the deal, such as legal and regula-
tory (e.g. anti-trust) matters. The deals are often highly complex and the task of
the manager is more related to risk management than to stock picking.12
Important elements of the analysis include the following:

■ Analysis of publicly available information on the companies involved in the

transaction. This includes the competitive environment in which they operate,
their financial position, assessment of the quality of management and anti-
trust concerns.
■ Assessment of the buyer’s motivation for the merger or acquisition as well as
the impact of the contemplated transaction on his reputation, competitive
position and financial performance. A review of the buyer’s past acquisition
history might provide interesting insights.

.... 61 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 62

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Evaluation of potential information regarding business developments at the target

company which might undermine the proposed transaction. Many uncompleted
mergers fail because the target company’s situation deteriorates unexpectedly.
Detailed analysis of the business and competitive environment as well as the
financial situation of the target company is essential for an assessment of this risk.
■ Assessment of the efficiency of any anti-takeover devices available to the target
company (poison pills, golden parachutes etc.).
■ Assessment of the strategic rationale of the transaction. The merging
companies should realize valuable strategic opportunities, such as geographic,
product or distribution expansion, new technological integration or
applications, significant cost savings or other synergies.
■ Financial analysis (cash flow estimates, ratio analysis etc.) of the post-merger
outlooks (pro-forma financial forecasts). This includes an analysis of the
amount of funding available for the merger and in what form funding is
available (e.g. stock, cash, LBO). Detailed financial analysis provides important
insights for assessing the probability that the merger will be completed.
■ Consideration of the background of the merger. The operational and financial
history of the companies have to be examined. The manager should know
about the merger history of the involved companies, who initiated merger talks
and whether there have been discussions about a merger with other companies
before. It is also important to examine whether key management interests have
been addressed. Some mergers fail because of disagreement over the future roles
of senior management (often CEOs).
■ Review of valuation parameters. The valuation analysis should determine
whether the price being paid for the target company is excessive relative to
the buyer’s business strategy, the market valuation of comparable companies in
the particular industry and the target company’s competitive position and
financial performance.
■ Calculation of the spread on the deal. The spread is the difference between the
price offered for the acquired company and the current market price. Based on
the spread, the manager calculates the return expected on closure of the deal.

.... 62 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 63

3 ■ AIS: Hedge funds and Managed Futures

■ Examination of merger conditions. The terms of the transaction can contain

conditions which may significantly impact the likelihood of the transaction
completing and which therefore must be carefully assessed. Such conditions
may include walk-away clauses, price provisions related to the acquirer’s
stock, financing contingencies, earnings tests, pending completion of due
diligence, minimum shares to vote in favour of the merger, revenue or profit
hurdles or possible debt covenants.
■ Estimate of the likelihood of regulatory interference with the deal and the
likely outcome of such. The constant monitoring of applications for regulatory
clearances is vital for Risk Arbitrage strategies. Different regulatory agencies
have to be considered (from the perspective of a US merger): Federal Trade
Commission (FTC)/Justice Department, SEC (must approve merger proxies),
individual states, specific industry (banks: Federal Reserve, utilities: Federal
Energy Regulatory Commission, communication: Federal Communications
Commission) and foreign regulatory agencies (e.g. EU, specific countries).

All this information is necessary for making an informed assessment of the risk of
a proposed transaction failing. A transaction can fail for a wide range of reasons
due to anything from failure to achieve the required shareholder approval, to neg-
ative earnings news from the target company and a significant drop in the stock
price of the acquiring company. Along with the assessment of the probability of a
transaction failing, one must estimate the financial impact of such an event. The
stock price before the deal announcement can serve as a general guide as to where
stock prices might go in case the deal does not go through. However, in many
cases applying a discount to the target company’s pre-announcement stock price is
appropriate. The failure of a merger is often caused by fundamental issues (e.g.
earning revisions of target company) and the market will probably value the target
company lower than before.
Information sources for this last analysis are stock market valuation, trade data
sources, past enforcement policies of justice department (DOJ), Federal Trade
Commission (FTC) and anti-trust agencies, discussions with management, invest-
ment bankers and legal advisors, court hearings and press publications.

.... 63 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 64

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Risk Arbitrage managers differ in the amount of deal risk they are willing to
take as well as the level of leverage and diversification employed. Most managers
only invest in announced transactions and some even invest only in deals that are
sufficiently progressed towards completion. Others enter in positions with higher
deal risk (and therefore wider spreads) and a few base their investment decisions
on rumours or speculation about possible transactions. Managers also distinguish
themselves with respect to sectors and geographical location.
The two basic types of merger offers are a cash tender offer, where a fixed amount
of cash is offered for the acquired company’s stock, and a stock swap, where stock of
the acquiring company are being offered at a fixed ratio in exchange for the stock of
the acquired company. More complex features of a merger transaction are cases in
which the exchange ratio is based on the price of the acquiring company’s stock when
the deal is closed or those in which the target firm can call off the merger if the
acquirer’s stock falls below a certain value. In these situations, the manager has to
adjust its hedges constantly and unwind positions if the price falls below the floor
value. The chosen merger procedure can affect the balance sheet of the combined
company drastically. While cash offers have to be accounted for by the ‘purchase
method’ in the USA, in mergers processed through an exchange of stock, the compa-
nies can under some circumstances employ the ‘pooling method’.13

EXAMPLES
The following two examples illustrate the way returns can be achieved for
a stock swap transaction and a cash tender offers.14

Stock swap
Assume Bank A and Bank B announce a merger structured as a stock swap
in which each holder of one share of Bank B is entitled to receive 0.84
shares of Bank A. The merger is to be completed in 90 days. Bank A and
Bank B pay dividends of 39 cents and 37 cents respectively during this
period. Assume a manager buys 10,000 shares of Bank B at 281/2 per share
for a total cost of $285,000 and sells short 8,400 shares of Bank A at 351/2

.... 64 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 65

3 ■ AIS: Hedge funds and Managed Futures

per share for a total credit of $298,200. Because of the leverage (2:1), the
manager will be required to post 50% of the value of the long position
($142,500) and 50% of the value of the short position ($149,100) with the
prime broker, thus resulting in a total cash outlay of $291,600. Assume the
manager finances the balance of the long position ($142,500) at an annual-
ized rate of 63/4 % for 90 days for a total cost of ($2,371.75). Further
assume that the manager receives a short stock rebate of 5.1% annualized
on $298,200 for the period of 90 days, which is $3,749.97 and that the
manager’s transaction costs equal ($184) or one cent per share. The net
profit on this transaction upon completion of the merger is as follows:

Bank A / Bank B spread $ 13,200.00

Transaction costs (184.00)
Bank A dividend (short) (3,276.00)
Bank B dividend (long) 3,700.00
Financing costs (2,371.75)
Short stock rebate 3,749.97
–––––––––
Profit $ 14,818.22

$ 14,818.22 365
Annualized return is ––––––––– × ––– = 20.61%
$ 291,600.00 90

Cash tender offer

Assume X Corp. receives a $17.50 cash tender offer from Y Corp. on
February 1 and that the tender offer expires on March 5 . Further assume
that the price of X Corp. was 17.25 on February 1 and that the manager
purchases 500,000 shares at that price. The cash outlay for a manager will be
50% (leverage 2:1) of the purchase price of the stock ($4,312,500). Assume
the manager can finance the remaining balance of $4,312,500 for 34 days at
a rate of 63/4% per annum for a cost of $27,115.58 and that the manager’s
transaction cost is one cent per share ($5,000).

.... 65 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 66

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The net profit therefore is:

Net cash proceeds $ 125,000.00
Financing costs (27,115.58)
Transaction costs (5,000.00)
––––––––––
Profit $ 92,884.42

Annualized return is $ 92,884.42 365 days

––––––––– × –––––––– = 23.12%
$ 4,312,500 34 days

Sources of return
The inherent return of Risk Arbitrage strategies is the risk premium for taking
exposure in deals with uncertain closing. This is referred to as the ‘deal risk pre-
mium’. The Merger Arbitrage manager assumes the risk that the deal could fail in
which case the manager assumes large losses. The strategy is thus similar to a
short option on the close of the deal.

Risk factors
The downside risk of a Risk Arbitrage position is large compared to the expected
return if the merger succeeds. The premium paid by the acquirer in a merger deal,
which is the difference between the acquired company’s market price right before
the merger announcement and the offer made by the acquirer, is usually large
compared to the remaining spread between offer price and market price after the
announcement of the deal. The risk profile of a typical Risk Arbitrage position is
thus comparable to a short position in an option. The manager receives a (more
or less predetermined) risk premium for entering the position in return for accept-
ing the risk of significant losses if the merger fails.
Risk Arbitrage positions are a ‘bet on the merger completion’. The strategy’s
principal risk factor is thus deal risk. Deal risk includes everything that affects the
completion of the merger or the timing of it, e.g. the risk of the acquirer getting

.... 66 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 67

3 ■ AIS: Hedge funds and Managed Futures

out of the deal, the risk of lack of shareholder support, or the risk of (federal or
state) regulators not approving the deal. These risks are influenced by a number of
different factors:

■ Financial position of the acquirer. Certain market factors (interest rates rise,
stock market downturn) can worsen the financial status of the acquirer leaving
him unable to finance and complete the transaction.
■ Earnings. A sudden decline in the target’s earnings may make a deal fail.
Different market factors can cause earnings to deteriorate.
■ Size of the premium. The higher the premium (difference between offer price
and market price before the merger announcement) paid in the transaction the
higher the downside risk.
■ Market sentiment. A general downturn in broad equity markets can increase
insecurity about mergers going through.
■ Regulatory concerns. Regulators influence the deal uncertainty spreads as they
affect directly the likelihood of deal completion. An example of regulatory
disapproval causing a deal to fail is the General Electric/Honeywell merger in
June 2001. When the European Commission announced that it was unwilling
to approve the merger for anti-trust reasons, the spread widened from 5 to
36% within two days (see Figure 3.6).
■ Consideration. The form of the proposed transaction (stock swap, cash or any
combination) can influence the deal probability.
■ Time to completion. A delay in the completion of the deal can significantly
reduce the expected return.

Cash deals are not completely market neutral, because the manager cannot hedge
market risk with short position in the acquirer’s stock. In case the merger is not
completed the position is additionally exposed to the risk of market declines. Other
risk factors of the strategy are liquidity risk and deal flow risk. Lack of sufficient
market depth and liquidity can become a problem for small cap positions. The
returns of Risk Arbitrage managers depend on the number and the ‘quality’ of
merger transactions. In times when merger activity is high (as in 1999) Merger

.... 67 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 68

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Arbitrage is very profitable compared to environments where deal flow is declining

and deal quality deteriorating (as in early 2001). The returns earned by Risk
Arbitrage strategies are not independent of general equity valuation levels. Deal flow
and quality are usually higher in equity bull markets. Companies are more likely to
initiate mergers in times of rising equity markets, when the economic outlook pro-
vides the necessary comfort and high stock prices provide ready currency for
mergers. Sharp equity declines can widen spreads and increase deal risk.
Consequently, Risk Arbitrage strategies face market risk in forms of extreme volatil-
ity and deteriorating deal flow. The strategy is also exposed to interest rate risk, as
significant and unexpected rises in interest rate increase borrowing costs and make it
more difficult for the acquirer to finance the deal. Furthermore, higher interest rates
might negatively affect the acquired company’s earnings.
As Merger Arbitrage strategies involve short selling of stock, the strategy is
exposed to the specific risks of Short Selling (shorting risk) connected to the fact
that the short seller has to borrow stock (e.g. short squeezes, risk of dividend
increase, execution risk). These risks will be discussed in detail in the section on

60 45%

40%
50
35%

40 30%

25%
30
20%

20 15%

10%
10
5%

0 0%
22.06.01
15.06.01
08.06.01
01.06.01
25.05.01
18.05.01
11.05.01

GE HON Spread

FIGURE 3.6 ■ Price and spread development for the GE–Honeywell deal (rejected by the European
Commission on 15 June 2001)

.... 68 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 69

3 ■ AIS: Hedge funds and Managed Futures

Short Selling. Operational risk factors include unfavourable trade executions (bad
fills, slippage, clumsy order execution by the broker) and wrong information (this
includes fraud risk due to manipulated financial statements).
Highly volatile markets combined with increasing interest rates create an envi-
ronment where Risk Arbitrage strategies are the most vulnerable. Insecurity about
earnings, stock prices and future economic outlook usually decrease the number
of announced deals dramatically and mergers are more likely to break up. An
example for an unfavourable period for Risk Arbitrage strategies was the first half
of 2001, when Risk Arbitrage managers suffered significant losses. Generally, peri-
ods of falling interest rates and normal to low market volatility are positive for the
strategy. It is interesting to note that positive and negative market environments
for Risk Arbitrage strategies are somewhat the opposite to those for Convertible
Arbitrage strategies. A combination of these two strategies provides good diversifi-
cation benefits in a multi-strategy AIS portfolio.

Event Driven – Distressed Securities

The strategy
Distressed Security strategies invest in or sell short securities of companies that
have been or are expected to be affected by financial difficulties. ‘Distress’ can be
caused by any number of the following: liquidity problems; excess debt; opera-
tional or strategic shortcomings; changes in the competitive marketplace; legal or
regulatory difficulties. Distressed Securities managers invest in bank debt, bonds,
subordinated debt, trade claims, letters of credit, common stock, preferred stock
or warrants. There may also be opportunities to buy specific assets being spun off
by distressed companies that need to raise cash.
While the level of interest in Distressed Securities has increased over the years,
the market remains relatively illiquid. Many private and institutional investors are
not willing to hold Distressed Securities in their books. For years distressed investing
was seen as a ‘dirty’ business because investors attempt to profit from other’s mis-
fortune. Now, however, distressed or ‘vulture’ investors have begun to lose their
‘bad guy’ image, as many institutions have come see the benefits of an active second-

.... 69 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 70

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

ary market for distressed debt. This has increased the overall amount of debt avail-
able for purchase and improved liquidity in primary markets.
Investing in Distressed Securities requires specialist expertise and extensive
pre-investment analysis in order to allow for an accurate assessment of the impact
of restructuring negotiations and/or bankruptcy proceedings on the value of a
company and its underlying securities.
Managers generally take a core position in a distressed company’s securities
and hold it throughout the restructuring or bankruptcy process. The industry dis-
tinguishes between managers who participate actively in creditor committees and
assist the recovery or reorganization process and ‘passive managers’ who buy and
hold Distressed Securities until they appreciate to the desired level. Some strate-
gies are particularly focused on short-term trading in anticipation of specific
events, such as the outcome of a court rule or important negotiations. Managers
may use some leverage, but generally the level of leverage is fairly low. Another
form of a Distressed Securities strategy is ‘Capital Structure Arbitrage’, where the
manager purchases undervalued securities and sells overpriced securities of the
same firm, e.g. long mezzanine debt and short common stock. Fund managers
may run a market hedge using S&P put options or put options spreads. Distressed
managers can benefit substantially from creative financial engineering.
Examples of investment opportunities in distressed investments include the
following:

■ Anticipated Restructuring. A corporate restructuring event caused by financial

problems is pending. This can be the announcement of a restructuring plan, a
change in management or the board, a refinancing or even a
merger/acquisition. Often the restructuring event will lead to a significant
change in the management of assets and allocation of capital. Careful analysis
may identify value post-restructuring that is not yet reflected in the price of
the company’s securities.
■ Chapter 11 Reorganizations. The final stages of bankruptcy reorganizations may
create attractive investment opportunities. The securities of companies undergoing
Chapter 11 Reorganizations are often underfollowed by Wall Street analysts and
may trade at depressed valuations versus the firm’s fundamental value.

.... 70 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 71

3 ■ AIS: Hedge funds and Managed Futures

■ Legal/Regulatory Overhang. Legal or regulatory issues which are not fully

understood may depress the valuation of the securities to values below their
fundamental fair value.
■ Corporate Restructurings/Spin-offs. An investor may provide buyout capital in the
form of equity or debt for privatizations (leveraged buyouts from management),
spin-offs, acquisitions or takeovers. Significant changes in the management of
assets and allocation of capital can amplify and accelerate positive financial trends.

EXAMPLE
Company ABC is unable to meet its debt amortization and is negotiating a debt
restructuring with a consortium of banks. The long-term debt of ABC with an
11% annually paid coupon is trading at 55% of par value due to the firm’s liq-
uidity problems and fear of bankruptcy. The manager buys the company’s debt
for 55% of par value after determining that the company is likely to reach an
agreement with its banks to stretch out debt repayments and therefore avoid
bankruptcy, and that the longer term prospects of the firm are good. After 12
months, due to a restructuring of the firm’s debt and first signs of financial
recovery, the long-term debt can be sold for 70% of par value and the firm is
able to meet interest payments on its debt. The annualized return of the invest-
ment (capital gain of 27.3% plus interest income of 20%) equals 47.3%.

Sources of return
Distressed Securities strategies earn a risk premium for being exposed to credit
and other company specific event risks. Further, the investment is usually locked
in for an extended period of time, as the market for the investment is likely to be
very limited. This lack of liquidity corresponds to a liquidity premium earned by
the investor. Returns of Distressed Securities strategies strongly depend on the
skill and experience of the manager. There are many un- or under-researched and
attractive investment opportunities, as there are generally very few analysts look-
ing at such situations. Managers have thus the opportunity to capitalize on
restricted availability of information and other market inefficiencies.

.... 71 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 72

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Risk factors
Investments in Distressed Securities strategies are generally very illiquid. Extended
redemption periods are the norm for Distressed Securities investments and most
managers seek a long-term commitment of the investors’ capital. Bankruptcy pro-
ceedings are, by nature, very time consuming, as are restructuring negotiations
outside bankruptcy, and their outcome is very uncertain. Distressed Securities
investments have characteristics that are similar to private equity investments.
The principal risk of Distressed Securities is corporate and situation-specific
risk. The strategy requires correct valuation of the underlying business and its
individual securities using a ‘bottom up’ analysis. The release of adverse informa-
tion or the occurrence of certain events can be extremely damaging to the value of
the investment (event risk). Any investment in distressed debt involves significant
credit risk in the form of bankruptcy risk.
Investments in Distressed Securities bear high liquidity risk. Often there is no
regular market for the security, and the positions are extremely difficult to value
(mark to market risk). As with private equity investments, the investor has to
commit invested capital for a longer period of time and might suffer significant
losses if he redeems prematurely. The termination date of the investment is not
known in advance (timing risk). Further, for distressed bank debt, the manager is
exposed to settlement risk, as the trade documentation process can be rather com-
plex and takes a significant amount of time.
Distressed Security strategies are heavily exposed to legal risks. Legal questions
become particularly important once insolvency procedures or litigation are initi-
ated. Chapter 11 of the US bankruptcy law provides relief from creditor claims
for companies in financial distress in order to allow for an orderly restructuring
process. The objective is to save distressed companies from liquidation. In Europe
and most other countries, however, bankruptcy laws are less debtor friendly, so
the chances of a company surviving as a going concern are reduced. Important
issues in the restructuring process are debt-restructuring practices and laws, rank-
ing of creditor claims (particularly in relationship with employee and tax claims),
disclosure policies, voting rights of creditors, debt terms and the treatment of

.... 72 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 73

3 ■ AIS: Hedge funds and Managed Futures

secured claims. Further, distressed companies may face lawsuits or legal threats
that could significantly impact the company’s financial position. The ‘workout’
process requires legal expertise and the navigation of legal pitfalls.
Besides bearing large exposure to firm-specific risks, Distressed Security strate-
gies are also exposed to general market risk factors. Especially in environments of
falling equity markets and rising interest rates, the performance of distressed secu-
rities investment is highly correlated to the broader market. A generally negative
equity market perception worsens the outlook for distressed companies signifi-
cantly and rising interest rates make it more difficult for companies to refinance
themselves. Furthermore, long duration debt positions are exposed to duration or
interest rate risk. For non-publicly traded instruments such as bank debt and trade
claims, the investor is exposed to settlement risk as trades are complex to docu-
ment and can take months to settle. Besides foreign currency risk as part of the
operational risk of the company, the investor faces foreign exchange risk if the
company goes bankrupt and its claims are all converted into the applicable cur-
rency in the jurisdiction of the bankruptcy.
A market environment favourable to Distressed Securities strategies is generally
characterized by economic growth, rising equity prices and low interest rates. The
strategy is very vulnerable in environments of rapidly falling equity markets and
deteriorating economic conditions. An example was the second half of 2000/first
quarter of 2001, when Distressed Securities strategies had significant losses.

Event Driven – Convertible Debenture

Arbitrage (‘Regulation D’)
The strategy
Convertible Debenture Arbitrage (or ‘Reg D’) strategies involve investing in publicly
listed companies of micro and small capitalization through privately structured
investments. The investment is usually negotiated directly with the company and
takes the form of stocks, convertibles or other derivatives (options, warrants) in
return for an injection of capital. Investments are usually structured in the form of
privately placed unregistered high coupon convertible securities or debentures with

.... 73 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 74

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

maturities ranging from 18 to 60 months which can, upon registration with the
SEC, partially (e.g. 1/18th every month in case of an 18-month debenture) or
entirely be converted into ordinary shares of the issuer. Conversion usually happens
at a predefined discounted price. Investments are made pursuant to an exemption
from registration as provided by Regulation D of the US SEC Act of 1933.
Normally after 75 to 90 days, the SEC declares the registration of the convertible
securities effective. The Regulation D manager can then sell the fully tradable and
registered shares in the public markets and realize the spread between the market
price and the discounted price of the stock he converts into (usually about 15–20%)
as a return. The Convertible Debenture Arbitrage strategy can be seen as
Convertible Arbitrage with privately structured debentures.
Unlike standard convertible bonds or preferred equity, the exercise price can be
floating at a predefined discount or subject to a look-back provision. Convertible
debentures can also have a fixed price and it sometimes is at the investor’s discre-
tion to choose the fixed or floating price on conversion. A floating price has the
effect of insulating the investor from a decline in the price of the underlying
stock. Investments are therefore fully price hedged. In some cases the manager
also receives warrants on top of the debenture.
The process of Regulation D investments requires extensive bottom up analyti-
cal work. Pre-investment due diligence usually consists of:

1 thorough analysis of the firm’s financial position (earnings, cash flow, assets
and liabilities, ratio analysis)
2 examination of the company’s stock liquidity (volume, volatility, market
capitalization, number of market makers, diversity of shareholder, short
interest)
3 assessment of the business and the quality management (interviewing
management, examination of competitive environment, firm’s strategy,
product and market structure, country and sector analysis).

On the sale of the convertible security to the investors, the company begins the
process of filing for registration of the privately placed shares, typically an S-3.
Pursuant to the terms of the offering negotiated between the company and the

.... 74 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 75

3 ■ AIS: Hedge funds and Managed Futures

investor, there is a time period by which the company must submit the registra-
tion to the SEC and cause it to be declared effective (i.e. ‘the common shares
are registered and freely tradable’). The term is generally between 60 and 120
days. The debenture agreement can be structured such that, if the registration is
delayed beyond a certain time, the company has to pay a penalty to the investor.
On approval of the registration, the investors may sell the newly registered
common stock at any time after a specified holding period. The holding period
is negotiated such that the stock is not immediately ‘dumped’ on the market,
but rather sold in an orderly fashion. As a general rule, once the shares are reg-
istered and converted, it will take a month or more to sell the shares in the open
market. Experience shows that most Regulation D transactions, all in all, take
18 months to five years from the day of closing until the conversion and sale of
all shares is completed.
The most critical decision to be made during the negotiation process is the
price at which the investor will be able to convert his position into common stock.
In this regard, the investors must decide whether to fix a ‘price’ of the stock
(which is ‘higher risk – higher reward’), or the less risky approach of protecting a
fixed ‘discount’ to the market price.

EXAMPLE
Company ABC’s stock currently sells for $10.00 per share and on average
130,000 shares are traded per day. The company negotiates with an
investor (Hedge fund manager) for a private placement of a debenture
convertible into its common stock. ABC and its investors agree to a 15%
discount to the share price at the time of conversion with registration to
be completed in 90 days. The share price is defined as the average of clos-
ing bids for the five days prior to conversion. On completion of
registration the investors can exercise their conversion right and begin sell-
ing the shares in the open as described in the offering memorandum. In
this example, assuming that the share price remains at $10 and the invest-
ment is $10 mio, the discounted price would be $8.50 and the number of

.... 75 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 76

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

shares the investor receives equals 1,176,470 shares ($10,000,000/$8.50).

The investor trades out of the position and receives a total sale price
(assumed to be $10 per share) of $11,765,000 or a non-annualized
17.65% profit. If the stock drops to $8.00 before the investor can convert
his shares, under the agreement ABC would now have to issue enough
additional shares at the new discounted price of $6.80, i.e. 1,470,588
shares. Alternatively, the investor could have chosen to ‘lock in’ the share
price of $8.50 and assumed the price fluctuation risk. However, in most
cases, managers choose the ‘protected discount’ structure as opposed to
the ‘protected price’ alternative.

Sources of return
When public companies need to raise capital, e.g. in order to make an acquisition
that has to close quickly, they have three alternatives:

1 debt issuance
2 secondary offerings
3 private equity placements pursuant to Regulation D.

Each of the first two alternatives has certain requirements which make access to
the capital very difficult and time consuming and, in some instances, impossible.
These encumbrances include limited borrowing capacity, high expenses in the case
of a secondary offering and time constraints. These factors, along with other more
subjective issues, cause many companies to pursue the third option, which allows
smaller companies to raise equity capital more quickly, cheaply and conveniently.
Similar to the returns of Distressed Securities strategies, Convertible Debenture
Arbitrage strategies’ returns are mostly related to credit and liquidity premiums
earned for taking positions with lower credit quality that cannot be easily liqui-
dated. Large parts of the returns of the investments arise from smaller companies
conceding an advantageous price to investors willing to hold an illiquid invest-
ment for a certain period (until registration with the SEC occurs). Manager skill
and expertise in evaluating potential investment targets are essential for
Convertible Debenture Arbitrage investments.

.... 76 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 77

3 ■ AIS: Hedge funds and Managed Futures

Risk factors
A Regulation D investment is a debenture from a company in need of capital whose
debt is non-rated. The strategy faces a similar risk profile as private equity invest-
ments. The most significant risk factor of the strategy is credit risk, in the form of
default risk, on the principal amount or on the coupon payments. A debenture with a
floating conversion price provides some marginal protection from default risk (if the
default comes after registration) through the possibility of converting into the com-
pany’s equity. However, tradability and liquidity of the stock may be very low. In
cases where the market capitalization of the company drops to levels close to where
the entire company’s equity is insufficient to cover the debenture, the conversion right
no longer protects from the company’s default. In some cases default risk is decreased
by the issuance of a letter of credit from a bank for part of the debenture.
Important risk factors of the strategy are the registration risk and the related (pre-
registration) liquidity risk. While the usual process of registration with the SEC takes
about 90 to 120 days, significant delays in the process can occur if the company is not
providing necessary information or is facing deteriorating financial conditions. The
length of registration can then increase significantly with an uncertain outcome. Before
registration with the SEC there is no market for investors to liquidate their position.
In many instances, the need to protect their investment during the registration
or the ‘non-conversion’ period leads investors to hedge their investment through
shorting the firm’s stock. The credit risk related to the financial condition of the
company might mandate these hedging activities. This can result in a dramatic
reduction of the company’s market price. The consequence is a dilution of the
stock once the debenture is converted, i.e. the debenture will be converted to
more stock. The industry refers to this as the ‘death spiral’ of Regulation D deals
(and often refers to Regulation D deals themselves as ‘death spiral convertibles’).
Most Regulation D transactions involve companies with micro or small capital-
ization. The necessary trading volume equals a significant percentage of the firm’s
capitalization (especially after a significant stock price decline, when more shares
have to be issued against the Convertible Debenture). Trading out of the con-
verted stock can become rather difficult, despite an SEC registration. The strategy
faces a great deal of (post-registration) liquidity risk.

.... 77 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 78

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Due to the private nature of the transaction, Convertible Debenture Arbitrage

strategies are exposed to legal risk and also to fraud risk; for example, the com-
pany may misinform or use manipulated earnings statements.
Besides bearing large firm specific risks, the Convertible Debenture Arbitrage
strategy is exposed to general market risk factors. The strategy is vulnerable in
environments of rapidly declining equity markets, as falling stock prices increase
the number of shares received from conversion which make liquidation of the
position more difficult. The company itself might also face less favourable busi-
ness conditions in such environments. The duration risk of the debenture depends
on its time to maturity, but is usually rather low.
As for Distressed Securities strategies, a market environment that is favourable
to Convertible Debenture Arbitrage strategies is generally characterized by rising
equity prices and low interest rates. During the boom years of equity markets in
the second half of the 1990s, Convertible Debenture Arbitrage strategies were
among the most outstanding performers of all Hedge fund strategies. With falling
equity prices in 2000/2001, however, Convertible Debenture Arbitrage strategies
suffered significant losses.

Opportunistic – Global Macro

The strategy
Global Macro strategies are characterized by making (generally leveraged) bets on
anticipated movements of equity, interest rates, foreign exchange rates and com-
modity prices. Global Macro managers employ a fundamental and/or systematic
‘top down’ approach and may invest in any market or instrument in order to
profit from expected market moves. These moves may result from forecasted
shifts in world economies, political developments or changes of global supply and
demand for physical and financial resources. Managers typically earn returns by
having access to better information or by their ability to analyze existing informa-
tion more accurately than others.
Global Macro strategies assume that prices are determined by fundamental eco-
nomic factors that affect the supply and demand of a financial asset. They consider
investors to be rational in the long term and believe that prices will sooner or later

.... 78 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 79

3 ■ AIS: Hedge funds and Managed Futures

converge to levels consistent with investor rationality as determined by fundamental

economic factors. In the short run, however, prices can deviate substantially from
their true economic equilibrium in which all available information is correctly and
fully discounted. This occurs because information is not a free good. Investors must
spend time and resources to acquire and properly interpret new data. New informa-
tion is incorporated gradually rather than instantaneously into market prices.
Furthermore, price volatility may cause investors to react emotionally in ways not
consistent with a dispassionate analysis of economic factors. Finally, trade restriction
(e.g. currency regimes) can prohibit prices from moving to their real economic
value. The goal of Global Macro strategies is to identify long-term macroeconomic
trends in times when prices deviate substantially from the price that corresponds to
their true fair value. The Macro strategy does not necessarily try to make money
from market inefficiencies, but rather by anticipating price changes (reversals) early.
Global Macro managers employ macroeconomic forecasting models, which can
be divided into technical and fundamental models. Technical analysis predicts
market returns by looking for trends and patterns in historical prices.
Fundamental approaches use different micro- and macroeconomic indicators to
predict expected returns. Often a combination of the two is used.
Global Macro managers usually have one of two backgrounds: Long/Short
Equity or derivatives (mostly Futures) trading.15 The first sort of manager operates
in liquid equity markets, mostly large cap stocks in the USA, Asia and Europe. The
second class of manager invests in liquid derivatives markets such as fixed income,
foreign exchange and equity index Futures. Both use exchange-traded and over-
the-counter derivatives to magnify returns. Some currency and Futures managers
declare themselves Global Macro traders, as they trade based on a fundamental
analysis of underlying economies. Currency and Futures managers are examined in
the Futures strategy section that follows later in the chapter.
An example of a famous Global Macro trade was the short position on the
British pound sterling against continental European currencies (mainly the Deutsche
Mark) Global Macro managers held in 1992. Despite increasing pressure for a
devaluation of the British pound, the British central bank and other European cen-
tral banks continued to intervene in order to support the currency and keep it
within the 3% boundaries defined by the European Monetary System. Eventually it

.... 79 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 80

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

became clear that the British currency had to be devalued. When the central banks
decided to let the currency float freely, the currency dropped about 20% in value
and pound short positions became extremely profitable. Soros’ Quantum fund is
said to have earned more than $1 billion in the course of the devaluation.16

Sources of return
Global Macro managers consider themselves acting in markets which are not com-
pletely efficient and claim to earn a return by exploiting temporary price anomalies.
Their goal is to anticipate long-term macroeconomic trends in times when prices devi-
ate substantially from the price that corresponds to their true fair value, which can
happen for a variety of reasons (slow information flow, emotions, regulatory con-
straints). They try to anticipate the economic developments in the coming months and
to exploit the predictability of economic cycles (it is disputed among economists,
though, that there is such predictability). For most opportunistic strategies it is difficult
to identify specific risk premiums earned. Global Macro managers move between a vari-
ety of different strategies and capitalize on mispricings and price moves depending on
available opportunities. Their strategies rely heavily on the manager’s skill, which is the
most important source of returns generated, but also most difficult to assess properly.

Risk factors
Global Macro strategies involve leveraged directional bets on market moves.
Managers move ‘from opportunity to opportunity, from trend to trend, from strat-
egy to strategy’.17 There is therefore no homogeneous risk profile of Global Macro
strategies. Global Macro strategies are exposed to most of the typical Hedge fund
risk factors (market risk, credit risk,18 liquidity risk, event risk, corporate risk, for-
eign exchange, operational risk, fraud risk, regulation risk, legal risk).
Ultimately, the success of the strategy comes down to a manager’s ability to
predict market moves. The strategy is especially exposed to market risk, i.e. the
market moving against the directional positions. Market risk often comes in the
form of event risk, i.e. the risk of macroeconomic and market developments
coming out differently than expected by the manager. This includes political
developments, sudden currency moves that are often related to regime changes

.... 80 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 81

3 ■ AIS: Hedge funds and Managed Futures

(fixed to floating), sudden earnings revisions by key companies, important merger

announcements or key firms’ changes of management.
In March 2000, Julian Robertson announced the closure of the Tiger Funds blam-
ing ‘irrational markets’ for the funds’ poor performance. His statement read: ‘Earnings
and price considerations take a back seat to mouse clicks and momentum.’ At the same
time George Soros cut back his $8.5 billion Quantum Fund after having faced double-
digit losses. Evidently the two largest long-term Macro players were unable to cope
with ‘irrational markets’ during the second half of 1999 and first quarter of 2000.19
Global Macro strategies act worldwide. Investments in foreign securities bear
foreign exchange risk. Some managers invest specifically in foreign exchange mar-
kets predicting depreciation or appreciation of specific currencies. Others invest
in foreign equities and bonds bearing the implicit risk of the underlying currency
moving against the position.
Global Macro strategies also bear model risk, which is the risk that the quanti-
tative method or the fundamental model is inappropriate for the instruments or
market invested in. Execution risk in the form of slippage or clumsy order execu-
tion by the broker is a further operational risk factor. Global Macro funds have
been among the largest Hedge funds in size, a fact that magnifies execution risk
and exposes the investments to liquidity risk. Some Global Macro strategies
involve short selling of stock causing the strategy to be exposed to the specific
risks of Short Selling, i.e. short risk (see section on Short Selling strategies).
Many investors believe that the closure of the Tiger Funds and the downsizing of
George Soros’ Quantum Fund provide evidence that the days of successful Global
Macro investing are over. Global Macro strategies are most successful in inefficient
markets. One can observe increased market efficiency (mainly through modern
communication and information distribution tools like the internet) throughout
global financial system, a development that clearly works against the Global Macro
strategy. The year 2001 has seen a revival of Global Macro managers, however,
which have been among the best performing during the equity bear market in
2000/2001. New immature markets with a lower degree of efficiency (such as
financial products related to electricity or weather) have emerged. Investors’ money
has flown back to the strategy sector and only time will tell whether Global Macro
managers can continue to perform in increasingly efficient markets.

.... 81 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 82

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Opportunistic – Long/Short Equity

The strategy
Long/Short Equity (also called ‘Equity Hedged’) is by far the most prevalent strategy in
the AIS industry. New investors in AIS tend to think of Long/Short Equity strategies as
synonymous with Hedge funds. The Long/Short strategy consists of investing in a core
holding of equities and then partially hedging the long equity position with short sales
of other stocks and/or stock index derivatives (this corresponds in principle to the
‘Jones strategy’ of 1949). In contrast to the Equity Market Neutral strategies, oppor-
tunistic Long/Short Equity strategies generally have a long bias, i.e. their long market
exposure is not completely offset by short positions.20 The focus of the strategy is stock
selection with regional, sector-specific or particular style emphasis (e.g. US equity,
European equities, high tech focus, value versus growth) rather than market timing or
prediction of the broad stock market direction. Ideally, the return profile of a
Long/Short Equity manager resembles a call option with full or partial participation
(and possibly out-performance) of the broad market in bull markets and protection
against losses (and ideally net gains from the short positions) in bear markets.
The Long/Short Equity sector is far from homogeneous.21 The amount of net
exposure to the broad equity markets varies widely among different managers as
do concentration levels in specific industry sectors. Conservative managers miti-
gate market risk by maintaining net market exposure ranging from 0 to 100%
while aggressive funds may magnify market risk by employing leverage exceeding
100% market exposure, e.g. 160% long and 50% short. In some rare instances,
managers maintain a short net exposure.
The Long/Short Equity community distinguishes between ‘bottom up’ strate-
gies and ‘top down’ approaches. The former follow individual companies and
invest on the basis of fundamental analysis. The net equity exposure of a manager
following a bottom up approach is based on his views regarding the extent to
which the stocks of individual companies are over- or undervalued rather than
being a function of expected broad market developments. The latter try to iden-
tify economic trends and upcoming ‘investment themes’ in certain asset classes
and industry sectors (e.g. technology, biotech, growth companies) and change
their allocation and net long exposure based on expected trends. Fundamental

.... 82 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 83

3 ■ AIS: Hedge funds and Managed Futures

analysis, nevertheless, is an important element of top down investing. Some

Long/Short Equity managers possess detailed knowledge and experience in spe-
cific industries and therefore focus their efforts on single sectors such as
technology, healthcare, energy or financial companies, while others invest across
various industries. Some managers have a particular focus on micro and small cap-
italization stocks, where potential inefficiencies are most prevalent.
Compared to Long Only strategies, Long/Short Equity strategies have signifi-
cant flexibility regarding the use of leverage and short positions. The manager
targets returns on the long side as well as on the short side, which ideally results
in the doubling of alpha. Short positions have three different purposes and man-
agers differ by the varying degrees they emphasize each one of these. First, the
short positions are intended to generate ‘alpha’; second, the short positions serve
the purpose of hedging market risk; and third, the manager earns interest on the
shorts, as he collects the short rebate. Some managers use Futures and/or puts on
the S&P500 index for hedging purposes.
Fundamental analysis refers to the examination of the different factors that affect a
company’s stock price. Sources of information are research reports from banks, jour-
nals and newspapers, trade shows and industry contacts, company visits and personal
interviews with management, conference calls and press conferences, as well as discus-
sions with competitors, suppliers and distributors. Managers examine the individual
company in terms of earnings, cash flow, balance sheet quality, assets and liabilities,
ratio analysis etc. This goes along with an analysis of the industry sector with focus on
the industry’s competitive environment, the demand and supply situation, industry
profitability and foreign influences. Qualitative analysis involves a detailed due dili-
gent process and analysis of management, the product and the market.
As most managers have limited capacity to conduct fundamental research, they
often focus on a limited universe of stocks, which can be defined through a
screening process involving market capitalization, industry sectors, value versus
growth or particular investment themes. Long/Short Equity managers diversify
their portfolio by holding a larger number of different positions and many man-
agers diversify through different industries and sectors. The construction of a
well-diversified portfolio requires careful attention to the relationship between
the different holdings of the portfolio.

.... 83 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 84

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

EXAMPLES
Typical examples of Long/Short Equity trades are the following:

■ Long undervalued countries/sectors/individual stocks, short overvalued

countries/ sectors/individual stocks.
■ Core long positions with a partial hedge overlay with short index
Futures positions, long out-of-the-money puts or short covered call
positions.
■ Short-term trading of initial public offering stocks (‘hot issues’).

Sources of return
Long/Short Equity strategies participate in the economic function of capital for-
mation and distribution. The managers are normally net long and are therefore
exposed to broad market risk. A part of Long/Short Equity returns is therefore
market risk premium, i.e. ‘beta return’ (depending on the net long position). Beta
has arguably been a major source of the high absolute returns for Long/Short
Equity strategies in the equity bull market of the 1990s. But Long/Short Equity
managers claim to return significant alpha, too. The discussion about the man-
agers’ alpha generation can follow the same lines as earlier for Equity Market
Neutral strategies. ‘Alpha’ is partly a consequence of particular risks taken such as
small firm risk, recession risk and company leverage risk. Certainly, the specific
manager’s stock-picking skills and trading talents are a meaningful source of gen-
erated returns, but the precise contribution of manager skill versus risk premiums
to Long/Short Equity returns remains an open question. Market inefficiencies
(and the possibility of generating alpha by detecting and exploiting them) are
most prevalent for stocks of companies with small capitalizations, as those are
often not well covered by analysts. (‘Wall Street is quite lazy’, I once heard a man-
ager saying with respect to small and micro cap stocks.)
One important feature of Long/Short Equity strategies is that managers are not
limited in generating ‘alpha’ by the constraint of having to ‘hug a benchmark’.

.... 84 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 85

3 ■ AIS: Hedge funds and Managed Futures

They do not carry ‘dead weight’ portfolio position in which they have no insight
or conviction for the sole reason of controlling idiosyncratic risk (i.e. the risk of
underperformance relative to an index). Only positions for which the manager
has performed his research and has a strong conviction about are held. The art of
the manager is to control non-market residual risk through other less capital-
intensive techniques than holding ‘dead weight’ positions.
The ability to sell short creates unique profit opportunities. Many investors face
psychological or regulatory barriers to engaging in Short Selling. Regulatory author-
ities or internal compliance rules prohibit many financial institutions and asset
managers from Short Selling. This creates market inefficiencies and price anomalies
which Long/Short managers benefit from. When Short Selling is restricted, market
prices might not be entirely efficient, as without complete freedom to sell short, the
pessimism of some investors is not fully represented by security prices and some
stocks will be overpriced relative to others. Bankruptcy proceedings or restructur-
ing can create unique profit opportunities for Short Selling.

Risk factors
Long/Short Equity strategies usually have a net long exposure. The resulting correla-
tion to the broad equity market or particular sectors causes the strategy to be exposed
to market risk. The determination of what level of net exposure to the market to hold
requires the trading manager to perform some assessment of the future direction of
the market. Declining equity markets usually lead to losses for the strategy, although
less so than Long Only Equity strategies because of the partial hedge with short posi-
tions. For the evaluation of a Long/Short Equity manager, it is important to
understand the source of a strategy’s alpha in order to distinguish the managers who
benefit mostly from beta from those which are ‘real alpha generators’.
Depending on the manager, Long/Short Equity strategies have specific exposures
to certain industry sectors (sector risk), countries (country risk) and currencies
(FX risk), company sizes (market capitalization) and styles (value/growth).
Ultimately, the success of the Long/Short Equity strategies comes down to stock
selection. The strategy is exposed to (corporate or sector) event risk in its individual

.... 85 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 86

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

holdings or the corresponding sectors invested in. Specific-company or sector risk is

anything that affects the values of the company or the sector in an unforeseen way,
e.g. the risk of surprise announcements like earnings revision, merger announce-
ment or changes of management. An unbalanced and non-diversified portfolio
enhances the risk of significant losses from individual stock events.
As Long/Short Equity strategies involve short selling of stock, the strategy is
exposed to the specific risks of Short Selling (short risk) connected to the fact that
the short seller has to borrow stock (e.g. short squeezes, execution risk). It is
important to note that the process of shorting stocks requires a unique set of skills
and experiences, which are different from buying stock on the long side only. A
pure fundamental approach to picking short candidates is much less appropriate
than the equivalent strategy for long only strategies. The internet excitement in
1999 and early 2000 provides a good example of a situation in which managers
who sold stocks short because of fundamental overvaluation got badly hurt and,
in many cases, went out of business. Experienced short sellers look for a catalyst
that will expose the poor fundamentals of the corresponding company. Many
Hedge fund managers coming from Long Only asset management have gone
through painful learning experiences in handling the short side. An investor has to
be aware of the ‘shorting learning curve’ and should avoid letting a Hedge fund
manager gather his first experience with Short Selling at his expense.
If the strategy trades small cap stocks (on the long or the short side), liquidity
risk becomes a considerable risk factor. Some Long/Short Equity managers hold a
significant number of positions simultaneously, so their strategy can bear signifi-
cant operational risk.
Long/Short Equity strategies tend to perform very well in periods of rising
equity markets benefiting from a net long exposure, while during periods of
falling equity markets the same net long exposure usually leads to losses.
Nevertheless, even in falling equity markets, Long/Short Equity managers are able
to generate alpha, i.e. outperform an equity market index (but as an ‘absolute
return strategy’ it does not have a benchmark).

.... 86 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 87

3 ■ AIS: Hedge funds and Managed Futures

Opportunistic – Equity Market Timing

The strategy
Equity Market Timing strategies (also referred to as ‘Mutual Fund Timing’) utilize
a variety of statistically based technical trading models which screen the global
equity markets to identify temporal short-term opportunities (trends, momentum)
in individual industry sectors (e.g. financial, pharmaceutical, biotechnology,
telecommunication etc.). The managers opportunistically switch between long
equity exposure and money market instruments or short-term bond investments.
Long equity positions are taken in favourable moments, i.e. an established upward
trend market and the position is neutralized in less favourable moments, i.e. the
start of a downward trend. Trading mutual funds offers the most price efficient
method for the implementation of an Equity Market Timing strategy. Capital allo-
cation to different sectors is based on relative opportunity and level of volatility. In
rare cases, managers hedge using stock index Futures. This may occur in times of
suddenly perceived downside risk.
Typically, managers employ technical trendfollowing models based on short-
and mid-term momentum indicators to determine the direction of a sector and to
identify buy and sell signals. On an ‘upward move buy signal’, money is trans-
ferred from a money market fund into a specific equity sector mutual fund in an
attempt to capture a short-term capital gain. After a downward move the posi-
tions in the mutual fund are sold and the assets moved back into a money market
fund for safe keeping until the next potential upward move occurs. Typically, the
manager is invested in money market funds about 50% of the time and invests
only if the odds are highly in favour of an upward move on the following day.
Two types of Market Timing strategies can be distinguished:

■ Sector Timing: This momentum strategy aims at profiting from micro upward
trends in single industry sectors. These trends occur as good news (e.g. an
earnings surprise) about a particular company within a sector affects the stock
price of other companies in the same sector. The information is not processed
immediately, but over a time period of one to several days. Particularly, small

.... 87 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 88

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

capitalization stocks, which are infrequently traded, tend to lack the

immediate price action of the broad market, a phenomenon referred to as
‘stale prices’.22 This leads to Time Arbitrage opportunities that can be
exploited by shifting into a corresponding mutual fund on a rise in the sector.
■ Time Zone Arbitrage: This strategy exploits an inefficiency in the pricing of US
mutual funds that is based on the conditional price correlation between
different time zones. US-based mutual European and Asian equity funds can
be purchased right around US market close at European/Asian closing prices.
Particularly after weak European or Asian markets and a late US rally there is
a high probability that these funds trade higher on the next day. Typically, the
manager is invested in the equity market only 10–20% of the time and
remains in safe money market funds otherwise.

Trading highly liquid US mutual funds (such as Kemper, Franklin Templeton,

Pimco etc.) allows the manager to buy and sell market exposure without a price
spread (trading at mid-market price) or other transaction costs (no-load mutual
funds). This enables the trading advisor to earn a return even on micro moves, as
the margins are not taken away by trading costs.

EXAMPLE
A real example of a typical trade is illustrated in Figure 3.7. After a few
days of weak equity market performance, the US stock market, based on
incoming good news, rallies strongly one hour before market closes. The
manager invests in US-based international mutual funds shortly before the
market closes assuming that, on the next day, Asian and European markets
will perform similarly well. As the international funds are priced at the
local closing prices (e.g. evening European time, i.e. noon US time), there
is a high probability a profit can be taken, once these funds are sold at the
end of the following day.

.... 88 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 89

3 ■ AIS: Hedge funds and Managed Futures

S&P500 Index German DAX Index

Reversal after a sideways movement Break out after S&P500 turns broke out

After a couple of non-trending sideways Now look what happens the other day
days the S&P500 equity index breaks out (February 23rd) in Europe after the
very sharply on February 22nd S&P500 equity index broke out upwards
the night before!

FIGURE 3.7 ■ Typical Time Zone Arbitrage trade by Equity Market Timing strategy (Copyright 2002
Bloomberg LP. Reprinted with permission. All rights reserved.)

Sources of return
For Equity Market Timing, manager skill does not play as dominant a role as with
other opportunistic Hedge fund strategies. The strategy is reactive rather than
predictive. Equity Market Timing strategies participate in the economic function
of capital formation and distribution. Part of their return is the risk premium for
being exposed to the broad equity market, which in the terminology of the CAPM
is beta. But a significant part of the return is based on exploiting particular ineffi-
ciencies of equity markets generally, and the pricing of US mutual funds in
particular, which are expressed by certain statistical relationships involving differ-
ent geographical regions, sectors and firm sizes:
■ Individual equity sectors show trending behaviour.23
■ Small cap securities (and funds or baskets thereof) tend to lag the price action
of large capitalization securities by one to two days.24
■ Regional equity markets have different market opening hours (time zone
difference).
.... 89 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 90

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Note that it is likely that these inefficiencies will disappear or be reduced

over time.

Risk factors
Equity Market Timing strategies enter and exit equity positions very opportunisti-
cally. The strategy is mainly exposed to market risk or sector risk, whenever
positions are taken up on a signal. Note that the strategy is not exposed to general
market risk, e.g. the risk of a bear market, but rather to conditional market risk, e.g.
a sudden reversal of an up trend.
Mutual Fund Timing strategies are not very popular among mutual fund compa-
nies. An important challenge for the manager is to find the right funds for the
implementation of the strategy. Frequently, mutual fund companies try to make it un-
attractive for Equity Market Timing managers to invest in their funds by imposing
front-load fees, switching costs or redemption penalties. Recent developments in the
mutual fund industry are very unfavourable to the strategy. Mutual fund companies are
increasingly resistant to Mutual Fund Timing. The SEC recently issued a recommenda-
tion to mutual funds to perform fair pricing which will stop market timers using their
funds for Time Zone Arbitrage. The strategy is exposed to the operational risk of sud-
denly not being able to trade in specifically attractive mutual funds any longer. Further,
operational risk is present due to the daily switching within different mutual funds.
This leads to significant execution risk in the form of wrong or missed trades.
Unfavourable environments for Equity Market Timing strategies are choppy
market conditions with a high tendency for trend reversal on a time scale of one
day, i.e. strong upward trend followed by a strong downward move the next day.
Examples for such market behaviour are the months of February and March 2001,
when Timing strategies suffered unusual losses. Favourable markets are continu-
ously upward-trending markets, as seen in 1999 and early 2000. In extended bear
markets, which are not interrupted by short-term rallies, the strategy’s perform-
ance is largely determined by prevailing risk free interest rates, since managers
remain invested in money market funds. The strategy will, however, generate sig-
nificant alpha (outperformance compared to the broad equity market) in bear
markets that display periods of short-term rallies (i.e. a few days or weeks).

.... 90 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 91

3 ■ AIS: Hedge funds and Managed Futures

Opportunistic – Short Selling

The strategy
Short Selling strategies involve the sale of stocks not currently owned by the seller
in order to take advantage of their anticipated price decline. To initiate a short
sale, the seller has to borrow the securities from a third party (the securities lend-
ing department of a broker) in order to make delivery to the purchaser. From the
proceeds of that sale he receives cash, which earns interest called the short rebate
interest. At a later time, the seller returns the borrowed securities to the lender by
purchasing back the securities in the open market. If the short seller can buy the
stock back at a lower price, a profit results. If the price of the stock has risen, a
loss will occur. Theoretically, the short sale (just like a Futures contract) does not
require any initial capital, but in practice, the stock lending brokerage firm
requires substantial collateral. On top of the cash received from the sale, a short
seller usually must pledge cash or cash-like securities to the lender in an amount
equal to 30 to 50% of the stock’s market value.25 Some managers use derivatives
(forwards, futures, options on indices and stocks) to execute a Short Selling
strategy. Others partially hedge the short sale with long positions or out-of-
the-money call options, which makes the strategy similar to a Long/Short Equity
strategy with a short bias.
The income on short positions has two components: profit from buying back
stock at a lower price than it was sold at and the short rebate interest. Similar to
Long/Short Equity strategies, the investment process requires much analytical
work. Many managers follow a bottom up approach, where an individual com-
pany’s financial standings and business prospects are examined (earnings, cash
flow, assets and liabilities, ratio analysis, products, costs etc.), followed by an
analysis of the industry sector (competitive environment, demand and supply,
profitability, foreign influences). A common theme of Short Selling strategies is
the identification of companies using aggressive accounting techniques. But a pure
fundamental approach to picking short candidates is much less appropriate than
the equivalent strategy for long strategies. Most short sellers look for an addi-
tional catalyst that will expose the poor fundamentals of the company, e.g. an
earnings revision.

.... 91 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 92

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Short Selling requires a good relationship with an established prime broker.

Potential problems related to borrowing stocks are share availability, stability of
the borrowing (since the broker can always call back the stock), the level of the
short rebate interest earned and protection against short squeezes (the temporary
sudden demand for the stock forcing short sellers to cover their positions at
unfavourable prices).26 With respect to short squeezes, the important issue is the
broker’s buy-in policy, which determines how quickly a short seller has to cover
his positions.

EXAMPLE
An example of a successful short trade in 1998 was Pediatrix Medical Group
Inc.,27 a provider of physician management services to hospital-based neona-
tal intensive care units. The valuation of the company was based on a
projected growth rate that far exceeded the expected birth rate of new
babies. The research performed by Short Selling managers showed further
aggressive accounting practices and inappropriate charges to insurance carri-
ers. The stock price between September 1998 and March 1999 is displayed
in Figure 3.8. Short Selling managers started shorting the stock in late
1998/early 1999. On February 12, the company announced it would miss
earning expectations, after the company’s auditor expressed the opinion that
the accounting for acquisition costs in 1998 was going to be rejected by the
SEC. The stock dropped more than 50% following the announcement.

Sources of return
The unlimited risk of loss to which Short Selling is exposed creates psychological
barriers for many investors. As a result, most investors never consider Short
Selling. Further, analysts are hesitant to give a detailed analysis of a potential
‘Short Sell Candidate’ and only a few give explicit ‘sell’ recommendations.
Additionally, many financial institutions and asset managers are prohibited by reg-
ulatory authorities or by internal compliance rules from Short Selling. All this can

.... 92 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 93

3 ■ AIS: Hedge funds and Managed Futures

lead to market inefficiencies and good opportunities for Short Selling managers.
When Short Selling is restricted, market prices might no longer be efficient.
Without the freedom to sell short, the pessimism of some investors will not be
fully represented in security prices and some stocks can become overpriced. Short
sellers also contribute to the efficient capital allocation process within an economy
and thereby take corresponding risks. They can be seen as earning market risk
premiums analogously as long equity investments. Comparable to Long/Short
Equity strategies, the key performance driver of Short Selling strategies is security
selection. The manager’s stock-picking and portfolio management skills remain
the most important sources of generated returns.

FIGURE 3.8 ■ Profitable Short Trade: Pediatrix Medical Group Inc., 1998 (Copyright 2002 Bloomberg
LP. Reprinted with permission. All rights reserved.)

.... 93 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 94

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Risk factors
Short Selling strategies are exposed to the same risk factors as long equity investment
strategies: general market risk, specific stock risk and corporate event risk (unex-
pected good news, takeovers, split-ups), execution risk (bad fills) and liquidity risk.
On top of these risks, Short Selling strategies are subject to specific risks related
to the fact that the securities have to be borrowed before they can be sold. The
extra steps of borrowing stock, maintaining the corresponding liability and then
buying back the shorted stock creates a set of unique risks which I call the borrow-
ing risks (or short risks). Certain stocks that a manager would like to short might
be unavailable, which limits the execution of good investment ideas (availability
risk). Second, short sellers can be subject to ‘short squeezes’ caused by the sudden
increase of demand for the stock. This can create a vicious cycle. One short seller
after the other has to cover his position, creating further upward pressure on the
stock. Finally, a short sale requires collateral. As the market moves against the
short position, additional collateral is required. The manager faces liquidity risk in
the form of insufficient funding. The degree of overall borrowing risk is greatly
influenced by the manager’s relationship with his prime broker.
Short Selling strategies face specific portfolio risks. The number of available
stocks for Short Selling is usually limited. This can lead to difficulties in reaching
sufficient portfolio diversification. A particular feature of Short Selling strategies
is the inverse relationship between performance and exposure. A loss in a short
position leads to a larger overall portfolio exposure to the losing position. There
is an unlimited risk of loss when a stock that has been shorted continues to rise.
Inversely, many short positions face the problem that the more successful a posi-
tion, the less invested the manager becomes.
The usual execution risks apply to Short Selling strategies, in particular bad fills
due to poor liquidity (slippage), fast markets and clumsy order execution by the
broker. On top of these general risks a short sale is subject to the ‘up-tick rule’ set
by market regulators. A short sale can only be executed on a plus or zero-plus
tick. Because of this rule some of the short sale ideas with great profit potential
might never get executed.

.... 94 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 95

3 ■ AIS: Hedge funds and Managed Futures

The short sale of stock generates cash which is usually held as collateral. Most
of the interest earned on this cash goes to the short seller and forms an income
stream for the strategy, but a portion is retained by the broker. When a stock is in
limited supply, the broker can retain a higher portion of the interest and the short
interest rebate to the short seller is reduced. Short rebate income is also reduced
by stock dividends that the short position has to pay. Failing to properly negotiate
short interest rebates or an unforeseen dividend increase creates a significant
income risk to the strategy (short rebate risk).
Inverse to Long Equity investing, Short Selling strategies perform the most
poorly in rising equity markets and the most favourably in bear markets. The
strong equity bull market of the 1990s caused this strategy to be consistently the
worst performing Hedge fund sector of that decade and extinguished most Short
Selling managers. The correction in the stock market that started in March 2000
revived the Short Selling industry and made it among the best performing sectors
during 2000/2001. Nevertheless, Short Sellers displayed rather disappointing per-
formance in 2001 (negative returns according to the CSFB/Tremont index) when
global equity markets lost broadly.

Managed Futures and Currency strategies

Managed Futures strategies seek to generate returns through long and short posi-
tions in various Futures contracts. This includes Futures on commodities (metals,
grains, etc.) and financial securities (fixed income, foreign exchange and equity
indices). The strategies are in most cases executed by ‘Commodity Trading
Advisors’ (CTA) and ‘Commodity Pool Operators’ (CPO) who have to be regis-
tered with the National Futures Association (NFA) and the Commodity Futures
Trading Commission (CFTC) in the USA or the Securities and Futures Authority
(SFA) in the UK. Within the set of Futures strategies, I distinguish between
Opportunistic (or Active) strategies on the one side and Passive strategies on the
other side (see Figures 2.1 and 3.1). Opportunistic Futures strategies are character-
ized by either discretionary trading based on long-term fundamentals and
short-term information or proprietary model-based techniques such as trend fol-

.... 95 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 96

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

lowing, countertrend trading, spread trading and a variety of others. Passive strate-
gies, in contrast, aim to systematically capture returns available from the presence
of professional hedging activities in Futures markets. The passive character
expresses itself through very low trading frequency and the lack of sophisticated
modelling. The system is rather simple (and in most cases publicly known, e.g. the
MLM or the sGFII index) and operates on a long time horizon, i.e. several months
to a year. Correspondingly, the fees charged are significantly lower.
I also include Currency Trading programs in the category of Managed Futures,
as their underlying strategy and trading approaches are often quite similar.
Currency strategies include technical trendfollowing, discretionary trading (based
on fundamentals), information based short-term trading, complex option arbi-
trage, interest rate carry, and cross rate arbitrage. Note that some Managed
Futures and Currency trading programs are quite similar to Global Macro strate-
gies in that they evaluate broad macroeconomic information and make according
‘bets’ on certain particular asset moves. Correspondingly, the distinction between
some Futures strategies and Global Macro is not always obvious.
Most CTAs trade Futures contracts, some trade options on Futures and a few
managers deal with OTC derivatives. Managers are usually broadly diversified by
trading in a wide class of Futures across multiple asset classes. Futures and cur-
rencies allow significant leverage, as usually only a fraction of the exposure has
to be provided by the manager for margin purposes. Margin to equity ratios vary
from 5% to about 25%.
Most Futures strategies are ‘long volatility’, i.e. they profit from rising volatil-
ity in financial markets. Being long volatility can be achieved in two different
ways: by taking a long position in options, i.e. a straddle; or by implementing a
momentum-based trading strategy, e.g. by following moving averages or breakout
indicators. The return of most Futures strategies corresponds to a risk premium
earned for providing the economic function of price risk transfer in markets with
commercial hedging activity (e.g. commodities). Further Opportunistic Futures
strategies’ returns are dependent on the managers’ skills to detect predictable pric-
ing patterns or market inefficiencies.

.... 96 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 97

3 ■ AIS: Hedge funds and Managed Futures

Futures Active – Systematic Technical

The strategy
Systematic Technical Futures strategies utilize computerized mathematical models
to generate buy and sell decisions. Trading is, to a large degree, systematic and
occurs mostly in highly liquid markets with low transaction costs. Models are
based on quantitative analysis of technical factors and indicators. The most typical
examples of this class are trendfollowing or countertrend models, which are based
on momentum indicators.28
Almost all systematic models have a time horizon or a combination of different
time horizons defined by the inherent time constants of their indicators. For long-
term models, momentum-based trendfollowing is typically most widely employed.
Systematic long-term models are usually rather simple; they use a variety of differ-
ent moving average (momentum) indicators, which help to detect extended price
trends. Most models have been optimized and back-tested for a number of years.
The majority of trades executed with this strategy are unprofitable. Nevertheless,
such strategies can generate positive returns because they close losing positions
quickly (‘cut losses’) and remain in profitable trades (‘let profits run’).
Short-term models use a larger variety of statistical tools including momentum
and countertrend trading techniques, breakout indicators and pattern recognition.
More exotic analysis is sometimes performed with the help of non-linear models,
neural networks, genetic algorithms and frequency models. The models try to
exploit statistically measurable short-term market inefficiencies.

EXAMPLES
Examples of short-term Systematic Technical trades are:

■ short-term trendfollowing and countertrend following with automatic

stop-loss limits
■ directional volatility positions executed with options
■ systematic equity or commodity index (e.g. GSCI) arbitrage
■ trading based on a trained (optimized) neural network or other applica-
tions of ‘artificial intelligence’ or non-linear or time series analysis.29

.... 97 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 98

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Sources of return
The distinction between long-term Systematic Active models and Systematic Passive
models can be blurred. Their sources of return are quite similar. As with most
Managed Futures strategies, parts of the returns of Systematic Active strategies are
based on the risk premium for providing liquidity to Futures and derivatives markets
and providing commercial hedgers with the opportunity to transfer their natural
price risk. But active models also have a skill component. The correct systematic
prediction of trends requires good model development abilities. Short-term model-
based trading aims at exploiting market inefficiencies and inhibited information
flow in financial markets. Returns are here a function of the manager skill in devel-
oping models that detect complex predictable short-term price patterns.
Most Systematic Futures models are of a trendfollowing type and many managers
and producers of these models claim that trendfollowing models generate inherent
returns as a result of mass psychology, i.e. crowd behaviour (a clear violation of the
‘weak’ form of the efficient market hypothesis). For most financial markets, however,
academic research provides no clear support in favour of these claims for trend per-
sistence in security prices.30 Likewise, there is a claim for currencies to exhibit serial
correlation, but again, this is not convincingly supported by empirical studies.31

Risk factors
Managers usually have clear exposure to certain markets. Therefore, the predomi-
nant risk of systematic Futures and Currency strategies is market risk, i.e. the risk
of the market moving against the established position. Trendfollowing strategies in
particular are exposed to the risk of steady losses over an extended period of time
in non-trending, directionless market environments, which exhibit numerous price
reversals (‘whip-saw markets’). This is illustrated in Figure 3.9 (left-hand side).
The structure of financial markets changes continuously and models built in the
past might not be suited for markets in the future. Systematic models thus bear the risk
of structural changes in the underlying markets. It often occurs that models that have
worked very successfully for a period of time start showing persistent performance
problems. Systematic long-term trendfollowing models had excellent performance
during the 1980s and early 1990s when financial markets showed strong directional

.... 98 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 99

3 ■ AIS: Hedge funds and Managed Futures

trends. Fewer trends were observed in the later 1990s, and the models accordingly
showed much lower returns. In 2000/2001 performance was again good for many
Futures strategies. But it has to be noted that the variability of returns between
managers increased during 2000–2001, i.e. managers diverged in performance.
Systematic trading strategies rely solely on the functioning of models that were
developed as a result of research efforts performed during some past time period.
A very important risk factor is model risk, which refers to the risk of the model
being flawed or unsuited for the current market structure. Most models are opti-
mized on past data, i.e. the parameters of the model are determined such that the
model shows the best performance for particular periods in the past. This is a dan-
gerous undertaking and bears the risk of over-fitting, also referred to as ‘curve
fitting’. Provided with a sufficient number of free parameters, every model can be
tuned such that it shows excellent performance on any (even random) data set!
However, when used with data that the model was not optimized for (‘out of
sample test’), performance may be significantly worse, sometimes rendering the
model useless. The more complicated the model and the more parameters
involved, the higher the risk of over-fitting. Simple trendfollowing models bear less
risk of over-fitting than complicated ‘non-linear modelling approaches’ that have a
lot of different parameters. The investor should always inquire about the optimiza-
tion procedure that was undertaken during the development of the model and
about the performance on out-of-sample data or during real-time trading.
Systematic Futures and Currency trading bears operational risk. Most models
are run on computers and their failure or programming errors (‘bugs’), electricity
breakdown etc. can lead to trading losses. Furthermore, the trading of a large
number of different positions bears execution risk.
Because approaches are very diverse, it is difficult to determine general market
conditions in which all the different models perform consistently well or particu-
larly poorly. Strategies which exploit short-term market inefficiencies perform
best in less developed and therefore less efficient markets. Most strategies perform
poorly in directionless ‘whip saw’ market conditions. Long-term trendfollowing
models in particular need persistent trends for good performance.

.... 99 ....
8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 100

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Futures Active – Discretionary

The strategy
Active Discretionary Futures and Currency strategies are characterized by proprietary
approaches primarily employing fundamental analysis, sometimes in combination
with technical analysis. The most typical examples are directional long-term positions
based on fundamental forecasts or short-term trading based on information flow.
Trading decisions are largely based on the study of external factors that affect the
supply and demand of a particular security or asset.
Most of the Active Discretionary managers have extensive experience with com-
modity or currency markets. Having typically worked for years in these markets or
related environments, e.g. in an oil company, trading corn, trading inter-bank FX
markets, the manager’s competitive advantage is his knowledge of the essential funda-
mentals (e.g. worldwide inventory level of a commodity) and his ability to anticipate
changes in supply and demand relationships that affect security prices. By monitoring
relevant supply and demand factors, a state of dis-equilibrium may be identified that
has yet to be reflected in the price of the underlying commodity. Such factors include
weather, the economics of a particular commodity, government policies, domestic and
foreign political and economic events and changing trade patterns. Most managers
have fast access to relevant information which they are able to capitalize on.
Examples of Active Discretionary trades are:

■ Directional long-term position based on fundamental analysis, e.g. one

currency against another, in a particular commodity, in one country or firm’s
debt, or in a particular equity market sector (or country).
■ Short-term positions based on particular information flow, e.g. corporate
currency-hedging activity, a country’s central bank policy or the inventory
management of particularly influential global corporates.
■ Crossrate arbitrage, interest rate carry arbitrage, investing in high-yield currencies.
■ Inter-product spread trade between Futures of related commodities, currencies
or fixed income products, e.g. soybean versus soybean oil, Euro versus Swiss
franc, German Bund versus British gilts.

.... 100 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 101

3 ■ AIS: Hedge funds and Managed Futures

■ Directional volatility positions and other option strategies such as time

spreads, i.e. Futures with different maturity times.
■ Equity sector arbitrage, i.e. long undervalued sector Futures and short
overvalued sector Futures.
■ Convexity trades, e.g. long Eurodollar Futures and short Eurodollar forward.

EXAMPLE
An example of an actual trade for an Active Discretionary Futures strategy
is the spread trade (long soybean meal and short soybean) grain traders
established following the increasing fear of the role of fodder meal and
BSE in Europe during 2000.

Sources of return
Similar to Global Macro Hedge fund strategies, Active Discretionary Futures strate-
gies exploit temporary mispricings through superior and faster access to information.
The specific skills and experience possessed by Active Discretionary managers in cer-
tain markets are the most important source of returns generated. For example, they
might have knowledge about the inventory levels of oil or corn producers and can
interpret this information accordingly. Their activities often generate the necessary
liquidity for commercial market participants to carry out their business in an orderly
fashion. This activity is compensated with a corresponding risk (liquidity) premium.
Similarly, the returns of FX strategies can be interpreted as a premium paid by
commercial market players for the generation of liquidity and price continuity. In
anticipation of the deal flow created by a large transaction of a commercial
market participant (that finds no direct commercial counter-party), or the trend
induced by an action of a major market player, the speculators adjust the FX rates
to create the necessary liquidity. The following serves as an idealized example. A
large cooperation has the need to exchange a large sum of money from one cur-
rency into another (e.g. Germany-based Siemens wishes to exchange USD 1
billion into EUR). The market anticipates this deal flow and the price shifts
accordingly in advance of the transaction, which leads to the company paying a

.... 101 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 102

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

higher price for the desired currency in exchange for the requested liquidity.
Further, when major shifts in the supply–demand balance occur due to a macro-
economic event (e.g. Mexico crisis 1995), FX traders continue to provide
liquidity and price continuity in a fundamentally overvalued/undervalued market.
Currency ‘carry trades’ (investing in high-yield currencies where a return is gener-
ated by the interest rate differential) earn a risk premium by assuming exposure to
possible currency devaluations of a high-yield currency.

Risk factors
As for most Managed Futures strategies, market risk is the dominant risk con-
nected with Active Discretionary Futures trading, i.e. the risk of the market
moving against the established positions. Fundamental directional positions are
specifically exposed to event risk, the risk of a particular external event influenc-
ing the position unfavourably (e.g. unexpected cut in interest rates, change of
government or unfavourable weather conditions).
Most Active Discretionary Futures and Currencies programs rely on the skill
and experience of one or a few key people. On departure or unavailability (e.g.
disease or death) of a key person, the trading program might suffer significantly
worse performance (key people risk). Furthermore, on changing market conditions
key people can lose their trading edge. Finally, short-term trading in a high
number of different positions bears execution risk (e.g. bad fills, slippage).
Most managers perform well in markets that are not entirely information
efficient. By way of contrast, Active Discretionary strategies are less successful in
very efficient markets.

Futures Passive
The strategy
Equity markets offer investors who are willing to provide capital to companies the
possibility of earning a premium (relative to the risk free rate) on their invest-
ment. Futures markets also offer inherent returns to investors, but for a different
reason. The basic economic function of Futures markets is to act as a risk transfer

.... 102 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 103

3 ■ AIS: Hedge funds and Managed Futures

vehicle. Participants in Futures markets are generally one of two different kinds:
hedgers and investors (neglecting the role of arbitrageurs). Hedgers are these com-
mercial producers or consumers of particular commodities who want to transfer
their price risk by transferring it to another entity. Investors in Futures markets
provide these commercial hedgers with the possibility to transfer their unwanted
price risk. Passive Futures strategies follow the hedging patterns of commercial
‘natural longs and shorts’ (producers and consumers)32 in Futures markets and
systematically take the opposite position.33
Demand–supply mismatches in the market between ‘natural long hedgers’ and
‘natural short hedgers’ occur frequently. A demand to acquire price stability (or
‘price insurance’) for either the long or short hedger is almost never completely met
by the other side. This gives investors an opportunity to provide the marginal price
stability and be compensated accordingly. Investors thus provide ‘price insurance’ to
hedgers. Comparable to an insurance company, which earns a premium for insuring
its clients or a bank that lends long term and borrows short term (and thus earns a
liquidity and a duration risk premium), the investor should expect a commensurate
return. The more extended the supply demand mismatch, the higher the return.
Commercial hedgers naturally follow contrarian strategies, i.e. they are acting
against trends. A price increase (or more specifically, a price move out of a certain
recent price range) will lead to more natural long hedgers wanting to hedge com-
pared to the number of natural shorts, because they want to lock in a relatively
higher price. Inversely, a decrease in prices will motivate natural shorts to
hedge/lock in attractive prices. The resulting demand–supply mismatch is a func-
tion of the extent of the price move. Investors can take the opposite position of
hedgers by investing with the price momentum (i.e. trend). Most Systematic
Passive Futures strategies are therefore constructed on the basis of long-term
trendfollowing systems. Positions are taken according to a momentum indicator
on underlying prices. Another approach is taking positions according to the
degree to which the asset is trading in contango or backwardation, which gives an
alternative indication of the net direction of commercial hedging activities.

.... 103 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 104

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

EXAMPLES
Two examples of typical trades of a Passive Futures strategy in crude oil
Futures are:

■ The price for crude oil has risen substantially and is currently relatively
high. British Petroleum (BP) and other oil producers want to lock in
the high price level and sell short crude oil Futures contracts. As a
result a demand–supply imbalance occurs in the Futures market for
crude oil. The Passive Futures strategy takes the opposite position by
going long.

■ The price for crude oil has fallen substantially and is currently
relatively low. American Airlines and other airlines are afraid of a
potential price increase in oil. They buy energy Futures in the market
to protect the airline from possible higher prices, which leads to a
general demand–supply imbalance in the crude oil Futures market. The
Passive Futures strategy takes the opposite position and builds up a
short exposure to energy prices in the Futures markets.

The determinants of commodities prices vary among each other, so the correlation
between different commodities is usually low. This enables Passive Futures managers to
reach efficient diversification through investing in a broad basket of Futures contracts.

Sources of return
Passive Futures strategies take systematically the opposite position to commercial
hedgers and by accepting short-term price volatility they provide necessary liquid-
ity for the Futures markets. In return, the investors receive a premium similar to
an insurance company being paid a premium in return for accepting event risk.
Thus, the investor generates return by accepting the excessive price risk commer-
cial hedgers are naturally exposed to but unwilling to take.34 The risk premium
earned by Passive Futures strategies can also be referred to as ‘commodity hedging
demand premium’.

.... 104 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 105

3 ■ AIS: Hedge funds and Managed Futures

The economic function of risk transfer as provided by investors in Futures markets

is entirely different from the function of traditional investment markets (i.e. equity and
bond market). The latter is the formation and efficient distribution/allocation of capital
to companies and governments. The returns generated by Passive Futures strategies are
therefore inherently independent from traditional equity and bond returns.

Risk factors
The predominant risk for Passive Futures strategies is market risk, i.e. the risk of
the market moving against the established position. As the strategy takes the
opposite position of commercial hedgers providing them with an ‘insurance’
against price adverse moves, it will suffer losses whenever there is a move against
the hedgers’ natural position.
The strategy sector relies strongly on diversification, similar to insurance com-
panies, which collect premiums over many insurance takers before paying out
claims. An important risk factor is diversification risk, i.e. the lack of sufficiently
broad diversification over different Futures markets. But there is a trade-off
between the need for diversification and the available liquidity for Passive Futures
strategies. Investing in too many different markets also leads to liquidity risk, as
the additional Futures contracts may lack sufficient trading liquidity.
Passive Futures strategies bear model risk, which is the risk that the model is
unsuited to carry out the function of capturing inherent returns in Futures mar-
kets. The execution risk of the strategy is comparably low. The time horizon of
commercial hedgers is in the order of weeks to months, so most Passive Futures
strategies have a rather low turnover.
Passive Futures strategies face problems in non-trending, directionless market envi-
ronments exhibiting numerous price reversals, where trading occurs in a limited price
range over an extended period of time (‘whip-saw markets’). Contrariwise, persistently
trending markets constitute a favourable environment. This is illustrated in Figure 3.9.
The strategy sector shows strong performance in volatile and declining equity markets,
especially in periods of an equity market crisis (crash), when investors’ and hedgers’
risk aversion increases and thus risk premiums are high. It therefore offers an excellent
diversification to a Long Only Equity portfolio or equity-directional Hedge fund
strategies, such as Equity Market Timing and Long/Short Equity.

.... 105 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 106

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Unfriendly environment (silver) Friendly environment (crude oil)

FIGURE 3.9 ■ Profitable and unprofitable environments for Passive Futures strategies (Copyright
2002 Bloomberg LP. Reprinted with permission. All rights reserved.)

Notes
1. Classification schemes with different nuances are presented in ‘In Search for
Alpha’ by A. Ineichen, UBS Warburg (2000) and in ‘Alternative Asset Returns:
Theoretical Bases and Empirical Evidence’ by T. Schneeweiss and J. Pescatore,
Handbook of Alternative Investment Strategies, Institutional Investors Inc., New
York, 1999.
2. It is disputed, whether Long/Short Equity strategies should be classified as
opportunistic. There is a great amount of heterogeneity between individual managers
in this strategy sector. But characteristically, most managers have a long bias in their
strategy, i.e. they maintain a net long exposure. I follow the reasoning of A. Ineichen
(2000) here. A different possibility is to introduce a new category on the first level
and distinguish the different styles of Long/Short managers by different degree of net
exposure and leverage (as done by T. Schneeweiss and J. Pescatore, 1999).
3. A description of the different components of a convertible security can be found
in ‘Market Neutral and Hedged Strategies’, published by Hedge Fund Research, Inc.
The authors also provide a discussion of Convertible Arbitrage Strategies.

.... 106 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 107

3 ■ AIS: Hedge funds and Managed Futures

4. For more details about the pricing and characteristics of convertible bonds the
reader is referred to Pricing Convertible Bonds by K. Connolly.
5. See the article ‘Convertible Arb Funds Turn to Default Swaps’ by C. Schenk in
Risk Magazine, July 2001. For credit derivatives generally, see the book by
J. Tavakoli, Credit Derivatives: A Guide to Instruments and Applications, 2nd edn.
6. See F. Fabozzi in Fixed Income Analysis (Chapter II, 2) for more details about
valuing bonds with embedded options.
7. A further discussion of the numerous liquidity issues connected to the
Convertible Arbitrage strategy can be found in ‘Convertible Arbitrage’ by M. Boyd
et al. in Managing Hedge Fund Risk by V. Parker (ed.), Risk Books, 2001.
8. These are often based on quadratic optimization procedures. Quadratic
optimization corresponds to the Markovitz construction of efficient portfolios. For
more algorithmic details, see H. Markowitz, Portfolio Selection – Efficient
Diversification of Investment, John Wiley & Sons, New York, 1959.
9. See ‘Persuasive Evidence on Market Inefficiencies’ by B. Rosenburg et al. in
Journal of Portfolio Management, Spring 1985, 11, no. 3, p.9.
10. There has been a broad discussion in recent years on extending the standard
theory of finance, which is built on the ‘no arbitrage’ principles, portfolio theory and
capital asset pricing models, by what is called ‘behavioural finance’. Behavioural
finance aims at accounting for market anomalies by considering psychological models
of human behaviour. See H. Shefren in Beyond Greed and Fear: Understanding
Behavioural Finance and the articles by M. Stratman, ‘Behavioural Finance: Past
Battles and Future Engagements’; and R. Shiller, ‘Human Behavior and the Efficiency
of the Financial System’, in Handbook of Macroeconomics.
11. See E. Fama, K. French, ‘The Cross Section of Expected Stock Returns’.
12. A thorough description of the different aspects of the Risk Arbitrage strategy
is provided in Risk Arbitrage: An Investor’s Guide, by K. M. Moore, John Wiley &
Sons, New York, 1999.
13. The pooling method is no longer accepted under US GAAP for mergers announced
after July 1, 2001. In other legislations (e.g. the UK), pooling is still possible, although
International Accounting Standards (IAS) provide for pooling only in cases where the
acquirer cannot be determined (i.e. the transaction is really a merger of equals).

.... 107 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 108

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

14. In Deals, Deals, and More Deals, by R. Pitaro, there are many examples of
historical merger transactions. The book also offers an in-depth discussion of Risk
Arbitrage strategies.
15. These are further elaborated in ‘The Case for Hedge Funds’, Tremont,
Partners Inc. and TASS Investment Research Inc., 1999.
16. In ‘Measuring the Market Impact of Hedge Funds’, W. Fung and D. Hsieh
examine the role of Hedge funds during market turbulences of the 1990s.
17. M. Strome in Evaluating and Implementing Hedge Fund Strategies, by
R. Lake, Euromoney Books, 1996.
18. Credit risk comes mostly in the form of sovereign country risk. For example,
the Russian default in the summer of 1998 hurt many Global Macro managers.
19. Ironically, the markets turned strongly in Robertson’s favour almost at the
same time he made the announcement proving correct his prediction that ‘a Ponzi
pyramid is destined for collapse’.
20. Some categorize Long/Short Equity completely as a Relative Value strategy, e.g.
J. Bernard and T. Schneeweiss in ‘Alternative Investments: Past, Present and Future’, in
The Capital Guide to Alternative Investments, ISI publications, 2001. Here, I want to
distinguish between market neutral strategies and strategies with a (mostly long) bias.
21. I am trying to resolve the question of whether to classify Long/Short as
Opportunistic or Market Neutral by distinguishing Long/Short Equity with net
market exposure from Long/Short Equity Market Neutral (the latter is discussed as
a ‘Relative Value’ strategy).
22. An interesting study concerning the price action lag of small and mid-cap
stocks is presented in The Wildcard Option in Transacting Mutual-Fund Shares by
J. Chalmers, R. Edelen and G. Kadlec, Wharton School of Finance, 2000.
23. Note that this statement refers to sectors, not to broad markets. Whether
statistical properties like trend persistence are present in equity markets (a clear
violation of the weak form of market efficiency) is still subject to much discussion.
A textbook summary of the discussion on efficient markets with numerous
references can be found in Chapter 7 of Investment Analysis and Portfolio
Management by F. Reilly and K. Brown. A more recent discussion with further
references can be found in Dan Sullivan (ed.) The Chartist Mutual Fund Letter,

.... 108 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 109

3 ■ AIS: Hedge funds and Managed Futures

(PO Box 758, Seal Beach, CA 90740). Among the references given here is the
seminal paper by Brock, Lakonishok and LeBaron, Journal of Finance, December
5, 1992, where the authors found that techniques championed by market timers
appeared to provide useful market ‘buy and sell’ signals. Their study utilized the
Dow Jones Index from 1897 to 1986 using two methods: simple moving averages
and trading ranges. The moving average method produced the best performance;
the trading range break also produced positive returns, which is in sharp contrast
to the ‘random walk’ hypothesis articulated by protagonists of the (weak) form of
market efficiency.
Le Baron presents another interesting statistical effect of trend forecasting
conditional on volatility in his paper ‘Some Relation between Volatility and Serial
Correlations in Stock Market Returns’, Journal of Business, 1992, 65(2), p.199.
24. As discussed in The Wildcard Option in Transacting Mutual-Fund Shares by
J. Chalmers et al.
25. A discussion about Short Selling in a portfolio context can be found in
‘Efficient Sets, Short Selling, and Estimation Risk’, by G. Alexander, Journal of
Portfolio Management, Winter 1995.
26. A good discussion about the issues involved is given in ‘Short Selling: A
Unique Set of Risks’ by A. Arulpragasam and J. Chanos in Managing Hedge Fund
Risk by V. Parker (ed.).
27. This example is also presented by A. Ineichen; ‘In Search of Alpha’.
28. See Technical Traders Guide to Computer Analysis of the Futures Markets by C. Le
Beau and D. Lucas for more details about technical trading models in financial markets.
29. I generally believe that the methods developed for non-linear time series
analysis (a discipline that originated in applied mathematics and theoretical
physics) fall short of being applicable to financial modelling.
30. See Chapter 7 of Investment Analysis and Portfolio Management by F. Reilly
and K. Brown and references therein.
31. One way of reasoning as why currency markets should trend is that central
banks attempt to smooth foreign exchange rate movements through intervention,
see e.g. R. Arnott and T. Pham, Tactical Currency Allocation.

.... 109 ....

8409 Chapter 3 p38-110 11/4/02 1:13 PM Page 110

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

32. Examples are: 1. Oil: Natural shorts are airline, natural longs are oil producers.
2. Foreign exchange: Exporting companies are natural longs of foreign currency,
companies importing raw material from abroad are natural shorts of foreign
currency. 3. Interest rates: Firms with a high need of financing (e.g. construction and
development companies) are natural shorts of interest rate-linked instruments
(bonds), firms with a large amount of liquid capital (e.g. banks) are natural longs.
33. A detailed description of this type of strategy is presented in L. Jaeger,
M. Jacquemai and P. Cittadini, ‘The saisGroup Futures Index (sGFI) – A New Passive
Futures Investment Strategy’, saisGroup Research Paper, http://www.saisgroup.com,
Journal of Alternative Investments, accepted for publication.
34. J. M. Keynes originally articulated the theoretical argument for a risk transfer
premium in the Futures market. In addition to specifying the theory, however, he
also declared that he expected the premium to manifest itself as a consistent
downward price bias in the Futures relative to spot prices. He termed this
postulated price bias ‘normal backwardation’. See J. M. Keynes, A Treatise on
Money, Volume II, Macmillan & Co., New York, 1930.

.... 110 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 111

CHAPTER 4

Empirical Properties of
Alternative Investment Strategies

Attractive risk–reward characteristics and low correlations to traditional assets are

features that make investments in Hedge funds and Managed Futures strategies par-
ticularly attractive to investors.1 This chapter provides an overview of empirical risk
and return properties of AIS sectors as measured by certain benchmark indices2 and
illustrates the benefits of AIS in a traditional portfolio. Properties of most interest to
investors are return, volatility (most often measured in standard deviation of
returns), drawdowns, correlation attributes and, as a measure for risk-adjusted per-
formance, the Sharpe ratio. The latter is defined as the ratio between annualized
monthly returns less the benchmark return (usually the long-term average of the risk
free interest rate, here assumed to be 5%) and annualized monthly volatility.3
An interesting additional risk measure included in the following discussion is the
kurtosis (fourth moment of the return distribution), which measures the ‘tail fatness’
of a historical return distribution as compared to the normal distribution. A value
larger than zero indicates fatter tails than the normal distribution, i.e. a higher prob-
ability of extreme moves.4 One of the perceived advantages of AIS is that they
generally show much lower volatilities than equity investments when measured by
the standard deviation (second moment). But many AIS return distributions show
negative outliers, which exceed the range of three and more standard deviations
more frequently than stocks or bonds. This risk (‘tail risk’) is of major concern to

.... 111 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 112

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

investors and remains unrecorded if the standard deviation is used for risk quantifi-
cation only. It is therefore useful to include the kurtosis in the analysis of AIS
performance. A value above 5 to 10 indicates significant exposure to tail risk.

Challenges of AIS performance measurement

Before we engage in the detailed analysis of the risk–return properties of alterna-
tive investments, some words of caution are necessary. First, the variability of
manager performance within strategy sectors is significantly larger than for tradi-
tional equity and fixed income investments, which makes benchmark analysis less
representative. Second, unlike traditional asset classes (bonds and equities), where
performance data and benchmarks are readily and reliably available, the infra-
structure and reliability of performance data for AIS are still undeveloped. 5
Several data providers have recently began to emerge, but the data availability and
quality does not compare to those of traditional asset classes. For a list of data
providers, the reader is referred to the appendix of this chapter.6 We note at least
three important problems specific to measuring AIS performance in the form of
an index (or sub-index for particular sectors):7

1 Survivorship bias. Survivorship bias is the result of unsuccessful managers

leaving the industry (and the representative index) and the successful ones
remaining. This leads to a bias towards successful managers. The question is
whether the database represents the true returns of the strategy sector or the
returns of those managers who are successful enough to stay in business.8
2 Weighting within the indices. Some indices are asset weighted so that funds with
more assets under management are given a larger weighting than smaller funds.
Other indices are equally weighted giving smaller managers a relatively strong
weight. Asset weighting can lead to a few large managers dominating the index
(this is especially the case for Short Selling and Global Macro strategies), while
equal weighting might be less representative for the average investor. The specific
nature of the index should be considered when selecting and comparing indices.

.... 112 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 113

4 ■ Empirical Properties of Alternative Investment Strategies

3 Selection bias. The selection of the managers and sub-strategies included in the
index has a significant influence on the index characteristics. To a lesser
degree this also applies to traditional indices (equities and bonds) but the
larger degree of peer group performance variability leads to more significant
differences in index performances depending on how the particular strategy
sector is defined and which managers are included in the index. Furthermore,
some managers refuse to be included in any database.

In this chapter, I refer to four different index families: the Credit Suisse First Boston
(CSFB)/Tremont indices, the HFR (Hedge Fund Research) indices, the Zurich (for-
merly MAR) indices for Managed Futures performance and the Parker FX Index for
Currency Strategies (see appendix to this chapter for further details). The HFR Index
is equally weighted among managers and adjusted for survivorship bias since 1995,
while the CSFB Tremont indices and Zurich (MAR) Managed Futures indices are asset
weighted and adjusted for survivorship bias since inception. The HFR Index goes back
to 01/90 (for some strategies the HFR indices go back to 01/87, but I will only con-
sider the period from 01/90 here) and the Tremont indices started in 01/94. The
Zurich (MAR) indices partly go back to the beginning of the 1980s and the Parker
indices to 1986, but here also, I will consider the period starting from 01/90 only.

Risk and return

Table 4.1 presents the results of a risk–return analysis for AIS strategy sectors from
January 1990 to January 2002 based on monthly data. Returns were calculated as
geometric averages (cumulative returns) of the log-differences of consecutive
prices. The risk calculation is based on the standard deviation of returns. One can
observe that the risk–return properties of Hedge fund strategies are very heteroge-
neous. Additionally, in Figures 4.1 and 4.2 these performance numbers are displayed
graphically in a risk–return plot. I intentionally did not choose to plot the perform-
ance properties of individual strategies as risk–return coordinate points, but instead
show representative areas of return and risk (namely one standard deviation along
the main two axes of each performance ‘cluster’) to express the variability of man-
ager performance within each sector.

.... 113 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 114

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

TABLE 4.1 ■ Risk and return of AIS

Return Volatility Max. Drawdown Sharpe Ratio Kurtosis

Distressed Securities
HFR 15.10% 6.38% –10.78% 1.58 6.53
Long/Short Equity
HFR 20.22% 9.21% –9.29% 1.65 1.22
Tremont 13.15% 11.88% –14.21% 0.69 2.69
Equity Market Neutral
HFR 11.11% 3.24% –2.72% 1.88 0.35
Tremont 10.86% 3.17% –3.55% 1.85 0.05
Event Driven
HFR 15.90% 6.67% –10.78% 1.63 6.75
Tremont 11.87% 6.42% –16.04% 1.07 28.30
Macro
HFR 17.50% 8.82% –10.70% 1.42 0.25
Tremont 14.26% 13.28% –26.78% 0.70 1.33
Market Timing
HFR 14.49% 6.92% –5.50% 1.37 –0.61
Regulation D
HFR* 23.02% 7.18% –11.99% 2.51 2.04
Relative Value
HFR 13.58% 4.14% –6.55% 2.07 8.44
Convertible Arbitrage (HFR) 11.90% 3.39% –4.84% 2.03 3.61
Convertible Arbitrage (Tremont) 11.06% 4.82% –12.04% 1.26 5.20
Fixed Income Arbitrage (HFR) 8.59% 4.79% –14.42% 0.75 9.09
Fixed Income Arbitrage (Tremont) 7.57% 4.76% –12.47% 0.54 15.22
Short Selling
HFR 1.43% 22.73% –53.36% –0.16 1.44
Tremont –1.04% 18.27% –43.64% –0.33 1.35
Fund of Funds
HFR 10.97% 6.05% –13.08% 0.99 4.16
Composite
HFR 15.83% 7.27% –11.42% 1.49 3.44
Tremont 11.80% 9.21% –13.81% 0.74 1.15

.... 114 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 115

4 ■ Empirical Properties of Alternative Investment Strategies

TABLE 4.1 (Continued)

Return Volatility Max. Drawdown Sharpe Ratio Kurtosis

Managed Futures
Tremont 4.69% 11.59% –17.74% –0.03 1.34
MAR Systematic Active** 6.62% 11.05% –15.64% 0.15 0.78
MAR Discretionary Active 12.47% 6.87% –5.60% 1.09 1.13
MAR Trendfollowing 10.29% 16.14% –20.11% 0.33 1.39
Passive (sGFII) 13.16% 10.02% –9.09% 0.81 1.19
Currency Trading
Parker Index 11.11% 9.73% –8.86% 0.63 3.90
Equity
MSCI EU (incl. dividends) 8.79% 15.33% –33.03% 0.25 0.60
MSCI World (incl. dividends) 6.79% 14.65% –35.32% 0.12 0.63
S&P 500 (incl. dividends) 12.03% 14.52% –31.41% 0.48 1.13
Bonds
SSB World (10) Gov’t Bonds 6.60% 6.13% –5.92% 0.26 0.40
Commodities
Goldmann Sachs Index 3.10% 17.96% –48.25% –0.11 1.91

Source: Partners Group

Data: HFR (01.90–10.02), Tremont (01.94–01.02), MAR (01.90–01.02),
*01.96–01.02; **01.92–01.02

Focusing on Global Macro and Long/Short Equity strategies, one can clearly
observe that: (1) high returns came at the expense of high risk; and (2) there are
significant differences between the HFR and the Tremont benchmarks. For the
12-year study period, HFR show an average annual return for Global Macro
funds of 17.5% per annum. Along with these impressive returns came a relatively
high level of risk. With 8.9% standard deviation, Global Macro investments
ranked among the more volatile in the Hedge funds industry. The strategy has a
Sharpe ratio of 1.42 according to HFR. In contrast, the eight-year CSFB Tremont
Index for Global Macro shows an average annual return of 14.3%, a volatility of
13.3% and a Sharpe ratio of 0.7. Long/Short Equity strategies showed 20.2%
annualized returns, 9.2% volatility and a Sharpe ratio of 1.65 (Tremont: 13.2%
returns, 11.9% volatility, 0.69 Sharpe ratio). The kurtosis numbers for Long/Short
Equity strategies are slightly higher than those for Global Macro. One reason for

.... 115 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 116

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

40%
35%
30%
25%
Return

20%
15%
10%
5%

2% 4% 6% 8% 10% 12% 14% 16%

Global Equity Equity Relative Distress Event Convert Short

S&P500
Macro Hedge Mkt. Tim. Value Security Driven Deb. Arb Seller

FIGURE 4.1 ■ Risk and Return for Hedge fund strategies (01.90–01.02)
Source: Partners Group
Data: HFR, MAR

the deviation between the two indices for Macro and Long/Short Equity strategies
is that they had strong returns during the years 1990–94, a time period not
entirely reflected by the Tremont Index, which began coverage in 1994.
It is important to note the large deviation in the drawdown figures between the
HFR and Tremont databases for Global Macro, Long/Short Equity, Convertible
Arbitrage and Event Driven strategies, for which the different time periods cannot
be an explanation. While the ten-year equal-weighted HFR Macro Index shows the
most severe drawdown during the period to be –10.7%, the asset-weighted Tremont
Index (covering a shorter time period) shows a substantially worse –26.8%. The
deviations for Long/Short Equity and Event Driven strategies are less remarkable,
but also noteworthy (–9.3% HFR and –14.2% Tremont, –10.8% HFR and –16.0%
Tremont respectively). For Convertible Arbitrage, HFR displays the worst draw-

.... 116 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 117

4 ■ Empirical Properties of Alternative Investment Strategies

40%
35%
30%
25%
Return

20%
15%
10%
5%

2% 4% 6% 8% 10% 12% 14% 16%

System. Discret. System.

Passive Active Active S&P500

FIGURE 4.2 ■ Risk and Return for Managed Futures strategies (01.90–01.02)
Source: Partners Group
Data: HFR, MAR

down of –4.9% in comparison to a value of –12.0% for the CSFB/Tremont Index.

The wide differences between the two databases for this very important statistic fur-
ther illustrate the value of understanding the way data is collected and what the
information truly represents. My assessment is that the Tremont numbers in most
circumstances provide a better understanding of past performance characteristics
than the HFR indices. The Tremont data is asset weighted which provides better
insights into the performance for the average investor.
Providing slightly lower returns on average, but better stability on the risk side
are Market Timing, Relative Value, Distressed Securities and Event Driven strate-
gies. While Market Timing and Relative Value experienced single worst
drawdowns of –5.5% and –6.6% (HFR data) respectively, Event Driven and
Distressed Securities exhibited drawdowns of about double these amounts.
Looking further at kurtosis provides another view on risk: Market Timing (HFR):

.... 117 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 118

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

–0.6, Relative Value (HFR): 8.4, Distressed Securities (HFR): 6.5, Event Driven:
6.75 (HFR) and 28.3 (CSFB/Tremont). The large kurtosis for Fixed Income
Arbitrage (HFR: 9.1, CSFB/Tremont: 15.2), and for Relative Value in general, was
largely caused by the events in the fall of 1998.
The worst performing sector during the period is clearly Short Selling.
Suffering during a decade of bullish equity markets, short sellers returned an aver-
age of 1.4% with a volatility of 22.7% (HFR). The worst drawdown figure of
–53.4% sheds further light on the difficulties faced by this sector during the time
period considered here. Similar results are reflected by the Tremont Index. It is
not surprising, however, that in the 18-month period from April 2000 to
September 2001, short sellers were among the best performers in the industry
(but they lost significantly again between 10/01 and 01/02).
Equity Market Neutral strategies demonstrated impressive risk-adjusted
performance according to both the HFR and Tremont databases. With returns of
11.1% and low volatility of 3.2%, the strategies show a very impressive Sharpe
ratio of 1.88 and 1.85 (HFR and Tremont respectively). This strategy has the
lowest drawdown overall (–3.55% for Tremont and –2.72% for HFR). The
kurtosis for the strategies is zero, indicating low tail risk.
Convertible Debenture Arbitrage (‘Regulation D’) strategies showed average
annual returns of 23.0% and volatility of 7.2%, which gives a Sharpe ratio of
2.51. The worst drawdown stands at –12.0%. An important factor to consider
here is the lack of liquidity. The presented volatility is partly the effect of the
absence of mark to market pricing. With frequent changes in pricing absent,
volatility appears artificially low. The high returns partly represent a liquidity pre-
mium the investor receives for accepting a relatively long lock-up period.
The Managed Futures side of the AIS Universe is more homogenous than Hedge
funds. The standalone return to risk ratios of Managed Futures strategies are gener-
ally not as impressive when compared to Hedge fund strategies. Their main value
lies in their correlation attributes and diversification capacities within an overall
portfolio. Interesting because of their attractive return and risk properties (and their
usually lower fee structure) are Systematic Passive Futures strategies (as represented
by the sGFI)9 with 13.2% return and 10.0% volatility (Sharpe ratio: 0.81). The
maximal drawdown of this strategy is –9.1%. Active Discretionary Futures strategies

.... 118 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 119

4 ■ Empirical Properties of Alternative Investment Strategies

show a 12.5% return, 6.9% volatility and a –5.6% maximal drawdown. Systematic
Active strategies demonstrate 6.6% returns and 11.1% standard deviation, with a
–15.6% maximal drawdown. The pure trendfollowing strategies show a return of
10.3%, a volatility of 16.4% and a maximal drawdown of –20.1%. For complete-
ness, the Commodity Long Only strategy, as represented by the Goldmann Sachs
Commodity Index, is also displayed. This Long Only approach does not take advan-
tage of large swings on the downside and therefore demonstrates only 3.1% returns
and a very large 18.0% standard deviation. The worst drawdown of –48.35% is a
staggering number that further illustrates the benefits of having the flexibility to go
short in the Futures market.

Comparison with equities and bonds

The returns of AIS vary by strategy, but as a whole and individually compare very
favourably to the returns of traditional investments. In the equity markets, repre-
sented here by the S&P500, the MSCI Europe and the MSCI World, the return to
risk ratios are generally lower than their AIS counterparts. The decade of the
1990s saw outstanding unparalleled growth in equity markets, which ended in
2000/2001. Nonetheless, the 12.0% return and 14.5% volatility of the S&P500
does not exactly meet the standard of high return paired with low volatility. In
2000/2001 the S&P500 experienced a 31.4% drawdown. The MSCI Europe
Index’s average annual return of 8.8% with a volatility of 15.3% is even less
inspiring and the MSCI World, with 6.8% returns and 14.7% volatility is below
the range of reasonable performance for an investment with this level of risk. The
worst drawdowns for the MSCI indices stand at the very intimidating figure of
–33.0% for Europe and –35.3% for the MSCI World. Finally, the Salomon Smith
Barney World Government Bond Ten Market Index (in US dollar terms) repre-
sents the traditional complement to stocks in the form of fixed income (including
corporate bonds). Although the volatility level is comparable to Hedge funds
(6.1%), the return of 6.6% is about half of the return of the average Hedge fund.

.... 119 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 120

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Unconditional correlation properties

Besides the very attractive risk-adjusted return characteristics, one of the most
attractive features of AIS is their favourable correlation to traditional asset classes. A
correlation matrix with the different strategy sectors and traditional assets is pre-
sented in Table 4.2. (The numbers presented here are linear correlations. The use of
‘rank correlation’ might be more appropriate in many circumstances.) By means of
diversification, including AIS in traditional portfolios provides asset managers with
the opportunity to produce better returns with lower levels of risk.
Short Selling strategies obviously have the lowest correlation to the equity markets
with a –0.64 correlation to the S&P500. The strategy also demonstrates low correla-
tions to the MSCI World and MSCI Europe (–0.6 and –0.48 respectively), as well as a
–0.07 correlation to the world bond index. Fixed Income Arbitrage strategies have a
slightly negative correlation to equity markets (–0.04) and a significant negative corre-
lation to bonds (–0.28). It should be noted for the discussion of correlations that
given the amount of data available (12 years of monthly data), the confidence interval
for the correlation coefficient being statistically significant (on a 95% level) is about
[–0.15:0.15]. This means that only correlation coefficients smaller than –0.15 or
larger than 0.15 are statistically significantly different from zero and correlation coef-
ficients within the interval [–0.15:0.15] can be considered equal to zero.
Equity Market Neutral, Regulation D, Convertible Arbitrage, Relative Value in
general, Distressed Securities and Global Macro have the next to lowest correlations
to the S&P500. Their correlations to the equity markets are low enough to act as an
effective diversifier and hedge, with figures showing 0.11, 0.31, 0.33, 0.37, 0.37
and 0.43 respectively. While not negatively correlated to the equity markets, all have
negative or very low positive correlations to bond markets. This property is valu-
able, as bonds are usually an integral part of traditional portfolios.
The following Hedge fund strategies have rather high correlations to tradi-
tional asset classes: Event Driven with a correlation of 0.60, Long/Short Equity
with 0.64 and Market Timing with 0.66. Their low correlations to fixed income
asset classes gives some validation to their Hedge fund classification, but these
strategies are not ideally suited from the point of view of diversification and cor-
relation with traditional asset classes, and especially not for an equity portfolio.

.... 120 ....

 TABLE 4.2 ■ Correlations between AIS and traditional investment classes

CA FX DS LSE EMN ED FIA GM EMT RD RV SS FoF FSA FDA FTF FP ME MW S&P SSB GSC
Convertible Arbitrage 1.00 0.01 0.59 0.46 0.14 0.62 0.13 0.40 0.29 0.21 0.51 –0.37 0.48 –0.09 0.10 –0.09 –0.13 0.24 0.30 0.33 –0.02 0.05
Currency Trading 0.01 1.00 –0.02 0.03 0.17 –0.06 –0.06 0.27 –0.03 –0.05 –0.09 0.18 0.23 0.69 0.40 0.72 0.09 –0.07 –0.03 –0.03 0.13 –0.00
Distressed Securities 0.59 –0.02 1.00 0.57 0.16 0.78 0.37 0.46 0.35 0.37 0.64 –0.48 0.57 –0.26 0.04 –0.20 –0.27 0.34 0.34 0.37 –0.16 0.01
Long/Short Equity 0.46 0.03 0.57 1.00 0.32 0.75 0.08 0.60 0.68 0.57 0.51 –0.76 0.75 –0.08 0.13 –0.08 –0.20 0.53 0.60 0.64 0.03 0.20
Equity Market Neutral 0.14 0.17 0.16 0.32 1.00 0.17 0.06 0.22 0.14 0.13 0.14 –0.14 0.30 0.03 0.10 0.12 –0.01 0.11 0.09 0.11 0.14 0.17
Event Driven 0.62 –0.06 0.78 0.75 0.17 1.00 0.20 0.57 0.48 0.41 0.63 –0.60 0.63 –0.13 0.04 –0.17 –0.24 0.50 0.55 0.60 –0.06 0.04
Fixed Income Arbitrage 0.13 –0.06 0.37 0.08 0.06 0.20 1.00 0.12 0.01 0.13 0.30 –0.04 0.26 –0.14 –0.06 –0.16 –0.14 0.09 0.02 –0.04 –0.28 0.05
8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 121

Global Macro 0.40 0.27 0.46 0.60 0.22 0.57 0.12 1.00 0.53 0.37 0.37 –0.41 0.71 0.26 0.39 0.27 0.02 0.44 0.44 0.43 0.06 0.02
Equity Market Timing 0.29 –0.03 0.35 0.68 0.14 0.48 0.01 0.53 1.00 0.46 0.29 –0.67 0.51 0.03 0.04 –0.01 –0.18 0.59 0.65 0.66 0.10 0.07
Regulation D 0.21 –0.05 0.37 0.57 0.13 0.41 0.13 0.37 0.46 1.00 0.37 –0.39 0.52 0.01 0.11 –0.09 –0.04 0.38 0.36 0.31 –0.11 0.18
Relative Value 0.51 –0.09 0.64 0.51 0.14 0.63 0.30 0.37 0.29 0.37 1.00 –0.32 0.46 –0.14 –0.02 –0.23 –0.21 0.37 0.37 0.37 –0.12 0.12
Short Selling –0.37 0.18 –0.48 –0.76 –0.14 –0.60 –0.04 –0.41 –0.67 –0.39 –0.32 1.00 –0.49 0.22 –0.01 0.24 0.26 –0.48 –0.60 –0.64 –0.07 –0.04
Fund of Funds 0.48 0.23 0.57 0.75 0.30 0.63 0.26 0.71 0.51 0.52 0.46 –0.49 1.00 0.09 0.41 0.16 –0.05 0.40 0.42 0.43 –0.08 0.21

.... 121 ....

Futures Sys. Active –0.09 0.69 –0.26 –0.08 0.03 –0.13 –0.14 0.26 0.03 0.01 –0.14 0.22 0.09 1.00 0.47 0.92 0.46 –0.05 –0.01 –0.07 0.30 0.05
Futures Discr. Active 0.10 0.40 0.04 0.13 0.10 0.04 –0.06 0.39 0.04 0.11 –0.02 –0.01 0.41 0.47 1.00 0.49 0.27 –0.09 0.01 –0.06 0.08 0.16
Futures Trendfollow –0.09 0.72 –0.20 –0.08 0.12 –0.17 –0.16 0.27 –0.01 –0.09 –0.23 0.24 0.16 0.92 0.49 1.00 0.41 –0.17 –0.16 –0.13 0.22 0.08
Futures Pass. (sGFI) –0.13 0.09 –0.27 –0.20 –0.01 –0.24 –0.14 0.02 –0.18 –0.04 –0.21 0.26 –0.05 0.46 0.27 0.41 1.00 –0.19 –0.23 –0.26 0.07 –0.11
MSCI Europe 0.24 –0.07 0.34 0.53 0.11 0.50 0.09 0.44 0.59 0.38 0.37 –0.48 0.40 –0.05 –0.09 –0.17 –0.19 1.00 0.86 0.69 0.32 –0.08
MSCI World 0.30 –0.03 0.34 0.60 0.09 0.55 0.02 0.44 0.65 0.36 0.37 –0.60 0.42 –0.01 0.01 –0.16 –0.23 0.86 1.00 0.84 0.31 –0.02
S&P 500 0.33 –0.03 0.37 0.64 0.11 0.60 –0.04 0.43 0.66 0.31 0.37 –0.64 0.43 –0.07 –0.06 –0.13 –0.26 0.69 0.84 1.00 0.16 –0.03
SSB World Govt. Bonds –0.02 0.13 –0.16 0.03 0.14 –0.06 –0.28 0.06 0.10 –0.11 –0.12 –0.07 –0.08 0.30 0.08 0.22 0.07 0.32 0.31 0.16 1.00 0.07
Goldmann Sachs CI 0.05 0.00 0.01 0.20 0.17 0.04 0.05 0.02 0.07 0.18 0.12 –0.04 0.21 0.05 0.16 0.08 –0.11 –0.08 –0.02 –0.03 0.07 1.00
4 ■ Empirical Properties of Alternative Investment Strategies

Source: Partners Group

Data: HFR (01.90–01.02). MAR (01.94–01.02)
8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 122

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

In the Managed Futures Universe, non-to-negative correlation properties are

more widely observed. All are either negatively correlated or show zero correla-
tion to the S&P500. The Passive Futures strategy sector has a statistically
significant negative correlation of –0.26 to the S&P, followed by trendfollowers
(–0.13) and Discretionary and Systematic Active strategies (–0.06 and –0.07
respectively). The correlation factors of the last three are not significantly differ-
ent from zero. Furthermore, Managed Futures strategies demonstrate rather low
correlations to bond markets with figures ranging between 0.07 and 0.3. With
these correlation properties to equity and fixed income investments, Managed
Futures generally serve as an excellent hedge (i.e. diversifier) within the global
portfolios of investors.
AIS also display interesting correlation properties among themselves. Short
Selling strategies have negative correlations to all other Hedge funds and
Regulation D strategies have low correlations to Relative Value, Macro and Event
Driven strategies (correlation numbers for illiquid strategies might be biased due
to the infrequent marking to market)10. Market Timing strategies have low corre-
lations to Distressed Securities and Relative Value strategies. Relative Value and
Macro strategies also have low correlations to one another. The correlations
between Managed Futures and Hedge funds also range from negative to zero.
Passive Futures strategies, for example, are positively correlated to only one
Hedge fund strategy (Short Selling).
As a final note on correlations, I would like to outline the variability of correla-
tion characteristics (which makes the application of quantitative optimisation
techniques very problematic). Figure 4.3 displays the rolling 12-month correlation
between the S&P500 index and the HFR Composite. For illustration the 12-
month rolling return of the S&P500 is also displayed. One can observe the large
fluctuations in the correlation figures ranging from about 0.3 to above 0.9.

.... 122 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 123

4 ■ Empirical Properties of Alternative Investment Strategies

1 0.5

0.9

0.4
0.8

0.7
0.3
0.6

0.5 0.2

0.4
0.1
0.3

0.2
0
0.1

0 –0.1
Dec Dec Dec Dec Dec Dec Dec Dec Dec Dec Dec Dec
90 91 92 93 94 95 96 97 98 99 00 01

FIGURE 4.3 ■ Rolling 12-month correlation between the S&P500 and the HFR Composite Index (black)
and the 12 month rolling return of the S&P500 (grey)

Conditional correlation properties

The correlation properties presented thus far do not tell the entire story about
diversification benefits. What investors really want is positive correlation in peri-
ods when equity and bond portfolios advance, and negative correlation in periods
when the value of traditional investments decline. This can be measured by look-
ing at conditional correlation. Conditional correlation is defined as the correlation
between two different assets measured only during particular periods. It can be
particularly useful in evaluating a portfolio’s sensitivity to extreme market moves.
Figure 4.4 displays the correlation of various AIS in different equity market condi-
tion, as represented by the performance of the S&P500, while Figure 4.5 presents
a different view on the same subject of conditional correlations by showing the
performance of selected strategies during the months of worst S&P performance.

.... 123 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 124

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

1.0

0.8

0.6 Equity Market Timer

Correlation to S&P500 in good months

Long/Short
0.4 Equity

Futures
Systematic Equity Market
0.2 Global Event
Active Neutral Driven
Macro
Reg D
0.0
Currencies
–1.0 –0.8 –0.6 –0.4 –0.2 0.0 0.2 0.4 0.6 0.8 1.0
Commodity Relative
Futures Long –0.2 Value
Passive Futures Distressed
Discretionary Securities
–0.4
Short
Seller
–0.6

–0.8

–1.0
Correlation to S&P500 in bad months

FIGURE 4.4 ■ AIS Conditional Correlation Properties

Source: Partners Group
Data: HRF, MAR

Systematic Active Futures strategies provide an example of favourable conditional

correlation. The strategy has a positive correlation of about 0.2 with the S&P500
during periods when the S&P500 is up and a negative –0.37 correlation when the
index is down. This property serves as an excellent hedge in an equity portfolio.
Note that the confidence intervals for statistical significance are about [–0.2:0.2] for
positive S&P500 return periods, and [–0.25:0.25] for negative S&P500 returns.
Systematic Passive Futures strategies also display interesting conditional correlation
features, posting a negative –0.3 correlation to the S&P500 when the index is down
and a correlation that is statistically not different from zero (–0.17) when the index is
up. While the returns are not necessarily correlated in advancing equity markets, the
strategies’ returns can very well serve to hedge against losses in declining equity

.... 124 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 125

4 ■ Empirical Properties of Alternative Investment Strategies

10% 3%
1%
5% –1%
–3%
0%
–5%
–7%
–5%
–9%
–10% –11%
–13%
–15% –15%
Mar 01

Mar 94

Mar 97

Jul 99

Mar 01

Mar 94

Mar 97

Jul 99
Jun 91

Jun 91
Feb 01
Sep 01
Nov 00
Jan 90

Sep 00
Sep 90
Jan 00

Jul 96

Nov 91

Nov 94
Oct 97
Feb 99

Apr 00
Feb 94
Sep 99
Sep 94

Feb 01
Sep 01
Nov 00
Jan 90

Sep 00
Sep 90
Jan 00

Jul 96

Nov 91

Nov 94
Oct 97
Feb 99

Apr 00
Feb 94
Sep 99
Sep 94
Aug 01
Aug 97

Aug 01
Aug 97
Aug 90

Aug 90
Aug 98

Aug 98
SGFII S&P500 Relative Value S&P500

4% 2%
2% 0%
0% –2%
–2% –4%
–4%
–6%
–6%
–8%
–8%
–10% –10%
–12% –12%
–14% –14%
–16% –16%
Mar 01

Mar 94

Mar 97

Jul 99

Mar 01

Mar 94

Mar 97

Jul 99
Jun 91

Jun 91
Feb 01
Sep 01
Nov 00
Jan 90

Sep 00
Sep 90
Jan 00

Jul 96

Nov 91

Nov 94
Oct 97
Feb 99

Apr 00
Feb 94
Sep 99
Sep 94

Feb 01
Sep 01
Nov 00
Jan 90

Sep 00
Sep 90
Jan 00

Jul 96

Nov 91

Nov 94
Oct 97
Feb 99

Apr 00
Feb 94
Sep 99
Sep 94
Aug 01
Aug 97

Aug 01
Aug 97
Aug 90

Aug 90
Aug 98

Aug 98

Equity Hedge S&P500 Hedge Fund Index S&P500

FIGURE 4.5 ■ Performance of selected strategies during equity down months

Data: HFR, partners Group

markets. This is especially useful in periods of great market volatility and uncertainty
(e.g. August/September 1998, fall 2000, early 2001, September 2001). Equity Market
Timing strategies, in contrast, are highly correlated to the equity markets in periods of
rising equity market (0.55) and almost uncorrelated in poor market periods (0.14).
This property is equally desirable for an investor. Returns in positive equity market
periods are not sacrificed, while losses in declining equity markets are avoided.
Positive correlation in up markets and in down markets is a feature of most
Long/Short Equity strategies. Event Driven, Distressed Security, Relative Value
and Macro strategies move in line with equity markets during periods of declining

.... 125 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 126

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

equity markets, but are largely uncorrelated to the S&P500 in advancing equity
markets. Short Selling strategies are naturally negatively correlated to the S&P500
in both advancing and declining equity markets. Convertible Debenture Arbitrage
(Reg D), Currencies, Discretionary Futures and Long Only Commodity strategies
show no statistically significant conditional correlation, either in rising or in
falling equity markets.

Behaviour in extreme market situations

During the last ten years or so there has rarely been a year without a move in the
range of ten standard deviations in one or more financial or commodity markets.
Investors have been faced with a variety of extreme market environments (e.g.
Japanese stock market crash, Gulf War, oil price boom, recession, bond market crash,
Mexico crisis, Asian crisis, Russian default, Brazil crisis, TMT crash, WTC terrorist
attacks). For the following discussion, I selected five such periods in order to illustrate
the behaviour of AIS during extreme market turmoil: the rising interest rates (bond
crash) in early 1994, the Asian crisis in 1997, the Russian default crises in 1998, the
‘TMT crash’ in 2000/2001 and the terrorist attacks in September 2001.11 The per-
formance of each strategy sector during these periods is displayed in Table 4.3. I also
included the performance of the average fund of funds as a best indication for the
return provided to investors in multi-manager AIS portfolios. Finally, the performance
of the two traditional asset classes (equity and bonds) is given:

1 From February to April 1994 the US and European yield curves shifted
upwards by about 150 basis points within the short period of two months,
after the FED unexpectedly raised interest rates. This led to major losses in
traditional bond portfolios. Within the AIS Universe, Global Macro and
Convertible Arbitrage managers suffered most from this unusual interest rate
move, with losses of –10.7% and –4.6%, respectively. Long/Short Equity
managers lost –2.9% and Active Futures returned between –3.0%
(systematic) and –5.0% (discretionary). Fixed Income Arbitrage and short
sellers on the other side made significant gains with +3.6% and +16.2%,
respectively. The average fund of funds lost –5.6% due to the high exposure
to Global Macro strategies most AIS portfolios had at the time.

.... 126 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 127

4 ■ Empirical Properties of Alternative Investment Strategies

2 During the Asian crisis following the devaluation of many Asian currencies in the
second half of 1997, all AIS earned significant positive returns led by Futures
strategies (Passive +14.0%), Long/Short Equity (+13.5%), Regulation D (+13.8%)
and Risk Arbitrage (+9.9%). Fixed Income Arbitrage with +1.7% displayed the
lowest returns of all strategies. Fund of funds returned 6.2% on average.
3 The market crisis following the Russian default in August 1998 is also referred
to as a ‘Hedge fund crisis’. The average fund of funds lost –11.6% in the
three-month period from August to October! This is mainly due to large
allocations in Fixed Income Arbitrage strategies (–13.2% from August to
October 1998) in many AIS portfolios at that time (the strategy had shown
very high performance in the two-year period preceding August 1998).
Distressed Security strategies also lost significantly (–12.4%). But the period
also had some clear winners. Futures strategies returned +8.9% (Systematic
Active), +4.8% (Systematic Passive) and +1.0% (Systematic Discretionary).
4 The equity market downturn and the burst of the technology bubble (‘NASDAQ
crash’) that started in March 2000 and led to a decline of 30.5% in the S&P500
during the following 19 months is largely seen as one reason for the tremendous
increase of investors’ interest in AIS. Interestingly, the performance of an average
fund of funds was actually negative (–2.6%) during this period! Again, this was
caused by the high exposure in many funds of funds to a strategy sector that lost
significantly; this time it was the Long/Short Equity sector that lost –7.9% in the
18-month equity bear market until September 2001. But the period also saw a
number of very successful strategies. Relative Value strategies provided very high
returns to investors (more than +20% for Convertible Arbitrage and Equity
Market Neutral, and +7.6% for Fixed Income Arbitrage). Futures strategies also
had a period of good performance (ranging +7.3% to +28.7%). But the stars of
the industry were short sellers with +95.6% return (according to HFR). Note that
the performance numbers for Short Selling strategies deviate significantly between
different data providers (CSFB/Tremont shows a return of about +43.7% during
this period). This can be explained by the fact that short selling managers are not
very numerous, and the information reported by each index depends strongly on
what managers included and how they are weighted in the index (see preceding

.... 127 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 128

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

discussion). Global Macro strategies were able to make a comeback. Note also here
that the CSFB/Tremont numbers (+28.8%) are significantly different from the
HFR numbers (+4.0%) presented in Table 4.3. A possible explanation here again
is the different weighting scheme. The Tremont indices are asset weighted, which
puts most emphasis on just a few large Hedge funds.
5 Finally, the month of September 2001 proved to be another real stress test for AIS,
which the industry, generally speaking, passed well. The average fund of funds lost
less than –2% compared to –8.2% for the S&P500 and +0.7% for bonds (Lehman
US Government Bond Index). Long/Short Equity, Fixed Income Arbitrage and Risk
Arbitrage lost –3.8%, –2.9% and –3.1% respectively, while Futures again served as
an excellent hedge in the AIS portfolio with returns of +2.9% (Systematic Passive).

TABLE 4.3 ■ AIS returns in extreme market environments

Bond Crash Asian crisis Russian default TMT Crash Terrorist attack
02.94–04.94 07.97–12.97 08.98–10.98 04.00–09.01 September 01
Convertible Arbitrage –4.62% 5.67% –4.69% 20.74% 0.72%
Fixed Income Arbitrage 3.58% 1.70% –13.18% 7.57% –2.92%
Equity Market Neutral 1.94% 7.31% –1.48% 21.43% 2.27%4
Risk Arbitrage 0.70% 9.91% –2.00% 13.78% –3.05%
Distressed Securities –1.38% 6.98% –12.43% 8.32% –1.02%
Regulation D N/A 13.83% 2.08% –7.42% –1.22%
Global Macro –10.70% 8.88% –5.93% 3.98%2 2.72%
Long/Short Equity –2.85% 13.50% –2.38%1 –7.89% –3.78%
Equity Market Timing –2.08% 8.01% 5.72% –1.68% –0.96%
Short Selling 16.15% 1.97% 4.15% 95.56%3 3.22%
Futures Systematic Active –2.98% 4.60% 8.91% 7.34% 0.19%
Futures Discretionary Active –4.98% 4.53% 0.99% 13.09% 0.83%
Futures Systematic Passive 0.71% 14.00% 4.81% 28.65% 2.87%
Fund of Funds –5.59% 6.23% –11.60% –2.62% –1.76%
S&P500 –6.37% 9.63% –1.96% –30.54% –8.17%
Lehman US Government Bond Index –9.34% 12.55% 6.62% 18.72% 0.74%
1 7.65% in 08.98
2 28.76% for the Tremont index
3 43.72% for the Tremont index
4 For Stat. Arbitrage: –1.95%

.... 128 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 129

4 ■ Empirical Properties of Alternative Investment Strategies

Benefits of AIS in a traditional portfolio

Risk–return and correlation properties have already given a clear indication of the
benefits of AIS in a generic traditional portfolio. In order to demonstrate the
effects of including AIS in a traditional portfolio of bonds and equity, Figure 4.6
presents the efficient frontiers of a traditional portfolio with and without invest-
ments in Hedge funds and Managed Futures. For the representation of a traditional
portfolio, the S&P500 Index, the MSCI World Index, the MSCI Europe Index and
the Salomon Smith Barney (SSB) World Government Bond Index were selected.
For Hedge fund representation, the HFR Composite Index as an equally weighted
basket of managers from all different Hedge funds strategies was chosen. For com-
parison I also included an efficient frontier analysis with the HFR Fund of Funds
Index (FoF) instead of the HFR Composite Index. Finally, for Managed Futures
strategies, I chose the Zurich Managed Futures Index and the sGFII. It should be
noted that the quantitative portfolio optimization techniques applied are subject to
certain assumptions about the risk and return inputs (see Chapter 7 for a more
detailed discussion on the pitfalls of portfolio optimization of AIS portfolios).
15%
Stocks/Bonds
14% Stocks/Bonds/HFR
Stocks/Bonds/HFR/Futures
13% Stocks/Bonds/FoF/Futures

12%
Return

11%

10%

9%

8%

7%
4% 6% 8% 10% 12% 14%
Volatility

FIGURE 4.6 ■ Efficient Frontier Analysis with and without AIS

Data: Time period: 01.90–01.02. HFR, MAR and SGFII were taken for AIS; MSCI, S&P and SSB bond index were taken for traditional
asset classes (see text)

.... 129 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 130

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The traditional portfolio displays annual returns ranging from just over 7% to 12%
(the S&P500 return), with a volatility ranging from about 6% to 14.5%. The top right
corner of the curve is a result of the equity contribution of higher return and higher
risk. On the bottom left corner of the curve, the efficient frontier displays the contribu-
tion of bonds to the portfolio; risk is minimized with returns being sacrificed as well.
The efficient frontier of the traditional portfolio changes dramatically when com-
bined with AIS. The addition of Hedge funds to a portfolio of equity and bonds
reduces the levels of volatility to a range from 4.7% to 7.4%. This substantial drop
in volatility occurs simultaneously with a dramatic rise in returns to a greatly
improved range of 10.4% to 15%. When Managed Futures are further added to the
equity, bond and Hedge fund portfolio, the risk levels are even more reduced, while
the returns are enhanced along the efficient frontier (the two efficient frontiers meet
in the point of highest return, which is the HFR return). The result is that the effi-
cient frontier curve for the traditional portfolio plus Hedge funds and Managed
Futures stretches from 11.5% to 15.0% in the return dimension, with only 3.8% to
7.4% volatility. The lowest level of return of the portfolio enhanced with Hedge
funds and Managed Futures is only slightly lower than the highest level of return for
the traditional portfolio without AIS. The same can be said for the risk dimension.
The highest risk point of the curve combining equity, bond, Hedge funds and
Managed Futures is only about 1% higher than the lowest risk level for the tradi-
tional portfolio without AIS, which corresponds to a 100% bond portfolio.
Also interesting is the comparison if the average fund of funds performance
instead of the HFR Composite is used. This combination also shows a significant
improvement compared to equities and bonds only, but compared with the analy-
sis using the HFR Composite Index, the portfolio with fund of funds has lower
return with about the same level of risk. The extra layer of fees that must be paid
to fund of funds managers explains a part of this difference.

Summary of empirical properties

The broad range of AIS demonstrates very attractive risk–reward characteristics.
Strategies with negative, zero and low correlations to traditional assets and/or to
each other offer attractive capacities for diversification and efficiency enhance-
ment within an investor’s global portfolio.

.... 130 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 131

4 ■ Empirical Properties of Alternative Investment Strategies

It is interesting to look at how AIS investors fared during times of market tur-
moil. Although quite diverse in their returns, Hedge funds and Managed Futures
generally showed much better performance than traditional investments during
these periods. Interestingly, in difficult market environments funds of funds have
shown much worse returns than the average strategy sector (even after accounting
for the extra fee level). I believe this performance pattern mirrors the fact that AIS
allocators tend to behave in a pro-cyclical way, over-weighting the best perform-
ing strategy sectors of the recent past. With too much money flowing into these
strategies they are often the first to experience problems during market turmoil.
Futures strategies generally performed well during these critical periods. This jus-
tifies their presence as a ‘hedge’ in a multi-manager AIS portfolio.
AIS investments generally have a meaningful impact on the efficient frontier for
investors who usually hold traditional assets such as stocks and bonds only. For peri-
ods when traditional markets suffer, some AIS display consistently positive returns. All
the benefits of AIS work dramatically to improve the efficient frontier of the tradi-
tional portfolio, producing significantly higher returns with substantially lower risk.

Appendix: Data providers for AIS

Altvest, New York (www.altvest.com) tracks 13 strategies from a database that
allows managers to input their own data. Each fund is assigned to the category in
which the largest percentage of its assets is invested. There are no performance
criteria for inclusion in the index. Index results are based on reports from more
than 1,400 Hedge funds in a database of 1,800 funds. Data goes back to 1993.
Altvest is owned by InvestorForce and the website requires registration in order to
gain access to the indices.
Capital Markets, EACM (www.eacmalternative.com), is based on 100 funds
selected to be representative of 13 strategies, arranged in five clusters. The index is an
equally weighted composite of unaudited performance information provided by the
funds. EACM bases its results on the same funds from month to month, allowing no
manager who had a bad month to avoid inclusion. Funds are assigned categories on
the basis of how closely they match the strategy definitions. Names of the funds are

.... 131 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 132

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

not disclosed. The index is rebalanced annually. It was launched in 1996 with data
going back to 1990. Capital Markets is owned by Evaluation Associates.
Credit Suisse First Boston/Tremont (www.hedgeindex.com) covers nine strategies
and is based on 340 funds, representing $100 billion in invested capital, selected from
a database of 2,600 funds. It is the only asset- (capitalization-) weighted Hedge fund
index. The CSFB/Tremont Index discloses its construction methods and identifies all
the funds within it. CSFB/Tremont accepts only funds (not separate accounts) with a
minimum of $10 million under management and an audited financial statement. If a
fund liquidates, its performance remains in the index for the period during which the
fund was active in order to minimize survivorship bias. The index was launched in
1999, with data going back to 1994. It incorporates the TASS+ database.
Hedge Fund Research (www.hfr.com) includes 29 categories plus subtotals.
The index is equally weighted and based on 1,100 funds, drawn from a database
of 1,700 funds. Funds of funds are not included in the composite index. Funds in
the database represent $260 billion in assets. The index was launched in 1994
with data back to 1990. Funds are assigned to categories based on the descriptions
in their offering memorandums. The indices aim at eliminating the survivor bias
problem by incorporating funds that have ceased to exist.
Hedgefund.net (www.hedgefund.net) covers 31 strategies arranged into three
subtotals. Called the Tuna Indices, they are updated from a database of 1,800 Hedge
funds and funds of funds. The data goes back to 1979 and managers select their
own categories. HedgeFund.Net is operated by Links Securities LLC, an NASD reg-
istered broker–dealer and owned by Links Holdings and Capital Z Investments.
Hennessee Group (www.hedgefnd.com) reports 22 investment-style categories.
The indices were created in 1987 and first published in 1992. Results are based on
450 funds, including 150 in which Hennessee clients are invested, from a database
of 3,000 funds. Assets of $160 billion are represented in the index. Each reporting
fund is placed in the category that reflects the manager’s core competency.
LJH Global Investments (www.ljh.com) tracks 16 different equally weighted
indices, each composed of approximately 50 managers. It is the only index that
presents performance exclusively in graph form. To be included, funds must have
an audited statement and have passed some level of LJH due diligence. Funds are

.... 132 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 133

4 ■ Empirical Properties of Alternative Investment Strategies

assigned to categories based on LJH’s screenings. The index is rebalanced quar-

terly or semi-annually, depending on the strategy. For a fee, LJH provides data on
index components.
MAR Futures (www.marhedge.com) reports especially on the performance of
Managed Futures strategies in each of 15 categories, ten of which are combined into
four sub-medians. The variety of Zurich (formerly MAR) index databases contains
1,300 funds. Managers usually select their own categories. The firm’s website iden-
tifies the number of funds and assets in each category. MAR, the former publisher of
the index, sold its database business to Zurich Financial Services in spring 2001.
A new index from Morgan Stanley Capital was announced to be available in
the second half of 2001. It will not be investable initially, although the option
remains under consideration. Morgan Stanley Capital claims that its index will be
‘institutional-brand quality’ and that it will be ‘different’. The database for the
new index will come from Financial Risk Management in London, an institutional
asset manager investing in Hedge funds.
The Parker Currency Trading Index (http://www.parkerglobal.com/fxindex.htm) is
a performance-based equally weighted benchmark that measures the returns of global
currency managers. In addition to the Parker FX Index, there are two style-driven
sub-indices: the Parker Systematic Index and the Parker Discretionary Index. The
former tracks those managers whose decision process is rule based and the latter
tracks those managers whose decision process is judgmental. The Parker FX Index
currently includes 43 programs managed by 37 firms located in the USA, Canada, UK,
Ireland and Switzerland. The 43 programs manage over $8 billion in currency assets
and include a combination of 33 programs that are systematic and ten programs that
are discretionary. Disciplines include technical, fundamental and quantitative.
Van Hedge Fund Advisors International (www.vanhedge.com) is derived from
the performance of an average of more than 750 funds separated into US and off-
shore funds covering 14 strategies and combined into a separate global index. There
are no performance or size criteria and funds are assigned to categories based on
their offering memorandums and interviews with the individual managers.
Zurich Capital Markets (http://www.zcmgroup.com) provides the Zurich Hedge
Fund indices in partnership with TRS Associates to track the performance of various
Hedge fund strategies. They differ from existing Hedge fund indices by focusing

.... 133 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 134

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

only on those funds/managers most likely to be considered for investment by institu-

tional and other sophisticated investors. Focusing is on those funds/managers that:
(1) are strategy pure in their style; (2) have a two-year minimum performance track
record; and (3) have sufficient assets under management to demonstrate organiza-
tional and managerial infrastructure, scalable strategies and the ability to raise funds
from sophisticated investors. Highly leveraged strategies and strategies involving
complex derivatives are excluded. There is no requirement that these funds be open
to new investments, because the indices themselves are not designed to be directly
investable and are constructed only to reflect the returns of a particular Hedge fund
strategy. The indices are equally weighted.

Notes
1. There are numerous empirical studies showing the benefits of AIS in
traditional portfolios. Please refer to ‘The Benefits of Alternative Investment
Strategies in the Institutional Portfolio’ by L. Jaeger, saisGroup Research Paper,
2001, available on http://www.saisgroup.com; ‘The Benefits of Hedge Funds’ by
T. Schneeweiss and G. Martin, Lehman Brothers Publications, 2000; ‘Portfolios of
Alternative Assets: Why not 100% Hedge Funds?’ by R. McFall Lamm Jr., The
Journal of Investing, Winter 1999; ‘The Performance of Hedge Funds: Risk
Return, and Incentives’, Ackermann et al., Journal of Finance, 1999.
2. An excellent and more detailed discussion of the empirical properties of AIS is
provided by A. Ineichen, ‘In Search of Alpha’, October 2000.
3. The Sharpe ratio, however, is not the only and arguably not the most
appropriate measure for risk-adjusted performance. There is a large variety of
different performance measures. Please refer to Dacorogna et al., ‘Effective Return,
Risk Aversion and Drawdowns’, Physica A, January 2001, 289, p.229–48 for an
excellent discussion and further references.

∫(
x–m 4
4. Kurtosis is most often defined as k
____
σ )
dx (where m is the mean, σ the
standard deviation and k a normalization factor). The kurtosis of the standard
normal distribution is three. In the numbers presented here, kurtosis is defined as

∫(
x–m 4
)
k ____ dx – 3 .
σ

.... 134 ....

8409 Chapter 4 p111-135 11/4/02 1:13 PM Page 135

4 ■ Empirical Properties of Alternative Investment Strategies

5. See ‘Performance Characteristics of Hedge Funds and Commodity Funds:

Natural versus Spurious Biases’ by W. Fung and D. Hsieh in the Journal of
Financial and Quantitive Analysis, 2000.
6. See also the work by Schneeweiss Partners, A Review of Alternative Hedge
Fund Indices.
7. A discussion of the different issues in AIS index construction is also presented
in ‘Hedge Fund Indices’ by G. Crowder and L. Hennessee, Journal of Alternative
Investments, Summer 2001; see also ‘The Benchmark Bane’ in The Economist,
August 31, 2001.
8. See also the article by G. Amin and H. Kat, ‘Welcome to the Dark Side:
Hedge Fund Attrition and Survivorship Bias over the Period 1994–2001’.
9. sGFI stands for ‘saisGroup Futures Index’, a systematic Passive Futures
strategy index developed by the author and his partners at saisGroup (now
Partners Group). Please refer to the following article for further details about this
index: L. Jaeger, P. Cittadini and M. Jacquemai, ‘The saisGroup Futures Index
(sGFI) – A New Passive Futures Investment Strategy’, available on
http://www.saisgroup.com.
10. See the article by C. Asness et al., ‘Do Hedge Funds Hedge?’
11. The study was published in German in the Neue Zürcher Zeitung (Jan 8,
2002): ‘Hedge Funds in Marktturbulenzen’.

.... 135 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 136

CHAPTER 5

Risk in Alternative
Investment Strategies

Alternative Investment Strategies entail the search for ‘alpha’ combined with the
proper management of the underlying risks. Whereas for traditional investments,
sector and instrument selection next to portfolio diversification are the key per-
formance drivers, for AIS it is strategy sector selection together with downside
risk management, which are the most critical determinants of investment perform-
ance. In Chapter 3, I outlined the return drivers and risk characteristics of the
single AIS sectors. In this chapter, I discuss general risks and the strategy-specific
risk management principles of AIS. In Chapter 7, I go on to outline the top down
sector and bottom up manager selection process in an integrated risk management
framework for the AIS portfolio. Important here will be the distinction between
‘pre-investment’ risk management (‘top down’ strategy allocation, ‘bottom up’
manager due diligence, portfolio diversification) and ‘post-investment risk man-
agement’ (monitoring, active risk control).

Risk factors in AIS investments

AIS investments are inherently more complex than traditional investments. Hedge
funds and Managed Futures trade a wide spectrum of different asset classes and
instruments and include features such as leverage, Short Selling, complex deriva-
tives trading, spread positions and low liquidity. While also present in traditional

.... 136 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 137

5 ■ Risk in Alternative Investment Strategies

portfolios, many risk factors are more difficult to analyze in an AIS context
because of their complex interaction with each other.1

■ Market risk: Market risk is the risk of loss due to unexpected and adverse
price moves or changes of volatility in the broad markets or single sectors.
■ Credit risk: Credit risk is the risk of counter-parties defaulting on their obligations
or of changes in the market’s sentiment about the probability of their default.
■ Liquidity risk: Liquidity risk is twofold: (1) The risk of loss due to the
(temporary) inability to unwind a position at a normal bid/ask spread; (2) the risk
of not being able to fund investment leverage. Liquidity risk in AIS is magnified
by a certain herding behaviour of AIS managers in times of turbulence.2 An
example is October 1998, when many Hedge funds exited their yen carry trades
simultaneously, which led to a 15% rise of the Japanese currency against the US
dollar (a normally extremely liquid market) within just a few hours.
■ Common factor risk: Common factor risk is the risk inherent in some, but not
all, securities. (e.g. industry-specific risk, geographic risk, etc.).
■ Operational risk: Operational risk is the risk of failure of internal systems,
technology, people, external systems or physical events.
■ Event risk: Event risk is the risk of an extraordinary political or economic
event, e.g. unexpected election outcome, military events, sovereign default.
■ Corporate event risk: Corporate event risk is the risk of loss due to an exposure
to a particular firm and a specific event affecting its value, e.g. surprise
announcements like earnings revisions, mergers or changes of management.
■ Model risk: Model risk is the risk of a model mis-specification, which can lead
to incorrect valuation of a financial instrument or the model not being suitable
for the particular trading approach.3

An AIS risk manager has to understand the complex relationship between market risk,
manager risk, liquidity risk, counter-party risk, pricing risk and leverage.4 Liquidity
risk is often connected to the leverage employed by the fund. While it is still current
practice to consider market risk, credit risk and liquidity risk separately, risk experts
increasingly argue that risk management should treat these risks in combination.

.... 137 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 138

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Because Hedge funds and Managed Futures have a much higher transaction
turnover and usually fewer back-up staff compared to traditional asset managers,
operational risks (e.g. exceeding position limits or erroneous execution) can be
significantly higher. Further, many Hedge funds depend on systematic quantitative
models for buy and sell signal generation. Inappropriate models can expose AIS
investments to model risk. AIS are also subject to event risk of many different
kinds. The summer/fall of 1998 showed that tumultuous financial markets can
affect Hedge funds just as negatively as traditional investments.
But AIS risks extend beyond the risks shared with traditional investments.
Investors and risk managers must consider the following additional risks:

■ Lack of transparency: Despite growing investor awareness of its importance,

there remains strong resistance against transparency in the AIS industry. There
are good reasons why investment and trading activities at financial institutions
(banks, insurances, brokers, pension funds) are generally subject to disclosure
to clients and regulatory authorities. The same logic should apply to the
trading activities of AIS managers; however, transparency in AIS is far from
universal. Lack of transparency and insufficient investor control are among the
main reasons for the following risk factors.
■ Manager risk: AIS returns are to a high degree a function of the individual AIS
manager’s skill. Much discretionary decision-making power is concentrated in one
or a few individuals. Overconfidence, in extreme cases hubris, is one of the
most important origins of unexpected losses encountered by single AIS
managers. The legal set-up of a Hedge fund gives the manager almost unlimited
freedom in their investment process. Typical disclosure documents provide for
broad manager authority along the lines of the following:

The program’s objective is to achieve substantial long-term capital appreciation

while offering an efficient diversification investment vehicle to traditional
assets such as bonds, equities or real estate based on the systematic application
of computerized trading strategies or fundamental models. There are no
restrictions or limitations to the markets, exchanges, futures, options and
forward contracts that can be traded by the advisor.

.... 138 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 139

5 ■ Risk in Alternative Investment Strategies

The wide freedom provided to the manager creates idiosyncratic risks such as
single-person dependency, style drift and ‘bets’.
■ Leverage risk: For AIS investments, leverage is not always undesirable and in
some cases even necessary to achieve return targets. An excessive amount of
leverage can be disastrous, however. In volatile markets, AIS easily becomes a
captive of leverage. Leverage risk has two components: volatility and
financing. To the same degree that leverage enhances returns, it also increases
investment risk. The problems of leverage become apparent with the advent of
the unexpected, when leverage can quickly turn moderate losses into
investment disasters. For strategies that use external financing for leverage,
sudden unavailability of financing can cause significant problems. A single
margin call on a position can destroy an entire portfolio, if the manager is
forced to sell at an inopportune time. It should be noted that prime brokers
tend to change their haircut policies and withhold financing exactly at those
moments when it is most needed.
■ Counter-party risk: There are numerous incidences where counter-parties’
change of policies and withdrawal of funding support caused severe losses to
AIS funds. This usually occurs in stress situations, when several risk factors act
simultaneously. The story of MKP Capital Management LLC, a mortage-backed
Securities Arbitrage manager, provides an illustrative example of counter-party
risk in combination with leverage, liquidity and legal risks. The manager was
forced to liquidate a significant part of his $370 million fund in October 1998,
when the prime broker, Salomon Smith Barney (SSB), stopped accepting certain
securities as collateral for MKP’s margin payments and at the same time refused
to give credit to MKP for unrealized returns generated by securities held in
SSB’s account. Furthermore, according to the lawsuit filed by MKP, SSB refused
to confirm the trades that the manager was able to negotiate in the process of
the liquidation, claiming lack of administrative support for the heavy trading
volume. The fund ended up losing hundreds of millions of dollars. Three years
after the event the case is still with the courts.5,6

.... 139 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 140

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Risk of asset–liability mismatch: The investor should make sure that the
investments of the AIS manager are liquid enough to fulfil redemption
obligations to investors. It is possible for Hedge funds to have a mismatch
between portfolio liquidity and redemption policies.
■ Capacity risk: Most AIS strategies have limited investment capacity, i.e. they can
only handle a certain amount of money without significantly diluting their
performance. Investors might not be able to invest with the managers they find
most attractive. High quality managers tend to close their program quickly. The
investor should be aware of the capacity limits of the strategy. Some managers
dislike rejecting new investors’ money when their capacity limit is reached.
■ Fraud risk: In an unregulated industry with about 5,000 Hedge fund products
offered, a certain number of these unavoidably involve fraud. Although a
rather rare occurrence, fraud in the form of false performance reports and
audits or Ponzi schemes have in the past resulted in significant losses for
investors. The names John Natal (Cambridge Fund) and Michael Berger
(Manhattan Fund) are linked with Hedge fund fraud.7
■ Data risk: Unlike traditional asset classes, where performance data and benchmarks
are readily and reliably available, the infrastructure of performance data for AIS is
still rather undeveloped and reliability of the data is low.8 The results of
performance studies for AIS often depend on the database used (see Chapter 4 for
a more detailed discussion). AIS data is biased in several different ways. It usually
does not go back far into the past and is largely drawn from the period of the
exceptional equity bull market in the late 1980s and 1990s. The resulting bias can
be illustrated by strategies containing ‘short option’ or ‘short volatility’ elements,
which generate steady returns containing a certain but low probability of large
losses. Losses may simply not yet have been observed during the measurement
period. Further, the survivorship bias (i.e. the bias due to liquidated strategies
dropping out of the pool of reported managers) can lead to reported performance
being 2–3% higher than actual returns. Finally, not all AIS managers disclose their
performance to reporting services (see Chapter 4 for a more detailed discussion).
■ Performance measurement risk: Traditional equity or bond managers’ performance
is usually measured against specific benchmarks (indices). In contrast, AIS are

.... 140 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 141

5 ■ Risk in Alternative Investment Strategies

considered ‘absolute return’ investments, i.e. they do not have explicit

performance benchmarks. Benchmarks as used in traditional asset management
are unlikely to be developed for AIS. Their definition and use requires (a) an
unambiguous classification of the strategy universe, (b) a good representation of
the investment universe benchmarked and (c) the possibility of being replicated.
These requirements cannot easily be applied to AIS investments.9
The emphasis of AIS managers is on how much absolute return their
strategy generates. Although the focus of many managers has shifted from
absolute return targets to quantifying performance in terms of ‘risk-adjusted
returns’, the best way of measuring risk-adjusted return is still the subject of
intense discussion.10 Many investors and managers use the Sharpe ratio or
one of its variations. The Sharpe ratio has its set of difficulties as a
performance measure, however.11 Most investments that earn premiums for
taking particular risks, and they can be exposed to asymmetric payoffs
(similar to a ‘short option’ position). Sharpe ratios are unsuited for
measuring the risk-adjusted performance of investments with skewed return
distributions, because the standard deviation does not account for risk
appropriately in this situation (see Chapter 2). Choosing inappropriate
criteria for non-benchmarked performance measurement imposes the risk of
introducing performance targets that do not match the investment
objectives. The examination of the performance track record of individual
managers has to be complemented with other types of analysis. A track
record should serve as a basis for questioning the past rather than for
predicting future performance.
■ Pricing risk: The pricing and NAV calculation for investment funds is not
guided by unique standards. Some Hedge fund strategies trade illiquid
instruments, which cannot be reliably priced on a regular basis and whose
market values the manager usually estimates himself. This leaves generous
space for performance smoothing. In some cases, the securities may be so
illiquid that they are not mark to market for several months. Thus, there may
be significant lags between reported returns and ‘real’ returns.12 An interesting
survey on NAV calculation and fair value practices used in the broad

.... 141 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 142

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

investment industry13 displayed that 13% of all funds (including Mutual funds,
Hedge funds and others) make some kind of adjustment to the prices they
receive from their valuation sources. This percentage is likely to be
significantly higher for Hedge funds only. The most frequently made
adjustments mentioned are write-downs for liquidity, price adaptations to time
zone and proprietary pricing models for complex derivative instruments.
■ Portfolio risk: As discussed, the risk profile of AIS can differ considerably from
traditional investments. The amount of variation in Hedge fund returns that can
be explained by the risk factors used in common asset-pricing models is rather
low.14 It is therefore difficult to map AIS investment risks onto standard risk
factors like equity, fixed income, FX, commodity risk or credit risk. As a result,
if included into the traditional portfolio, the heterogeneous nature of AIS
strategies makes it more difficult to obtain an overall picture of portfolio risk.
■ Legal risk: Hedge fund managers can find themselves in situations where they are
unable to enforce their claims other than through a lengthy legal process. This is
mainly true for very illiquid strategies such as Distressed Securities or Regulation
D strategies. Another aspect of legal risk concerns the relationship between the
investor and the fund. Due to the litigious nature of the US legal system, offering
memorandums for Hedge funds are almost always one-sided in favour of the AIS
manager. The legal documents, which should be carefully read and understood by
the investor prior to investment, are the fund prospectus, subscription agreement,
investment management agreement and advisory contracts.
■ Administrative risk: The Hedge fund investor relies strongly on the quality
and responsibility of administrators and auditors in controlling the assets and
the accuracy of financial statements. Often some basic infrastructural
requirements for the performance of these tasks are not present in offshore
locations that many AIS managers choose as fund domiciles, including
independent mark to market pricing (which is too often left in the hands of
the Hedge fund managers) and the auditing of financial statements. The case
of the ‘Manhattan Fund’ provides an illustrative example, where the fund
manager, Michael Berger, was able to deliver false account statements to the
administrator and auditor for an extended period of time.

.... 142 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 143

5 ■ Risk in Alternative Investment Strategies

■ Regulatory risk: Most AIS funds are registered offshore. The choice of
domicile for a Hedge fund or Managed Futures fund is critical. Changing
regulatory or tax requirements can be damaging to investors’ interests.
■ Systemic risk: Systemic risk involves the contagious transmission of a sudden shock
due to the financial system’s exposure to one or more failing institutions. This
shock might result in damage to the financial system to such an extent that the
economic activity in the wider economy can suffer. It is feared that the large size
of a single Hedge fund can expose the wider financial market to systemic risk. The
near collapse of LTCM is often referred to as an example of systemic risk.

Risk from an investor’s point of view

Most AIS investors are not able to devote sufficient resources to the selection of
strategies and managers and to the necessary degree of risk management. They there-
fore rely on a professional fund of funds manager to cope with the details of the
various risks outlined above. Therefore, many private and institutional investors have
a broader perspective on risk and are possibly concerned about the following issues
when investing in AIS (in order of least to most impact on the investor’s portfolio):

1 Investment underperformance: While the risk of investment performance falling

short of expectations can never be entirely eliminated, it can be reduced by a
thorough manager due diligence and strategy allocation process through which
the best trading talents are picked within the most attractive sectors.
2 Inappropriate strategy sector allocation: Inappropriate sector allocation may
lead to a risk–returned profile that is unsuited for the investor. This type of
risk can be minimized if clear investment objectives, (risk-adjusted)
performance targets and well-defined risk parameters are established and used
as the drivers of the sector allocation process. Further, sector allocation should
not be static; it must be subject to ongoing review.
3 Severe decline in account value: This is (besides fraud) the greatest fear of
investors. A sharp and significant drop in the portfolio value can be caused by
all of the following:

.... 143 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 144

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ a deviation from the stated strategy (style drift or an unexpected ‘bet’ by

the trading manager)
■ an existing risk factor that the investor did not understand or recognize
(most often an event or a market behaviour that was not anticipated)
■ risks that were known and accepted as part of the investment strategy.
In order to cope with the risk of severe loss, continuous monitoring of
exposure and risk is necessary. In addition to allowing for the early
recognition of deviations from a stated strategy (‘style drifts’) and the
identification of undesired ‘bets’, exposure and risk monitoring can help
detect previously unknown risks (particularly in extreme market
environments). However, risks as a result of previously known and accepted
risks cannot be entirely eliminated.
4 Fraud: This is a risk in any investment, but due to the specific focus on
individual trading managers and the participation in private and mostly
unregulated investment vehicles, this risk is more severe for AIS investments
compared to traditional investments. Fraud can occur in many different ways
ranging from rather mild forms (e.g. undisclosed kickbacks to third parties) to
very severe forms (e.g. manipulated performance statements). Exposure to
fraud is the ultimate nightmare for investors. In many cases of fraud the
investor loses all or a large portion of his investment. A good safeguard is to
know the individuals involved and to check every detail of their background.
Continuous monitoring of the manager’s trading activity decreases fraud risk
significantly. Detailed reading of the fund’s offering memorandum is essential.
Ultimately, the best way to safeguard against fraud risk is to separate the
trading manager from the money, i.e. establish a managed account.
From an investor’s perspective, the gradations between fraud and bad business
practice, between malicious intent and ineptitude may not be so relevant.
Investors would like to avoid any kind of irregularity that can potentially
harm them. One irregularity is a breakdown in the integrity of a manager. The
initial due diligence and constant monitoring should uncover potential
irregularities before they occur.

.... 144 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 145

5 ■ Risk in Alternative Investment Strategies

Effective risk management begins with the ability to control. In the process of the
institutionalization of the AIS industry and in light of the increasing number of AIS
investors with fiduciary responsibility, investing in a ‘black box’ must be considered
more and more unsuitable. The risk manager should have regular knowledge about
the activities of the AIS manager. At the same time, long redemption periods make
control impossible. Ideally, if the allocation is sufficiently high, the AIS allocator
should make an investment with an AIS manager through a managed account,
which provides him with daily transparency and the highest possible level of liquid-
ity (which, of course, ultimately depends on the traded securities).

Risk factors of the different AIS sectors

As discussed in Chapter 3, the degree and type of risk varies substantially among strat-
egy sectors. Global Macro, Long/Short equity, Short Sellers and Futures/Currency
strategies are strongly exposed to broad market risk, while credit risk is an important
element of the risk profile for Distressed Securities and Regulation D strategies. Some
Relative Value (Convertible Arbitrage) and Event Driven strategies (Risk Arbitrage
and Distressed Securities) are highly exposed to corporate event risk. Liquidity risk
requires significant attention for Fixed Income Arbitrage, Regulation D and Distressed
Securities strategies.
AIS strategies are often distinguished in terms of their exposure to market risk
(equity, interest rates, FX, commodities) and credit risk.15 Sometimes liquidity risk
is added as an extra dimension. Figures 5.1–5.10 introduce a further differenti-
ated view on AIS risk and illustrate the various risk types for each of the strategy
sectors using a scale from zero to five (zero represents no sensitivity to the risk
factor and five represents maximal sensitivity). The illustration is of rather a
descriptive nature here and should serve its illustrative purpose.

.... 145 ....

 Co Co
nv nv
er er
tib tib
Fix le Fix le
A

FIGURE 5.1
0
4
5

1
2
ed

3
ed

0
4
5

1
2
3
rb Ar

■
Inc itr Inc bit
om ag om ra
eA e ge
Eq Eq eA
u ity rb uit rb
itr itr
M ag yM ag
ar
ke e ar e
ke
t Ne tN
ut eu
Ris ra tra
kA l Ris
kA l
Di
str rb Di
str rb
es itr itr
ag es ag
se
d e se
d e
Se Se
8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 146

c ur cu
itie rit
Re
gu s Re ies
gu
lat lat
ion ion
Gl
ob D Gl D
ob
al al
Lo M Lo M
ng ac ng ac
/Sh ro /Sh ro

.... 146 ....

Eq or Eq or
u ity tE uit tE
M qu yM qu
ity ity

Equity Volatility
ar ar
ke ke
t
Equity Market Direction

tT

Arbitrage usually displays an inverse sensitivity to equity volatility)

Tim im
Fu Fu
tu Sh
ing tu ing
re o re Sh
sD rt s Di or
isc Se sc tS
re llin re ell
Fu t ion g Fu t ing
tu tu ion
re ar re ar
sS yA s Sy yA
Fu ys cti cti
te ve Fu ste
tu m tu m ve
re at re at
s Sy ic s ic
MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Ac Sy
s te tiv ste Ac
tiv
m e m e
at at
ic ic
Pa Pa
ssi ssi
ve ve
Directional market risk: Sensitivity to falling (for Short Selling: rising) stock markets

FIGURE 5.2 ■ Non-directional equity risk: Sensitivity to stock market volatility (note that Convertible
 Co Co
nv nv
er er
tib tib
Fix le Fix le

FIGURE 5.4
A

FIGURE 5.3
ed ed Ar

0
4
5

1
2
3
0
4
5

1
2
3
rb

■
■
Inc itr Inc bit
om ag om ra
e ge
Eq eA Eq eA
u ity rb uit rb
itr yM itr
M ag ag
ar
k e ar
ke e
et tN
N eu eu
Ris tra Ris tra
kA l kA l
Di
str rb Di
str rb
itr es itr
es ag ag
se
d e se
d e
Se Se
cu cu
8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 147

rit rit
Re ies Re ies
gu gu
lat lat
ion ion
Gl
ob D Gl
ob D
al al
Lo M Lo M
ng ac ng ac
ro ro

.... 147 ....

/Sh /Sh
Eq or Eq or
uit tE uit tE
qu qu
Yield curve

yM yM
ar ity ar ity
ke ke
tT tT
Fu im Fu im
tu Sh
ing tu Sh
ing
re re
sD o rt s or

Credit risk (default and credit spreads)

Di tS
5 ■ Risk in Alternative Investment Strategies

isc Se sc ell
re llin re
Fu tio g Fu t ion ing
tu na tu
re r re ar
sS yA s Sy yA
Fu ys cti Fu ste cti
te ve m ve
tu m tu
re at re at
sS ic s Sy ic
ys Ac Ac
Yield curve risk: Sensitivity to parallel shift in interest rate term structure

te tiv ste tiv

m e m e
at at
ic ic
Pa Pa
ssi ssi
ve ve

Credit risk: Sensitivity to increasing default probability and widening credit spreads
 Co Co
nv nv
er er
ti tib
Fix ble Fix le

FIGURE 5.5
ed Ar ed Ar

0
4
5

1
2
3
0
4
5

1
2
3

■
Inc bit Inc bit
om ra om ra
ge ge
Eq eA Eq eA
u ity rb uit rb
itr yM itr
M a ag
ar
k
ge ar
ke e
et tN
Ne eu
ut tra
Ris ra Ris
l
kA l kA
Di
str rb Di
str rb
itr es itr
es ag ag
se
d e se
d e
Se Se
8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 148

c ur cu
itie rit
ies
Re
gu s Re
gu
lat lat
ion ion

systems or physical events (including model risk)

Gl
ob D Gl
ob D
al al
Lo M Lo M
ng ac ng ac
/Sh ro /Sh ro

.... 148 ....

Eq or Eq or
u tE uit tE
Liquidity

ity qu yM qu
M ity ity
ar ar
ke ke
t tT
Fu
Tim
Fu im
tu Sh
ing tu Sh
ing
re re
sD or s or

Operational risk (including model risk)

tS Di tS
isc sc ell
re ell re
Fu t ing Fu t ion ing
tu ion tu
re ar re ar
s Sy yA s Sy yA
Fu ste cti Fu ste cti
Liquidity risk: Sensitivity to liquidity decreasing in financial markets

tu m ve tu m ve
re at re at
sS ic s ic
A Sy
MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

ys cti ste Ac
te m
tiv
m ve
at e
at ic
ic
Pa Pa
s siv ssi
e ve

FIGURE 5.6 ■ Operational risk: Risk of failure due to internal systems, technology, people, external
 Co Co
nv nv
er er
tib tib
Fix le Fix le

FIGURE 5.8
FIGURE 5.7
ed

0
4
5

1
2
3
Ar

0
ed

4
5

1
2
3
Ar

■
■
Inc bit Inc bit
om ra om ra
ge ge
Eq eA Eq eA
u ity rb uit rb
itr
itr yM
M a ag
ar ge ar e
k et ke
tN
N eu eu
t ra Ris tra
Ris
kA l kA l
Di rb Di
str rb
str itr es itr

Lack of transparency
es ag ag
se
d e se
d e
Se Se
cu cu
8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 149

rit rit
Re ies Re ies
gu gu
lat lat
ion ion
Gl D Gl
ob D
ob
al al
Lo M Lo M
ng ac ng ac
ro /Sh ro

.... 149 ....

/Sh
Eq or Eq or
uit tE uit tE
yM qu yM qu

Transparency
Idiosyncratic equity risk: Risk of corporate event
ar ity ar ity
Corporate event

ke ke
tT tT
Fu im Fu im
tu ing tu Sh
ing
re Sh re
sD or s Di or
tS
5 ■ Risk in Alternative Investment Strategies

isc tS sc
re ell re ell
Fu tio ing Fu t ion ing
tu n tu
re ar re ar
sS yA s Sy yA
Fu ys cti Fu ste cti
te ve m ve
tu m tu
re at
re at
ic s ic
sS Sy Ac
ys Ac ste
te tiv m
tiv
e
m e at
at ic
ic Pa
Pa
ssi ssi
ve ve
 Co Co
nv nv
er er
tib tib
Fix le Fix le

FIGURE 5.9
ed

0
4
5

1
2
3
Ar

0
ed

4
5

1
2
3
Ar

FIGURE 5.10
■
Inc bit Inc bit

■
om ra om ra
ge ge
Eq eA Eq eA
uit rb uit rb
yM itr yM itr
ag ag
ar
ke e ar
ke e
tN tN

Leverage risk
eu eu
Ris tra Ris tra
kA l kA l
Di
str rb Di
str rb
es itr itr
ag es ag

Capacity constraints
se
d e se
d e
Se Se
8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 150

cu cu
rit rit
Re ies Re ies
gu gu
lat lat
ion ion
Gl
ob D Gl D
ob
al al
Lo M Lo M
ng ac ng ac
/Sh ro /Sh ro

.... 150 ....

Eq or Eq or
uit tE uit tE
Leverage

Capacity
yM qu yM qu
ar ity ar ity
ke ke
tT tT
Fu im Fu im
tu Sh
ing tu ing
re re Sh
sD or
tS s Di or
isc sc tS
re ell re ell
Fu tio ing Fu t ing
tu
re n ar tu
re
ion
ar
sS yA s Sy yA
Fu ys cti cti
te ve Fu ste
tu m tu m ve
re at re at
sS ic s ic
MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

ys Ac Sy Ac
te tiv ste tiv
m e m e
at at
ic ic
Pa Pa
ssi ssi
ve ve
8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 151

5 ■ Risk in Alternative Investment Strategies

Broad equity market (‘beta’) risk comes in two forms: directional (price) and non-
directional (volatility) risk (Figures 5.1 and 5.2). Next to the ‘beta’ risk (also called
‘systemic risk’), financial market theory (the CAPM) differentiates the idiosyncratic
(residual) portion of equity risk, which is denoted ‘corporate event risk’ in Figure
5.7. Interest rate risk is a function of possible yield curve shifts (Figures 5.3). Credit
and liquidity risk (Figures 5.4 and 5.5 respectively) often appear simultaneously
(except in the case of short sellers, where liquidity risk represents the risk of a ‘short
squeeze’). Figures 5.6, 5.8, 5.9 and 5.10 display the different strategies’ exposure to
operational risk, the risk of insufficient transparency, leverage risk and the degree of
capacity constraints.
Directional and corporate specific equity market risk ranges from high for
opportunistic equity strategies (Long/Short Equity, Short Selling) and certain Event
Driven strategies (Regulation D, Distressed Securities) to low or non-existent for
most Futures strategies. By making bets on the direction of certain equity market
sectors and other financial markets (e.g. currencies) Global Macro strategies are
directly exposed to directional price risk (in equity, fixed income, FX markets).
Convertible Arbitrage, Equity Market Neutral and Fixed Income Arbitrage in con-
trast bear little exposure to directional price moves in the broad equity markets and
Risk Arbitrage strategies find themselves in an intermediate range for directional
equity risk. But some Relative Value strategies (Convertible Arbitrage and Equity
Market Neutral) are exposed to changes in the volatility of equity markets as well as
corporate event risk. Note that Convertible Arbitrage carries an inverse sensitivity to
equity market volatility, performing well in periods of increasing volatility and
showing lower returns in periods of decreasing volatility.
Yield curve risk is most prevalent for Fixed Income Arbitrage strategies, which tend
to also be exposed to credit and liquidity risk. The most illiquid strategies, namely
Distressed Securities and Regulation D, also come with significant credit risk.
Operational risk in the form of model risk is present in those strategies that rely on
pricing and technical trading models, which are mostly Relative Value and Futures
strategies. Finally, the strategies with the lowest capacities are the illiquid Event Driven
strategies (Distressed Securities, Regulation D) and Equity Market Timing. This last
strategy is facing increasing problems in its operation due to the fact that Mutual fund
managers are less willing to accept market timers trading in their funds.

.... 151 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 152

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Risk management principles for the different AIS sectors

An understanding of AIS risk and risk management principles on a strategy-by-
strategy basis is important for three reasons. First, it provides the allocator with the
possibility to decide during the sector allocation process which risks he is willing to
bear for an appropriate expected return and to avoid the ones he is unwilling to
accept. Second, it allows the AIS allocator to critically review a manager’s own
internal risk controls as part of the manager due diligence process. Third, it enables
the allocator to monitor and control risk on the portfolio level more effectively
(see Chapter 7 for a more detailed discussion).
Active risk management generally includes measures such as establishing maxi-
mal position sizes, maintaining defined allocations to sectors and asset classes,
keeping a minimum number of different positions, and applying strict position
stop-loss limits. But the most important element of risk management on the level
of the individual strategy as well as the multi-manager portfolio is diversification.
While multi-manager funds reach diversification by investing with a variety of
strategies, individual AIS managers diversify by trading their particular strategy in
various instruments and asset classes. The following presents the most important
risk management guidelines for each strategy sector.

Convertible Arbitrage
Convertible Arbitrage is a ‘long volatility strategy’. The strategy is naturally exposed
to ‘vega’ risk, i.e. the risk of unexpected changes in volatility. This risk is generally not
hedged other than through sector and issuer diversification. But most managers seek
to implement a delta neutral hedge, i.e. they avoid exposure to price moves of the
stock. For this purpose the manager has to determine the appropriate number of
stocks to be sold short against the long convertible position (as determined by the
hedge ratio). Scenario analysis and stress tests are performed to determine how prices
and hedge ratio behave under different market scenarios. Besides delta hedges for the
equity component, interest rate derivatives (options, futures, swaps) are used to hedge
exposure to changes in interest and foreign exchange rates (if relevant). Recent devel-
opments in credit derivatives enable Convertible Arbitrage managers increasingly to

.... 152 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 153

5 ■ Risk in Alternative Investment Strategies

hedge credit risk through total return swaps credit default swaps and credit options.16
This still happens mostly on security baskets, as individual credit default protection
for below investment-grade securities remains costly. But as the credit derivatives mar-
kets are developing, single issuer protection might become more easily available in the
future. Convertible Arbitrage managers who build up an expertise in trading credit
derivatives can potentially build up a real edge in their sector.
Convertible Arbitrage managers have to perform in-depth, traditional funda-
mental analysis on the issuing company’s stock and debt. Credit analysis is
especially important in cases where credit quality is low. The focus of most man-
agers has traditionally been on financial ratios (earnings per share, ROE, Altman’s
Z-score, interest rate coverage, balance sheet leverage etc.). Recently, quantitative
credit risk models have evolved which can aid the risk manager in the analysis of
default risk (see Chapter 6 for a discussion of these models).
Diversification is an important component of Convertible Arbitrage risk man-
agement. The manager can diversify across industries, sectors, credit qualities and
moneyness of convertible securities (out-of-the-money convertibles are more
duration sensitive, while in-the-money convertibles bear higher equity sensitivity).
The manager should stress test the portfolio for sensitivities to interest rate risk
(rho risk), vega (volatility) risk, spread widening and credit default risk and equity
price moves. Setting stop-loss limits and maximum sizes for exposures to individ-
ual companies can further control risk.
Convertible Arbitrage managers have to deal with liquidity risk. The manager
has to be aware of funding risk, the exposure to short squeezes and possible
decreases of short rebates, especially in situations of financial market turmoil,
when spreads increase and liquidity dries up. ‘Flight to quality’ can become a seri-
ous issue at moments when the extension of credit, including the financing
provided by prime brokers, is rationed and investors avoid riskier assets in favour
of higher quality issues. A good relationship with the prime broker decreases these
financing and ‘shorting’ risks. Leverage control constitutes an important guideline
for risk management. The manager should always consider the following ques-
tion: ‘In extreme circumstances, can the position be unwound, and at what price?’

.... 153 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 154

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Fixed Income Arbitrage

Knowledge about the present and the possibility of future portfolio exposure and
risk (VaR and stress tests respectively) is extremely important for the Fixed
Income Arbitrage manager in order to be able to react to developments in finan-
cial markets in a timely fashion. The continuous analysis of the strategy’s risk
profile requires very sophisticated quantitative tools. ‘Variance-based VaR’, ‘multi-
ple standard deviation’ and ‘95% confidence interval’ approaches are insufficient
for assessing the market risk of a typical Fixed Income Arbitrage strategy. It is the
‘5% tail risk’ that requires the risk manager’s particular attention. Monte Carlo
simulation, extreme market stress tests (e.g. the scenario of the emerging
market/Hedge fund crisis of 1998, unexpected cuts/raises of interest rates by the
Fed or sovereign default), and Extreme Value Theory (a concept widely applied in
insurance mathematics) are examples of more appropriate risk analysis tools.
A Fixed Income Arbitrage manager relies strongly on his counter-parties as
lenders as well as borrowers. A thorough due diligence of their creditworthiness
as well as their reliability is necessary. A variety of portfolio credit risk tools have
been developed recently (see Chapter 6). The strategy can easily become a captive
of extreme leverage, where a single margin call on a position can destroy an entire
portfolio. The liquidity risk has to be controlled by ensuring sufficient funding
and maintaining appropriate leverage levels. Generally, managers try to eliminate
interest rate risk by structuring their portfolios to be near zero duration (except
when managers execute a particular view on the future shape of the yield curve).
Convexity risk is less often hedged.
Fixed Income Arbitrage managers often employ sophisticated quantitative
models to identify pricing anomalies. These models require constant reassessment
to ensure that the required level of accuracy is maintained. Accurate understand-
ing and valuation of prepayment options for mortage-backed securities (MBS) is
the key to appropriate risk management for MBS strategies. The main tool here is
an appropriate option-adjusted spread (OAS) model. OAS is the yield spread that
has to be added (to a binomial model or Monte Carlo simulation) in MBS valua-
tions accounting for the optionality of prepayment.17 Prepayment risk is usually
difficult to hedge.

.... 154 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 155

5 ■ Risk in Alternative Investment Strategies

Equity Market Neutral

The key to risk management for Equity Market Neutral strategies is diversification.
Exposure should be spread widely across industries and sectors. Managers have to
narrow down the number of stocks considered for trading to 200 to 1,000 out of
the tens of thousands of companies listed on international exchanges. For this, they
use criteria such as sectors, size, liquidity and market depth, ability to short and cor-
porate event risk. Statistical Arbitrage managers usually trade more liquid stocks,
because their turnover is generally high and positions need to be readily bought and
sold. Corporate event risk requires special attention. Most managers avoid stocks
involved in mergers or restructuring, buyback programs or heavy (legal) insider
trading (e.g. employee stock plans). Particularly ‘hard to borrow’ stocks are also
excluded from the list of eligible stocks. Important features of risk management are
stop-loss limits and maximum sizes for individual positions. As for Fixed Income
Arbitrage, the ‘5% tail risk’ calls for the particular attention of the risk manager,
who should also consider the correlation features of specific stocks and sectors. In
most circumstances variance-based VaR with 95% confidence intervals is insufficient
to assess the risk of a typical Equity Market Neutral strategy.
Once the ‘eligible stock list’ is defined, the manager begins the process of select-
ing the individual stocks to be included in his portfolio. The investment
decision-making process might require an in-depth fundamental analysis of compa-
nies. For Statistical Arbitrage strategies this often happens based on factor models.
Many Statistical Arbitrage managers have no qualitative component to their deci-
sion-making process, i.e. they use quantitative models exclusively. These rely mostly
on past statistical properties of single stocks and stock sectors. Most market neutral
Long/Short Equity managers, on the other hand, perform discretionary evaluations
or use a qualitative discretionary overlay on top of quantitative analysis. They
review the buy and sell lists to identify specific factors that models cannot cover,
such as pending mergers, late breaking news, rumours or simply faulty data.
Models and discretionary techniques do not work equally well in all sectors, so
managers have to select the sectors for which their valuation models and analysis
techniques work best. High-tech stocks require different valuation approaches from
low tech, utility or railway companies. Sophisticated quantitative models that are

.... 155 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 156

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

employed to rank stocks relative to one another and to identify equity price anomalies
require constant monitoring and reassessment to ensure their predictive accuracy.
Continuous structural changes of financial markets usually affect the predictive qual-
ity of the models and necessitate their frequent re-evaluation and dynamic refinement.
The manager has to determine when to drop certain factors of the model that have
lost their predictive power and when to include new ones. Out-of-sample testing
should always follow model optimization in order to limit ‘over-fitting’.
Portfolio construction is a very important component of return and risk man-
agement for Equity Market Neutral strategies. Managers often utilize sophisticated
computer algorithms, so called quadratic optimization procedures, for this pur-
pose. Such optimizers are widely commercially available (BARRA, Mathlab, etc.),
but some managers construct their own proprietary tool sets. Constant portfolio
monitoring is necessary to ensure neutrality against the exposures that strategy
seeks neutrality against (e.g. market beta, sector or industry, size, P/E levels etc.).
With a high position, turnover efficient and reliable execution and back office
facilities are necessary to cope with operational risks (e.g. execution risk). The rela-
tionship to the prime broker is important for ensuring continuous funding of
leverage, the availability of shorting capacity and protection against ‘short squeezes’.

Risk Arbitrage
Risk Arbitrage can be seen as equivalent to writing a put option on a merger deal.
The primary risk of the strategy is the deal risk arising from financial, legal or reg-
ulatory challenges to a merger going through. When a merger actually fails, the
long position usually drops significantly. In order to avoid a single deal failure
spelling disaster for the entire portfolio, diversification across a number of merg-
ers is the first principle for risk management in Merger Arbitrage. Many managers
use position limits on individual deals in the portfolio. Some managers spread risk
across different industry sectors, while others have particular expertise in one
industry sector. In this case, the investor should seek diversification within his AIS
portfolio through diversification across various managers with different sector
focuses. Concentrating on deals in the late stage of the merger transaction process
decreases deal risk, but also decreases the deal premium spread (expected return).

.... 156 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 157

5 ■ Risk in Alternative Investment Strategies

Many managers enter into positions related to announced deals only and do not
speculate on possible merger candidates prior to their announcement.
Risk Arbitrage managers have to perform in-depth analysis of all the issues
involved in the merger and constantly monitor the evolution of the merger process
in order to control the deal risk. This includes the financial situation of the two
companies (earnings, balance sheet leverage, growth etc.), regulatory issues, share-
holder resistance and the development of the deal premium (spread). Important
parts of the due diligence and risk control process are studies of the financial state-
ments, interviews with management, attending anti-trust court hearings, analysis of
market sentiment and stock market valuations, studying trade data sources and past
justice department and anti-trust agencies’ enforcement policies.
Most Risk Arbitrage managers use some leverage. Leverage control generally consti-
tutes an important guideline for risk management, especially in periods of decreased
deal flow. Taking equal size long and short positions usually controls exposure to the
broad equity market. In some cases the precise value of the merger transaction is
unknown or changes occur in the course of the transaction. In those cases, delta neutral-
ity should be monitored diligently. Liquidity is usually not a major risk factor, if the
involved companies have sufficient market capitalization. In the case of smaller deals the
manager should monitor whether the stocks are traded with sufficient liquidity. The
ability to sell the acquirer short can be dampened by the stock being ‘hard to borrow’.
A note on VaR: VaR techniques are difficult to apply to Risk Arbitrage strate-
gies, as the prices of the two stocks (especially the stock of the acquired company)
detach from their price history once a merger has been announced. The deal spe-
cific character of a contemplated transaction generally makes VaR-based
quantitative risk analysis less useful. Nevertheless, recent efforts by CMRA18 are
directed towards developing a ‘database that will allow investors to apply value-
at-risk to portfolios with investments in Merger Arbitrage Hedge funds’.

Distressed Securities
Distressed Securities strategies are exposed to a very unique set of risks, mostly com-
pany specific. Liquidity is inherently low, as restructuring negotiations and/or
bankruptcy proceedings often stretch over an extended period of time with uncertain

.... 157 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 158

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

outcome. Therefore, diversification across a broad range of companies is an important

part of risk management. Once invested, however, liquidity constraints impose limita-
tions on active risk management. Nevertheless, constant monitoring or involvement in
the ongoing restructuring process or bankruptcy proceedings is important.
Managers must engage in in-depth fundamental analysis of individual companies in
the pre-investment phase and should have extensive experience in assessing troubled
companies. Credit risk, the probability and default and the impact of an actual default
must be assessed in a ‘bottom up’ analysis. Quantitative credit risk models are of lim-
ited use, as most of these models are not specifically designed for distressed investments
and do not distinguish sensitively enough in the range of high default probability.
In order to value equity and debt securities of troubled companies, the analyst
often prepares both a ‘going concern’ valuation and a valuation based on liquidation
and then assigns probabilities to possible outcomes. In the case of a debt security,
this includes an assessment of the debt’s collateral and other terms. In addition to
looking at recent and projected financial results, other factors such as liquidity, com-
petence of management, credibility of restructuring plans and company strategy and
external market factors should be taken into account. Bankruptcy laws in relevant
jurisdictions and the impact of a bankruptcy must be assessed along with any other
relevant legal issues. The valuation of the specific securities also requires an assess-
ment of the ranking of debt or equity claims relative to other stakeholders as well as
an analysis of specific legal issues that may impact the enforceability of the claim.

Convertible Debenture Arbitrage (Regulation D)

Similar to Distressed Securities or Private Equity investments, Convertible
Debenture Arbitrage strategies are exposed to a unique set of risks, mostly specific
company and liquidity risks. Until SEC registration takes place, there is basically no
liquidity for the investment and no guarantee can be given that SEC registration
will eventually occur. This severely constrains the possibilities of pre-registration
active risk management once an investment is made; therefore, diversification
across a broad range of companies is critical to control risk.
Based on ‘bottom up’ analysis, the financial condition, the firm’s management com-
petence and its credit risk (the probability of default) must be assessed prior to
investment. The financial analysis should consist of a review of historical financial per-

.... 158 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 159

5 ■ Risk in Alternative Investment Strategies

formance, the company’s current pro forma income statements, Z-score and ratio
analysis (mostly focusing on current ratios and liquidity ratios) and an assessment of
the appropriate valuation. Recently developed credit risk models based on the ‘Merton
model’ (for more details about structural credit risk models, see Chapter 6) can pro-
vide for a quantitative assessment of the probability of default for the company.
Further firm and management evaluation should include due diligence on management
quality and experience, personal background checks, industry and competitive analysis,
supply capacity/supplier agreements, client/buyer analysis and performance trends of
the firm and its industry. Finally, ‘downside’ assessment of the impact of a possible
bankruptcy along with the relevant legal issues should be completed.
Once registration with the SEC is completed, the converted parts of the debenture
become freely tradeable. If the conversion price is floating, the manager is largely
hedged against a declining stock price. An analysis of post-registration risk should
include historical liquidity analysis (e.g. volume, bid–ask spreads), historical volatility,
short interest, share overhang, market makers and specialists who are actively trading
the stock and institutional and insider ownership and their recent activity. The man-
ager also has to assess the nature and investment approach of other possible investors
in order to evaluate the risk that the company enters into a ‘shorting death spiral’, i.e.
too many investors shorting the stock at the same time to hedge their exposure.

Global Macro
Global Macro managers invest very opportunistically in a variety of different
asset classes and instruments. It is difficult to describe general risk management
guidelines for this strategy, as the risk profile of Global Macro strategies gener-
ally extends over the wide range of AIS risks (market risk, liquidity risk, event
risk, corporate risk, foreign exchange, operational risk, credit risk, fraud risk,
regulation risks, legal risk, model risk). Knowledge about current and possible
future portfolio exposure and risk (VaR and stress testing respectively) provides
the manager with the ability to react in a timely fashion to undesired risk expo-
sure. The key risk management tool, however, is effective diversification across
asset classes and instruments. This includes exposure limits and risk buckets.
Many managers impose stop losses on individual positions. Leverage limits are
important for controlling risk.

.... 159 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 160

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Long/Short Equity
Long/Short Equity strategies face broad equity market risk to the extent of their net
market exposure. Determining the net exposure is one of the key issues of the
Long/Short strategy sector and requires solid macroeconomic research. The manager
can mitigate the broad market risk by monitoring the relevant macro factors in rela-
tionship to stock market valuations and adjusting the net long and short exposure
accordingly. In bull markets exposure is usually net long, but in bear markets net long
exposure should be avoided. Sector risk can be reduced or eliminated by trading pairs,
i.e. the manager takes a long position in a company with a favourable outlook in a cer-
tain industry and sells short a stock with less favourable outlooks in the same industry.
These two positions offset each other in case the entire sector experiences a downturn.
Most managers predefine a certain set of stocks to pick their positions from.
Often ‘liquidity filters’ are employed which take the form of required minimum
free float market capitalization or trading volume, before actual fundamental
research is performed. Stocks with large capitalization have the advantage of
being more liquid. But at the same time more analysts follow large cap stocks, i.e.
the manager is less likely to have any informational advantage. Small cap stocks
usually involve more risk, as they are unknown and have unproven earning
streams. Generally, the choice of a stock universe and an ‘investment theme’ has
great influence on the risk profile of the particular strategy.
Picking a stock for a long or short position means taking on specific company
risks such as unexpected change of profit margins and earnings (or earnings out-
looks), unfavourable industry trends, discontinuity of management, emergence of
new technologies, unfavourable international trends, loss (gain) of market share,
changing commodity prices, mergers and others. Diversification across positions
in various sectors is of key importance and ensures that the manager’s portfolio is
not too strongly affected by events related to individual companies or industries.
The construction of an efficient individual portfolio requires that careful attention
be paid to correlations between stocks.
A key tool for managing the company-specific risk is thorough fundamental
research that allows the manager to make predictions about the company’s future earn-
ings and stock price based on all currently available information. Important research

.... 160 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 161

5 ■ Risk in Alternative Investment Strategies

sources are journals and newspapers, data providers (Bloomberg, Reuters etc.), indus-
try contacts and conference calls with management, company visits, reports provided
by brokerage houses and the internet. Technical analysis can be employed to determine
the price momentum of the specific stock or to support a fundamental view.
For the individual long and short positions, stop-loss limits and profit targets are
important risk management measures. They enforce discipline in situations where
the stock does not behave as expected or profit targets have been reached. Further,
as most managers employ leverage in order to increase their return potential, this
leverage has to be controlled. Finally, Long/Short Equity strategies are exposed to
the risk connected to short selling. A good relationship with the prime broker and a
solid short selling experience base are necessary to minimize common ‘short selling
risks’ (e.g. ‘short squeeze’, ability to borrow stock).

Equity Market Timing

Equity Market Timing strategies are among the least complex strategies in the AIS
universe. Usually, rather simple indicators determine whether to enter or exit a posi-
tion. Standard VaR or volatility-based approaches can be used for the monitoring of
market risk. Diversification across various equity sectors and industries is important
for risk reduction. A risk management measure used by some Equity Market Timing
managers is the adjustment of position size according to current market or sector
volatility. Studies by LeBaron19 have shown that trend persistence in equity markets
is conditionally correlated to volatility. Trends tend to be more persistent in periods
of lower volatility but are more likely to revert at times of high volatility.
The manager has to be certain that the selected instruments, mostly mutual
funds or stock basket, have the desired exposure to the market or sector. This is
not always guaranteed, as most mutual funds do not frequently disclose their cur-
rent holdings. Furthermore, most Mutual fund managers are resistant to allowing
market timers to use their funds. The manager must ensure that he can get into
the fund at the time he wants to with no time delay. This requires a good relation-
ship with the Mutual fund firms. Because many Equity Market Timing strategies
switch positions frequently, the manager has to guarantee reliable trade execution.

.... 161 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 162

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Short Selling
Short Selling strategies are exposed to the same risk factors as ‘long only’ invest-
ment strategies and require similar risk management approaches with respect to
broad market risk and specific stock risk/corporate event risks. Sound macroeco-
nomic and sector-specific research and fundamental analysis of individual stocks
are important elements of risk management. This has to go along with diversifica-
tion across different stocks and sectors.
In addition to the general equity investment risks, there are particular execu-
tion and liquidity issues associated with Short Selling. Short Selling strategies are
subject to a unique set of risks related to the fact that managers have to borrow
stocks before selling them (‘borrowing risk’). Borrowing risk can be greatly
reduced through the appropriate choice of and a strong relationship with a prime
broker. Important parameters are the prime broker’s capability to lend stock, his
buy-in policy and his funding practices. A good prime broker is proactive in secur-
ing stocks to loan out to the short seller and communicates changing stability of
the borrowing process quickly enough for the short seller to react.
Short Selling strategies face a particular form of portfolio risk through an inverse
relationship between performance and exposure. A loss in a short position leads to a
larger overall portfolio exposure to the losing position. Short sellers have to manage
this dynamic risk vigilantly. Rigorous adherence to maximum position size limits is
one way to mitigate this risk. Stop losses are often used to control the risk in indi-
vidual stocks. However, especially in volatile periods, stop losses can have an
undesired side-effect. Other market participants who know about the stop loss can
artificially induce the necessary short-term price move to trigger them.

Managed Futures
The risks of Managed Futures strategies (Active Discretionary, Systematic Active,
Systematic Passive) and the appropriate approaches to risk management are gener-
ally quite similar, so they all will be treated in the same subsection. The number
one risk management principle for a Managed Futures portfolio is diversification.
Most active as well as passive strategies invest in a large number of different
Futures markets. This ranges from trading the most liquid 20–25 Futures con-

.... 162 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 163

5 ■ Risk in Alternative Investment Strategies

tracts to investing in more than 100 markets across commodities, equities, fixed
income and foreign exchange. Note that there is a trade-off between diversifica-
tion and liquidity. A large universe of contracts includes those which are less
liquid. Unwinding larger positions in these markets might become difficult, espe-
cially at times of high market volatility. Liquidity risk and the relevant liquidity
indicators such as transaction volume and bid–ask spreads should in this case be
constantly monitored. Another useful measure for risk control is volatility-
dependent allocation, where the weight of the different contracts in the portfolio
is determined by their historical (or implied) volatility.
Quantitative analysis such as VaR and stress tests are important risk monitoring
tools for Futures strategies. Besides portfolio VaR, the large number of positions
in different asset classes necessitates the consideration of VaR on various aggrega-
tion levels (e.g. commodities only) to detect undesired risk concentrations. Some
managers apply limits on the notional exposure or risk to a single market/risk
factor (this is referred to as ‘risk budgeting’). Risk budgets can be based on VaR or
stress test losses. Limits on employed leverage (margin to equity ratio), stop losses
and profit targets constitute further useful risk control measures for Managed
Futures strategies. For contracts traded in different currencies, FX futures and
options can be used to hedge unwanted currency risk (however, currency expo-
sure remains unhedged in most strategies).
Systematic Managed Futures strategies rely on models that are the results of
past research efforts. These might be flawed or unsuited for current market struc-
tures. Generally, model development should be guided by four principles:

1 The more complex the underlying model, the higher the risk of ‘over-fitting’
(‘curve fitting’). The parameter optimization of models should be limited to
only a few parameters, which, if feasible, are chosen identically for each
traded market.
2 Choosing exactly those parameters which give the best performance without
consideration of the performance for nearby parameters is another recipe for
over-fitting. The optimal parameters should be located on a ‘plateau’ in
parameter space rather than on an isolated ‘performance peak’.

.... 163 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 164

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

3 A strict distinction between in-sample and out-of-sample data is essential.

Performance for optimization periods (in sample) and for model testing
periods (out of sample) should be compared and found not to be too different
from each other.
4 The target function of the optimization should be selected carefully. Risk-
adjusted performance measures are generally better suited than gross returns.20

Fully systematic trading approaches are widespread, especially among Managed

Futures strategies. But it is important to note that these principles of model opti-
mization and addressing model risk equally apply to model-based strategies in
other AIS sectors.
By way of contrast, non-model-based strategies (Active Discretionary) often rely
on the skill of one or several key persons. Appropriate backup should be available
should one of these people become unavailable. While some mostly Passive Futures
strategies have rather low turnover, there are strategies with hundreds of trades per
day. This requires efficient in-house execution and back-office facilities.

Portfolio diversification
Diversification is the most rudimentary, but at the same time one of the key ele-
ments of portfolio construction and risk management in finance. Modern portfolio
theory21 (MPT) states that correlation properties are the key to efficient diversifica-
tion in an investment portfolio. In the investment practice the calculation of
efficient frontiers is the most important application of MPT. Numerous studies22
(see also Chapter 4) have shown that due to low correlations to traditional assets,
AIS investments offer attractive possibilities for diversification and efficiency
enhancement in an investor’s global portfolio. At the same time, the performance
and correlation properties of AIS differ widely across the strategy sectors.23
A note of caution is necessary upfront: quantitative optimization for AIS port-
folios can be quite useful in determining the scope for improvement from adding
AIS to a global traditional balanced (equity and bond) portfolio, however, it
shows its limitations if used for determining the right AIS sector allocation within
a multi-manager AIS portfolio. The results of Markowitz-type optimizations are

.... 164 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 165

5 ■ Risk in Alternative Investment Strategies

very sensitive to the return and risk assumption used in the calculation. The avail-
ability and reliability of AIS performance data leaves much to be desired, as
discussed in detail in Chapters 4 and 7. Data is available only for a limited time
frame and is biased due to the specific market circumstances in the 1990s (bull
equity markets). The performance of the different strategy sectors is very hetero-
geneous and generally far from stationary, i.e. the mean return, standard deviation
and the correlations, all important inputs to portfolio optimization programs, vary
substantially with respect to chosen time periods, market environments and indi-
vidual managers. Taking average performance numbers over a time period of a
few years as representative for the strategy sectors in the portfolio optimization
does not account sufficiently for this variability. Performance and correlation
properties between the different strategies remain an important consideration in
setting up an AIS portfolio, but experienced AIS portfolio managers do not apply
quantitative portfolio optimization on a sole basis, rather they combine it with
qualitative considerations.
The diversification of equity risk in the traditional portfolio is one of the main
motivations for investing in AIS. But for this very purpose, certain combinations
of strategies are better suited than others. For example, Relative Value and
Managed Futures strategies show generally lower correlations to equity markets
than most Opportunistic strategies. Chapter 4 introduced the concept of condi-
tional correlations, which provides a powerful tool for examining the
effectiveness of portfolio diversification.
Figure 5.11 shows an example of two different sets of strategies (encircled) and
their past correlation properties conditional to equity market directions (compare
with Figures 4.4 and 4.5). The figure illustrates clearly that the second set pro-
vides better diversification in periods of negative equity performance than the
first. The first portfolio consisting of Long/Short Equity, Global Macro, Event
Driven and Distressed Securities is biased towards an environment of rising equity
markets and shows weaknesses during periods of negative equity market perform-
ance. In contrast, the combination of Managed Futures, Equity Market Timing
and Relative Value strategies (Convertible Arbitrage, Fixed Income Arbitrage,
Equity Market Neutral) shows strengths in various equity market environments.
In periods of falling equity markets, Managed Futures tend to perform well24 and

.... 165 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 166

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

1.0

0.8

0.6 Equity Market Timer

Correlation to S&P500 in good months

Long/Short
0.4
Equity

Futures Equity Market

0.2 Global Event
Systematic Neutral
Active Macro Driven
Reg D
0.0 Currencies
–1.0 –0.8 –0.6 –0.4 –0.2 0.0 0.2 0.4 0.6 0.8 1.0
Commodity
Futures Long –0.2
Passive Relative Distressed
Futures Value Securities
Discretionary
–0.4
Short
Seller
–0.6

–0.8

–1.0
Correlation to S&P500 in bad months
(a)

1.0

0.8

0.6 Equity Market Timer

Correlation to S&P500 in good months

0.4
Long/Short
Equity
Futures 0.2 Equity Market
Global Event
Systematic Neutral
Macro Driven
Active Reg D
0.0 Currencies
–1.0 –0.8 –0.6 –0.4 –0.2 0.0 0.2 0.4 0.6 0.8 1.0
Commodity
Futures Relative
Long–0.2
Passive Value
Futures Distressed
Discretionary Securities
–0.4
Short
Seller
–0.6

–0.8

–1.0
Correlation to S&P500 in bad months
(b)

FIGURE 5.11 ■ Testing diversification with conditional correlation (compare with Figures 4.4 and 4.5)
(a) Example for a portfolio with strong equity correlations in falling markets
(b) Example for a diversified portfolio

.... 166 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 167

5 ■ Risk in Alternative Investment Strategies

Equity Market Timing strategies are almost uncorrelated to equity markets. In

periods of rising equity markets, Equity Market timers show strong performance,
while Managed Futures are uncorrelated. Relative Value strategies generally dis-
play low correlation to equity market directions in ‘normal periods’ but tend to be
highly correlated in times of market turbulence (not shown in Figure 5.11).
Another example is the following: A combination of Convertible Arbitrage and
Risk Arbitrage can provide favourable correlation properties conditional to equity
market volatility. Convertible Arbitrage strategies show above average performance
during periods of increasing equity market volatility, when Risk Arbitrage strategies
struggle. Low volatility periods bear inverse performance behaviour. Risk Arbitrage
is attractive, Convertible Arbitrage less so.
Another important analysis refers to the returns and correlations conditional
on different bond market performance. Generally, a systematic study of condi-
tional returns and correlations in various market environments (e.g. with respect
to equities, bonds, FX) can lead the AIS investor to similar insights as multi-factor
models of returns.
Statistical analysis and correlation studies are only one way to approach the
subject of AIS portfolio diversification. Another approach, which aims at supple-
menting the results of statistical studies on past performance data, is based on
fundamental economic reasoning and looks more closely at the economic reasons
and sources of AIS returns. Understanding the return generation process for each
strategy can help to reach efficient portfolio diversification. As discussed in
Chapters 2 and 3, individual strategies can have very different performance driv-
ers, as they represent different economic functions in financial markets. The
following list builds on the discussion in Chapter 2 and summarizes the classifica-
tion of the AIS in terms of economic functions:

■ Capital formation (providing capital to companies): Equity Market Timing,

Long/Short Equity. The primary purpose of equity markets is the formation
and efficient distribution of capital needed by companies to foster innovation
and economic progress. Investors provide the needed capital and ensure its
efficient allocation to the most promising businesses.

.... 167 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 168

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Risk transfer (taking risk from commercial hedgers): Managed Futures. The
investor provides commercial hedgers with the necessary liquidity to perform
their hedging activities. Similar to insurance companies, they are compensated
with a risk premium.
■ Relative Value Arbitrage (making markets more efficient): Convertible
Arbitrage, Risk Arbitrage, Fixed Income Arbitrage and Equity Market Neutral.
Arbitrageurs25 take advantage of relative mispricings or differences in credit,
liquidity or other risk factors of different but generally closely related
securities. This provides financial markets with a higher degree of efficiency
and a less erratic price formation process.
■ Providing liquidity (taking positions in less liquid markets and instruments):
Distressed Securities, Regulation D, some Fixed Income Arbitrage. Managers
invest in less liquid instruments and earn a liquidity premium, thus providing
a market for instruments less frequently traded.
■ The economic function for Global Macro, Short Selling and some Long/Short
Equity strategies is less clear. The returns of these are based on the manager’s
skills in forecasting stock price developments and acting quickly rather than
earning particular risk premiums. Usually, these strategies involve a higher
degree of speculation.
Low statistical correlations to equity and other markets as well as between differ-
ent AIS sectors usually find their theoretical explanation in different economic
functions of the corresponding strategies. So for the example just discussed: the
set of Equity Market Timing (economic function: capital formation), Futures (eco-
nomic function: risk transfer) and Relative Value (economic function: market
efficiency) contains strategies with very different return generation processes.
Next to statistical correlation features, the AIS allocator should consider the eco-
nomic sources of returns for the purpose of AIS portfolio diversification.

Notes
1. See the article by L. Rahl, ‘Risk Budgeting: The Next Step of the Risk
Management Journey – A Veteran’s Perspective’, in L. Rahl (ed.) Risk Budgeting: A
New Approach to Investing.

.... 168 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 169

5 ■ Risk in Alternative Investment Strategies

2. See also the study ‘Hedge Funds and Financial Markets Dynamics’, by
B. Eichengreen and D. Mathieson.
3. This risk is sometimes classified as operational risk, but in our context it
should be seen as a risk type on its own.
4. See also report issued by a group of five Hedge fund managers, ‘Sound Practices
for Hedge Fund Managers’, section ‘Recommendations: Risk Monitoring’.
5. Personal communication with M. Perkins, MKP. See also ‘Prime Brokers Can
Make and Break Hedge Funds’ in Pension & Investments, September 3, 2001.
6. For more recent developments in the relationship between Hedge funds and
their prime brokerage counter-parties, see the article by Patel and Navroz,
‘Courting the Hedge Funds’ in Risk Magazine, November 2001.
7. See the article ‘Hedge Fund Disasters: Avoiding the Next Catastrophe’ by
D. Kramer in Alternative Investment Quarterly, October 2001.
8. See Chris Clair, ‘Hedge Funds Suffer Without a Benchmark’, Pension &
Investments, June 11, 2001.
9. See also the work by W. Fung and D. Hsieh, ‘Benchmarks of Hedge Fund
Performance: Information Content and Measurement Biases’.
10. Dacorogna et al. in ‘Effective Return, Risk Aversion and Drawdowns’ in
Physica A, January 2001, provide an excellent presentation of different
performance measurements.
11. See the articles ‘Life at Sharpe’s end’ and ‘Measure for Measure’ by H. Till in
Risk and Reward, September and October 2001.
12. A study ‘Do Hedge Funds Hedge?’ by C. Asness et al. examines the effect of these
liquidity-induced lags on correlation and performance properties of Hedge funds.
13. Capital Market Risk Advisors, ‘NAV / Fair Value Practices Survey Results’,
July 2001, available on http://www.cmra.com.
14. See B. Liang, ‘Hedge Funds: On the Performance of Hedge Funds’ in
Financial Analysts Journal, July 1999.
15. See the discussion in ‘TSS(II)-Tactical Style Selection: Integrating Hedge Funds
into the Asset Allocation Framework’ by Crossborder Capital, published by Hedge
Fund Research, August 2000.

.... 169 ....

8409 Chapter 5 p136-170 11/4/02 1:14 PM Page 170

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

16. See the article ‘Convertible Arb Funds turn to Default Swaps’ by C. Schenk in
Risk Magazine, July 2001.
17. See F. Fabozzi in Fixed Income Analysis (Chapter 11, 2) for more details on
OAS models.
18. CMRA stands for ‘Capital Markets Risk Advisors’; see article by G. Polyn,
‘Value-at-Risk for Merger Arbs’ in Risk Magazine, April 2001, p.6.
19. See the publications by B. LeBaron, ‘Forecast Improvements using a Volatility
Index’ and ‘Some Relation between Volatility and Serial Correlations in Stock
Market Returns’.
20. See the work by Pictet et al., ‘Real Time Trading Models for Foreign Exchange
Rates’, where possible target functions for model parameter optimization are discussed.
21. Modern Portfolio Theory (MPT) is a topic in almost all finance textbooks and
university classes. For an introduction into MPT refer to Investment Analysis and
Portfolio Management by F. Reilly and K. Brown.
22. See the article ‘The Benefits of Hedge Funds’ by T. Schneeweiss and
references therein.
23. For a discussion of diversification within a multi-strategy portfolio, see J. Park
and J. Strum, ‘Fund of Fund Diversification: How Much is Enough?, The Journal
of Alternative Investments, Winter 1999. See also R. McFall Lamm, ‘Portfolios of
Alternative Assets: Why not 100% Hedge Funds?’, The Journal of Investing,
Winter 1999, pp.87–97.
24. This statement applies when interest rates remain constant or fall. CTA
actually struggles when interest rates rise. See the discussion in Chapter 4 and also
the work by M. Caglayan and F. Edwards, ‘Hedge Fund and Commodity Fund
Investment Styles in Bull and Bear Markets’. The authors of this article found that
CTAs have higher returns than Hedge funds in bear markets and generally have a
negative correlation with equity returns in periods of falling markets.
25. I am not referring to arbitrage in the strictest meaning of the word here
(which is ‘generating a profit without risk’): With ‘Relative Value Arbitrage’ I
generally mean ‘buying relatively undervalued securities and selling overvalued
securities’. There is a risk involved, specifically, the risk that the undervalued
securities become even cheaper and the overvalued ones more expensive.

.... 170 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 171

CHAPTER 6

Tools and Principles of Modern

Financial Risk Management

Although there is no unique and widely accepted definition of risk in financial

markets, on a very general level, risk is the result of holding an investment (expo-
sure) combined with the uncertainty about future price developments (with risk
factor variability). Investment decision makers can control exposure, while uncer-
tainty is by nature uncontrollable. Risk measurement starts with the analysis of
these two components. Exposure analysis is the categorization of investment posi-
tions by risk factors, and uncertainty is often described by statistical estimates of
future price variability, i.e. volatility (as measured, for example, by the standard
deviation of historical returns).
It is acknowledged among risk professionals that the widespread perception of
risk and volatility as equivalents leads to a very incomplete picture of financial
risk. Statistical measures of volatility most often do not capture all the risks an
investor is facing (e.g. extreme market moves). Therefore, the emergence of vari-
ous risk quantification tools (such as VaR) notwithstanding, qualitative
considerations remain important for the complete assessment of risk, as some risk
factors defy accurate quantification. This is particularly true for AIS, where the
wide range of investment instruments and trading techniques exposes the investor
to more numerous and more complex risks compared to traditional investments
(see Chapter 5) and therefore makes risk quantification all the more challenging.

.... 171 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 172

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The foundation of effective risk management is a general risk policy outlining

what types of risk are acceptable and which are to be avoided. The policy must be
supported by appropriate tools that allow for monitoring and control and, impor-
tantly, by the actions and attitudes of key personnel. In the end, risk management
must be more than just risk quantification or risk measurement; it must involve spe-
cific actions, which directly influence the investment decision-making process and
which ensure that corrective action is taken when risk exceeds permitted levels.
This chapter aims at introducing the reader to the state-of-the-art financial risk
management tools and principles, including a discussion of their strengths and
shortcomings. It cannot claim complete coverage of this extensive subject.
Numerous articles, books and encyclopaedias have been published over the years1
and it is almost impossible to cover them all. The interested reader is advised to
read the references provided. They are, naturally, selective.2

The emergence of quantitative tools for risk management

During the last decade the role, importance and spectrum of tools for risk analysis
and risk management have evolved dramatically. The fallout from numerous finan-
cial fiascos, combined with the emergence of powerful quantitative models and
strong computing capacities, have led to what the financial profession regards as a
‘risk management revolution’. Today, the theory of risk management is considered
as a distinct field of its own within the world of finance.
The emergence of the ‘Value-at-Risk (VaR)’ changed the scope and character of
applied financial risk management. Developed initially as a concept for market
risk, it is now used for a variety of risk types (e.g. credit risk, liquidity risk, opera-
tional risk as well as firm-wide risk). Although it is difficult to pinpoint the precise
origin of VaR, it is undisputed that two events helped to bring this approach to
the attention of investment professionals globally. One is the Group of Thirty
Report in July 19933 and the subsequent use of VaR as a measure for market risk
by financial market risk managers and regulators (that ultimately defined the 1996
amendments, in particular the ‘internal models approach’, to the Basle accord for
the minimum capital requirements by internationally active commercial banks),
and the other is the introduction of RiskMetrics by J.P. Morgan in October 1994.

.... 172 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 173

6 ■ Tools and Principles of Modern Financial Risk Management

The emergence of VaR has led to new problems, however. Most VaR calculations
rely on certain assumptions which, at best, only approximate reality. Theoretical
discussions about VaR are ongoing, but it is widely accepted that VaR can actually
be quite misleading unless used with care. Much intellectual energy is currently
being dedicated to the development of more robust risk analysis tools than VaR.
On the application side, risk managers can today choose from numerous com-
mercial software tools with varying degrees of sophistication. The large, and, for
some, overwhelming, variety of available tools is a blessing and a curse at the
same time. Too often the key decision makers in financial institutions receive risk
information which is either too complex or too voluminous to allow for efficient
decision making and control. An important part of risk management is the presen-
tation of risk information in a manner such that the key decision takers can
understand and act on the presented information.

Pre-VaR risk management: Traditional risk measures

and ‘modern’ portfolio theory
On the most basic level, risk management consists of breaking down a portfolio by
holdings and analyzing risk on an instrument-by-instrument basis. Then exposures on
different aggregation levels (e.g. instrument, asset class, region, sector) are measured
and risk levels are quantified. Traditionally, volatility describing past fluctuations and
future uncertainty of security prices (usually quantified by the standard deviation of
returns) is the most important measure of risk. A further risk measure often used for
interest rate-sensitive portfolios (bonds) is duration (defined as the weighted average
time to maturity of cash flows).4 Such exposure analysis and volatility/duration meas-
urement on distinct aggregation levels is still widely used today.
‘Modern’ portfolio theory (MPT) provided a new approach to portfolio risk
analysis (‘modern’ is not quite accurate; the origins of MPT go back to the works
of H. Markowitz in the early 1950s, and W. Sharpe in the early 1960s). MPT does
not look at the volatility of individual positions, but considers the extent to which
individual holdings contribute to overall portfolio volatility. While a complete dis-
cussion of MPT is beyond the scope of this book, this topic is extensively covered
in most financial textbooks.5

.... 173 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 174

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

While not immediately resulting in a conceptual change of how risk is measured

(MPT still assumes risk can be completely described by the standard deviation of
returns), MPT set the stage for risk management on the portfolio level (and specifi-
cally the development of VaR) by introducing the concept of ‘correlation’. MPT
introduced the idea that highly volatile instruments may add no risk to a portfolio, if
they correlate negatively with the rest of the portfolio (i.e. if they are added to the
portfolio, the resulting portfolio volatility remains equal or decreases). For the calcu-
lation of correlations (and volatilities) the literature offers a rich universe of methods.6

Value-at-Risk (VaR)
VaR is today the most widely used quantitative analysis tool for financial risk. It
describes the maximal portfolio loss for a given confidence level over a specified
trading horizon. Mathematically speaking, VaR characterizes the tail (extreme
quantile) of the portfolio return distribution.7 Despite a number of pitfalls (see
later), VaR has introduced a new dimension of risk analysis into the financial com-
munity. The magic of VaR is that it introduces a uniform measuring system for the
various instruments in a portfolio by providing a method for comparing risk
across security and asset classes (prior to VaR, an investor was unable to quantify
the risk of a combination of a $1 million position in US equity and a $1 million
position in short duration German Government Bonds). The two important
parameters that a risk manager has to define for VAR are the time period and the
confidence level (e.g. 99% for a one-day horizon). VaR can be calculated on a
variety of different aggregation levels within the portfolio (fund-wide risk attribu-
tion, asset classes, sectors, instruments, trading managers, geographic regions).
In most respects VaR is a natural progression from MPT, but there are also
important differences:

■ MPT interprets risk in terms of the standard deviation of returns, while VaR
describes risk as the maximum likely loss.
■ The variance–covariance approach of MPT is just one of several methods to
calculate VaR (others are the Monte Carlo method and the historical method).

.... 174 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 175

6 ■ Tools and Principles of Modern Financial Risk Management

■ VaR can be applied to a range of different risks beyond market risk, while
MPT is limited to market risk. The concept of VaR can even be used for firm-
wide risk management.
■ VaR offers a wider frame for the incorporation of more complex statistical
methods such as non-normal returns.

While VaR is conceptually and intuitively quite simple, the technical details of its
calculation can be rather involved depending on the heterogeneity of the portfolio
and the distributional assumptions made. Detailing the technical aspects of portfo-
lio VaR calculation is beyond the scope of this book8 but a short overview is
provided in the following.
The three main elements of standard VaR calculation are:

■ mapping the portfolio positions to risk factors

■ calculating the risk factor covariance matrix
■ determining the VaR method and calculating VaR.

Mapping risk factors

Risk factor mapping is the decomposition of individual securities into components
over which the risk manager has control (the exposure) and exogenous influences
he cannot control (the risk factors). A risk factor is a variable which affects
directly the future value of a security (e.g. interest rates, equity market valuations,
FX rates etc.). The prices of thousands of worldwide available financial securities
are influenced by common risk factors.
The dependency of a security on risk factors (the security’s ‘sensitivities’) is
expressed by a ‘pricing function’ which can depend on the VaR method employed (see
below). Note that the concept of risk factors and sensitivities is not new. Sensitivities
have been used for the analysis of yield curve risks as well as in option theory for
many years (examples are the ‘value of a basis point move’ or the ‘Greeks’).
VaR calculations should account for every important risk factor in the portfo-
lio, which can be quite numerous (>1,000). The following provides a broad
categorization.9

.... 175 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 176

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Equity indices: The risk of an equity position can be separated into two
components, the sensitivity to one or several relevant market/sector indices
and an idiosyncratic residual risk. With a sufficient number of stocks in the
portfolio, the latter is diversified away. Commonly employed equity risk
factors are broad market or specific sector risks, expressed as ‘sensitivities’, or
in the language of the ‘Capital Asset Pricing Model’, ‘betas’.
■ Yield and credit spread curves: Fixed income risk arises from potential moves in
interest rate term structures (yield curves) and credit ratings. Each currency has its
own yield curve. Credit risk is expressed in terms of the spread to a ‘risk-free’ (i.e.
not defaultable) reference bond or probability of default. A portfolio exposure to
yield and spread curve risk can be described by certain representative maturity
points on the yield and credit spread curves (e.g. one-month, six-month, one-year,
two-year, five-year, ten-year, 15-year and 30-year maturities).10
■ Foreign exchange rates: Foreign exchange spot rates are risk factors in portfolios
that hold positions in currencies different from the investor’s home currency.
For positions in FX forwards and futures, the two underlying yield curves define
additional risk factors. The interest rate parity theorem provides the exact
relationship between the spot rates, the forward rates and the two interest rates.
■ Commodities: Spot commodity positions are exposed to single risk factors, the
respective commodity prices. Similar to yield curves defining risk factors
related to interest rates, commodity Forward and Futures contracts define a
term structure for each commodity which is determined by the costs of carry
(mostly related to interest rates) and the ‘convenience yield’.

One of the most common approaches to mapping positions to risk factors is via the
use of a security’s cash flow. This method assumes that all instruments can be repre-
sented by defined cash flows to the investor at present or specified times in the future
(note that this assumption is invalid for options). The cash flows are attributed to spe-
cific risk factors and discounted to present value at the prevailing interest rates. A
simple example is given by the cash flows of a (default free) bond with yearly coupon
payments and principal payback in ten years. The risk factors are represented by the
interest rates at various maturities on the yield curve. The present value of the cash
flow to the investor in a given year (coupon payments in years one to ten and

.... 176 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 177

6 ■ Tools and Principles of Modern Financial Risk Management

principal payback in year ten) is uniquely determined by the x-year interest rate. Cash
flows that fall in between two maturity points are split between the two nearby
points. A second commonly used approach is the sensitivity approach, which is based
on directly determining the portfolio’s sensitivity to changes in each risk factor.11

Calculation of covariance matrix

The next step in the calculation of VaR consists of estimating volatilities and correla-
tions of the risk factors. Making the standard assumption that security returns are
multivariate normally distributed, the correlation matrix (together with the vector of
expected returns) uniquely determines the distribution of portfolio returns, and
therefore portfolio VaR. There are various methods for calculating volatility (vari-
ances) and correlations (covariances). The following approaches are most common
(the RiskMetrics Technical document (1996) provides a discussion of the details):12

■ Equally weighted moving average of squared returns (variance) and cross-

products of returns (covariance).
■ Exponential moving average of squared returns and cross-products of returns
with specified decay parameter. This is the method chosen by RiskMetrics and
many others. The decay parameter most often chosen is 0.94 for daily
observations and 0.97 for monthly observations.
■ The family of ARCH models: ARCH (autoregressive conditional
heteroskedastic) and GARCH (General ARCH) models were developed in the
early 1980s and have become a starting point for sophisticated volatility and
correlation forecasting models.13 The exponential moving average model can
be seen as a special case of a GARCH model.

Calculation of VaR
After attributing cash flows to risk factors and calculating the risk factors’ covari-
ance matrix, VaR can be calculated using several methods which differ mainly in
respect to two factors.

1 Assumptions regarding security valuation as a function of risk factors: A

determination must be made as to how to model the sensitivity of security

.... 177 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 178

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

prices to changes in risk factors. The industry distinguishes between local

valuation and full valuation methods.
2 The distributional assumptions made: The industry distinguishes between
parametric (mostly normal) distributions versus historical distributions.
The three most popular VaR methods are:
■ Parametric approach: The idea behind parametric VAR models is to approximate
the pricing function of each instrument (i.e. the relationship between each
instrument and the risk factors), in a way that an analytical formula for VaR can
be obtained. The simplest parametric approach is the ‘delta method’, also referred
to as ‘variance–covariance-based VaR’.14 Assuming normal return distributions and
a linear relationship between risk factors and the securities in the portfolio, the
returns of the portfolio themselves are conditionally normally distributed (here
‘conditional’ refers to the standard deviation of this distribution changing over
time depending on the most recent volatility). The covariance matrix of risk
factors determines the standard deviation of portfolio returns. The method
possesses the charm of simplicity, but is severely flawed in circumstances where
the assumption of linearity does not hold. An enhancement of the delta method is
the ‘delta–gamma’ method, where second-order approximations are added to the
security pricing function. Both the delta and delta–gamma methods are ‘local’
valuation methods, as only the mark to market values of the portfolio positions
enter into the valuation of the portfolio (and no other price scenarios).
■ Monte Carlo simulations:15 The Monte Carlo method is a ‘full valuation’
method based on simulating the behaviour of the underlying risk factors
through a large number of draws produced by a random generator. Using
given pricing functions, the values of the portfolio positions are calculated
from the simulated values of risk factors. A possible non-linear relationship
between security and risk factors is fully accounted for. The positions in the
portfolio are fully revalued under each of the random scenarios. Every
random draw of risk factor values thus leads to a new portfolio valuation. A
high number of iterations (several thousand) provides a simulated return
distribution of the portfolio, from which the VaR value can be determined.
(The number of iterations necessary can be somewhat reduced by

.... 178 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 179

6 ■ Tools and Principles of Modern Financial Risk Management

‘deterministic sampling’, i.e. pseudo-random selection of values to ensure that

the full spectrum of relevant scenarios is selected.) The underlying distribution
of the randomly generated risk factor values can theoretically be chosen freely,
but in most cases the simulation draws from a normal distribution. The
random values are transformed such that the empirical correlations between
the risk factors are recognized (mathematically, this is achieved through a
‘Cholesky decomposition’ of the correlation matrix). The use of non-normal
distributions can quickly lead to complex mathematical problems.16
■ Historical simulation: Instead of simulating return distributions, the distribution
is determined by looking into the past. The historical method is also a full
valuation method and relies on the (unconditional) historical distribution of
returns by applying past asset returns to the present holdings in the portfolio. The
values of the portfolio positions are fully evaluated for each historical return by
the use of the specified pricing functions. This method has the advantage that no
explicit assumptions about the underlying return distribution have to be made.
The problem of the method is that it relies on historical price behaviour, which
might no longer be relevant in current market environments.

It was mentioned that the full security re-valuation employed by the Monte Carlo as
well as the historical method requires specific valuation models or pricing functions.
These models have to be correctly adjusted, which happens by calibrating the model
parameters to the market prices of specific benchmark instruments. Further, as long as
normally distributed draws are used, the Monte Carlo method does not address the
issue of non-normally distributed asset returns. The dependence and correlation struc-
ture of non-normal multivariate distributions is still the subject of intense research.
Stretching out beyond the normal distribution very quickly leads to intractable mathe-
matical problems, as non-normal multivariate distributions are much more difficult to
deal with (see the section on Extreme Value Theory later in this chapter).17
Because of complex and non-linear (option) positions that are present in most
AIS portfolios, the variance-based method can be quite misleading and should be
avoided. Historical simulations are attractive as they make no distributional
assumptions, but the results depend strongly on the historical period chosen for
the analysis. The Monte Carlo approach is usually the most reliable method, but is
at the same time computationally the most intense.

.... 179 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 180

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

VaR: Back testing and model verification

It is important to realize that the VaR figures any model generates are subject to
error. These errors have various sources (sampling error, model error, human
error etc.). It is essential that the VaR numbers themselves are monitored on a reg-
ular basis. As VaR values are not directly observable in the markets, this can only
happen by checking whether the calculated numbers are consistent with the subse-
quently realized returns. The most obvious way to do this is by investigating how
often losses exceed the calculated VaR levels, and whether such losses occur signif-
icantly more or less frequently than indicated by the chosen confidence level.
More advanced tests take into account the sizes of losses exceeding VaR. There
are a number of different tests available.18 Their common problem is that due to
sparse data, they tend to have a limited ability to distinguish between good and
bad models. Bad models are too often qualified as good ones.

The problems of VaR

VaR cannot cover all aspects of risk. It provides useful information about portfo-
lio risk under normal market circumstances, but does not address the extreme left
tail region of the return distribution (which becomes important in the case of
extreme market moves and turbulences). This is one of several problems with
using VaR as a sole risk quantification tool.

1 VaR measures potential losses that occur fairly regularly (e.g. once in 20 or
once in 100 days). It does not provide information about the extreme left tail
of the profit and loss (P&L) distribution and the expected size of an
experienced loss that exceeds VaR. Often it is the rare but extreme losses that
are of most interest for a risk manager. For the past 15 years, at least one
major market has moved by more than ten standard deviations in almost every
year. Examples include the 1987 stock market crash, the 1990 Nikkei crash
and High Yield tumble, the 1992 European currency crisis, the 1994 US
interest rate hike, the 1995 Mexico and South America crisis, the 1997 Asian

.... 180 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 181

6 ■ Tools and Principles of Modern Financial Risk Management

crisis, the 1998 Russian crisis and the fall of LTCM, the 1999 Brazilian crisis
and the 2000/2001 high-tech market downturn. These are market
environments where VaR fails to accurately account for risk.
2 VaR relies heavily on its assumptions. One major challenge in implementing
VaR is the specification of the probability distribution of extreme returns. As
mentioned earlier, in most VaR schemes the risk manager chooses a normal
return distribution with a stationary (i.e. structurally stable) covariance matrix
of risk factors. These can be good assumptions on some days and poor ones
on other days. Unfortunately, the latter are the moments when VaR is most
needed. Much has been written about the assumption of normality of the
returns, which is the most important of all assumptions made by VaR models.
It is well known in mathematics that the covariance matrix (next to expected
returns) uniquely describes a multivariate normal distribution. The assumption
of normality makes it easy to calculate VaR from the standard deviation at any
desired confidence level by simply multiplying the latter by a z-factor (1.65 for
the 95% confidence level and 2.33 for the 99% confidence level). There is
strong empirical evidence, however, that security returns are non-normally
distributed.19 Non-normal return distribution can be considered by
investigating higher moments, specifically the skewness (unsymmetrical
distribution) and kurtosis (‘fat tails’). The characteristics of VaR using non-
normal distributions are still subject to intense research.20
VaR based on full revaluation (like the Monte Carlo and historical
method) depends on the particular pricing models employed for the valuation
of financial instruments. Many derivative instruments are not very actively
traded, so they have to be marked to model rather than marked to market.
3 VaR relies on estimates of correlations and volatilities. These are usually based
on historical data. It is assumed that historical returns are approximately
stationary, i.e. their behaviour will not change structurally in the future. In
reality, correlations and volatilities are unstable and change constantly.
Correlation and volatilities in specific market environments (also called
‘conditional correlation/volatilities’) can be quite different from their average

.... 181 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 182

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

(‘unconditional’) values. Especially in times of financial crisis, volatility

increases well above historical averages and correlations tend to be
significantly higher than in normal market environments. Events tend to
happen discretely rather than continuously. In other words, the events that
expose the investor to extreme risks are often not predictable from historical
time series. Further, like all statistical measures, correlations and volatilities
have statistical errors. The smaller the sample of observation, the greater their
impact. VaR depends on how volatilities and correlations are calculated from
past data. Calculations based on equally weighted historical returns can
deviate from the results of exponentially weighted schemes.
4 VaR only captures certain systematic risk factors, such as market (equity, bond,
FX, commodity) or credit risk. VaR does not measure non-systematic risk,
which can present itself in a variety of disguises: idiosyncratic (e.g. corporate
event) risk, liquidity risk, credit spread risk, operational risk, political risk,
model risk. With a generally higher degree of non-systematic risk in their
portfolios, VaR is less likely to provide reliable approximation of total risk for
AIS investors. Event Driven strategies, such as Distressed Debt or Merger
Arbitrage strategies, provide good examples of this. They are exposed to a
large amount of deal- or company-specific risk that cannot be reliably
described by VaR.
5 VaR has an important and widely criticized theoretical shortcoming: It is not
additive with respect to sub-portfolios, i.e. the VaR of a portfolio consisting of
two sub-portfolios can be larger than the sum of the single VaR values of the
two sub-portfolios. Thus, VaR does not qualify as a ‘coherent risk measure’.21

Risk managers should be sceptical about breaking down risk into one number.
Risk has many dimensions and different degrees of severity. It is undisputed
among theoreticians and practitioners of risk management that VaR needs to be
supplemented by other analysis tools suited to cope with non-systematic risks and
extreme market conditions. VaR alone is not a sufficient means to quantify poten-
tial unacceptable losses.

.... 182 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 183

6 ■ Tools and Principles of Modern Financial Risk Management

Variations of VaR
There are other risk measures which are related to Value-at-Risk.

■ Marginal and Incremental VaR: Marginal VaR expresses the amount by which
the value of portfolio VaR is increased upon inclusion of a particular position
or sub-portfolio. Incremental VaR is a related concept and measures the
sensitivity of the portfolio risk to very small (incremental) changes in the
portfolio holdings. Incremental VaR is additive (Marginal VaR is not), i.e. the
sum of Incremental VaRs of individual positions equals the Incremental VaR of
the sum of the positions. Incremental and marginal VaR can be used to
evaluate the impact of a particular position or an individual trading manager
on the risk structure of the entire portfolio.
■ Expected Shortfall (Conditional VaR):22 Expected shortfall is the mean value of
the portfolio loss, conditional on the loss exceeding VaR (or any specified
threshold). It provides important information about the left tail of the P&L
distribution. Conditional VaR is a coherent risk measure, i.e. (among other
properties) it is sub-additive. Expected shortfall has desirable features for
portfolio optimization algorithms, in which risk is minimized under given
constraints (in particular, it is a convex function of portfolio weights).23
■ ‘Lower partial moments’ (LPM): A generalization of risk measures that
includes Expected Shortfall as a specific sub-case, are ‘lower partial moments’
(LPM). A set of lower partial moments can be defined by the nth power of the
loss exceeding a certain threshold:

LPM(n) = E[(return – threshold)n] for (return < threshold)

For n=1 this measure reduces to the Conditional VaR. For n=2 and the
threshold equals the expected return, this measure reduces to the semi-
variance, i.e. the variance with only returns below the expected return taken
into account (the square root of which is often referred to as ‘downside
deviation’). The use of lower partial moments in the (multivariate) portfolio is

.... 183 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 184

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

generally more difficult, because the portfolio risk measured by LPM is not
composed of the individual securities’ risk and correlations only. Thus,
although LPM capture more specific measures of risk, they are not as
convenient to estimate or to derive analytical solutions for.

Stress testing and scenario analysis

Scenario analysis consists of applying predetermined changes to the values of risk
factors and determining the resulting changes of portfolio value as a result.
Scenarios can be applied independently to each instrument/asset class or in con-
sideration of risk factors’correlations. The latter is more useful, as it accounts for
co-movement in risk factors other than those considered directly in the scenario.
Stress tests make use of extreme scenarios in order to ascertain coverage of the
impact of large price changes in financial markets. Such tests give insight into the
portfolio behaviour under extreme and low probability, but, nevertheless, plausi-
ble market conditions, which are outside the predictive capability of any statistical
model. Stress tests are usually performed by applying an appropriate stress event
and observing the resulting portfolio re-valuation. The choice of stress scenarios is
the crucial factor in performing these tests. Stress scenarios should include:

■ Appropriately sized shocks: Extreme events are characterized by large moves,

which should be accounted for in scenario definitions.
■ Test for asymmetries in return distribution: Skewed distributions are
omnipresent in financial markets.
■ Test for correlation breakdown: Extreme events usually lead to very different
correlation structures in financial markets. Investments that are uncorrelated
in normal market environments can be highly correlated during times of
financial crisis.
■ Stressing different combinations of asset classes separately and in combination:
The best way to perform stress tests is to include all assets in the scenario and
apply realistic correlations between instruments to define the full impact of a

.... 184 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 185

6 ■ Tools and Principles of Modern Financial Risk Management

specific scenario. For example, the behaviour of bonds, commodities and

foreign exchange rates during turmoil in equity markets should be considered
through certain assumptions about their correlations to stocks.

Scenario selection is the most important and most difficult part of stress testing
and is much more an art than a science.24 Three different sorts of scenarios can be
distinguished:

■ Historical scenarios: The replication of past events is a rather simple way to

define a scenario. Examples are the stock market crash in 1987 or the
emerging market crisis (Russian default) in 1998.
■ General user-defined market scenarios: In order to incorporate current market
structures rather than relying on historical events, the risk manager can define
specific changes to different risk factors (e.g. a drop of 15% in the broad
equity markets or a parallel shift of 100bp in the US dollar yield curve).
■ Portfolio-specific scenarios: The particular portfolio in consideration can guide
the definitions of scenarios. Risk managers should examine the impact of
extreme losses in particular positions in the portfolio or the result of a sudden
change in their correlations. Portfolio-specific scenarios can include expert
views on certain subsets of risk factors and the possible future correlation
scenarios between core risk factors.25

The second part of stress testing is the re-valuation of the portfolio based on chosen
scenarios. The valuation is usually based on the same models as used for the Monte
Carlo simulation. The stress test result is obtained by calculating the difference
between the portfolio value after the stress test and the current portfolio value.

Extreme Value Theory

Academics as well as practitioners in financial markets generally agree that financial
risk factors are not normally distributed. They exhibit fat tails, i.e. extreme events
are more likely to occur compared to what the normal distribution suggests. But
accurate prediction of extreme events is of central importance to risk analysis. As

.... 185 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 186

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

discussed, most VaR models have rather poor predictive capabilities for the tails of
the P&L distribution. Even historical simulations are subject to sampling problems
in the tails, as extreme events do not occur sufficiently regularly to apply valid statis-
tical judgement. In order to cover the fat-tailed distribution of financial risk factors,
a non-normal probability distribution is more appropriate to work with. A first
alternative choice is the student-t distribution, which shares some attractive proper-
ties with the normal distribution. Mathematically speaking, it falls into the class of
‘elliptical distribution’, i.e. the correlation matrix uniquely describes the dependence
structure of multivariate t-distributions (see below).26
The mathematical description of extreme events requires a solid mathematical
foundation and is outside the scope of this book. The following explanations
should serve the purpose of a short overview only. The interested reader is asked
to study the references provided for a more in-depth discussion. An understanding
of the mathematical discussion that follows is not essential to the further compre-
hension of this book.
The ‘central limit theorem’ provides the foundation for the statistics of nor-
mally distributed events, as it describes the ‘Gaussian’ (normal) distribution as the
generic form of the ‘middle part’ of many probability distributions. The ‘Fisher-
Tippett theorem’ in contrast provides the generic form of the asymptotic tails for
a wide class of probability distributions.27 The theorem is central to the mathe-
matical discipline of Extreme Value Theory (EVT), which provides the theoretical
foundation for the analysis of leptokurtic (i.e. fat-tailed) distributions. The three
generic distributions for extreme events (the ‘extreme value distributions’) are
those defined by the ‘Frechet’, ‘Gumbel’ and ‘Weibull’ distributions and all three
are represented by the ‘Generalized Extreme Value (GEV) distributions’, a frame-
work in which each distribution can be characterized by what statisticians refer to
as the ‘tail exponent’. The empirical models based on EVT are usually calibrated
to the extreme ends of the distributions and do not cover the more common
events.28 Risk managers therefore sometimes work with a mixture of different dis-
tributions for the different regimes.
Non-normal marginal (i.e. univariate) distributions for individual asset returns
in a portfolio lead to very complex mathematical problems, when the correspon-

.... 186 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 187

6 ■ Tools and Principles of Modern Financial Risk Management

ding multivariate portfolio distribution is examined. As already mentioned, for

multivariate normal distributions the correlation matrix uniquely defines the
dependency structure between the individual processes (asset returns),29 which
makes them very handy in practical applications. In financial markets, however, the
statistical properties of extreme events are too complex to be covered by the
assumptions of statistical normality. Multivariate normal distributions cover the
dependency structure of rare events insufficiently. Compared to what is empirically
observed in financial markets, joint extreme events of individual processes (asset
returns) occur too seldom in normally distributed multivariate processes. For non-
normal (generally ‘non-elliptical’) distributions the correlation matrix does not
describe the dependency structure of the multivariate process adequately. Two
strongly dependent variables could have a low or even zero (linear) correlation,
which can be illustrated with the following simple example: Consider a normally
distributed random variable X and the random variable Y=X*X. Although they are
perfectly dependent (as they can be determined from each other with deterministic
certainty) their (linear) correlation is zero!
Alternative measures for statistical dependency, especially for the dependency of
rare events, are the subject of intense research.30 One approach is to quantify
dependency through conditional quantiles. These describe the likelihood that one
variable will take on extreme values conditional on another variable. A further
method is rank correlations, which can be used as a more reliable correlation meas-
ure. They are computationally simple enough to be used by practioners. The general
theory of ‘copulas’ provides a way to separate the joint multivariate distribution into
two different parts: the marginal univariate distributions and the dependence struc-
ture, called the ‘copula’.31 The joint multivariate distribution can be described as a
convolution of the copula function and uniformly distributed marginal distributions.
Copula theory provides the possibility of including more complex dependence
structures into a Monte Carlo setting.32 It should be noted, however, that introduc-
ing non-normal distributions into financial risk management is a difficult and
complex undertaking. Most commercial risk management software providers do not
offer users the possibility of dealing with this degree of complexity.

.... 187 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 188

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

After determining the tail properties of the portfolio distribution, VaR can be
estimated as a quantile of the corresponding extreme value distribution.33 Studies
have shown that VaR calculated on the basis of EVT is considerably higher than
normal VaR.34 In other words, VaR calculated on the basis of standard (normal)
volatilities and correlations often severely underestimates real risk.

Liquidity Risk
Liquidity Risk takes two different forms: asset liquidity and funding liquidity risk.
Asset liquidity risk describes the potential inability to sell a position at fair value
because of insufficient trading activity. In order to examine the potential impact of
insufficient asset liquidity, a risk manager should run scenarios using prices that are
discounted from current market prices, i.e. ‘liquidation values’ rather than fair
value. Liquidation values should reflect the likely impact on prices upon forced liq-
uidations. The discount depends on the position size relative to trading volume and
bid–ask spreads.35
Funding risk is of central concern to AIS managers. A lack of funding can con-
tribute to a crisis situation where the Hedge fund has to liquidate positions to
raise cash, either because the fund must post margin with its counter-parties (usu-
ally as the result of a loss) or because of redemptions by investors. The manager
should constantly track the fund’s cash positions (cash and short-term securities of
high credit quality) and its borrowing capacities (margin rules and credit lines) on
the one side and the potential need for liquidity on the other. The level of liquid-
ity required is dependent on the nature of the instruments traded (e.g. Futures
require much less cash than other instruments), the level of leverage and risk (as
can be measured by VaR, scenario analysis or stress testing) and redemption poli-
cies. Possible funding liquidity measures include the cash/equity ratio, the
VaR/(cash plus borrowing capacity) ratio and the worst historical drawdown/(cash
plus borrowing capacity) ratio. Hedge fund managers should have good commu-
nication lines with their credit providers, giving them sufficient access to
information about the fund’s risk and liquidity levels.

.... 188 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 189

6 ■ Tools and Principles of Modern Financial Risk Management

Credit Risk
While the decade of the 1990s saw tremendous advances in the modelling of
market risk, more recently, quantitative credit risk modelling has emerged as a
major focus of risk management. Important advances in the analysis of credit risk
have led to the proliferation of a new breed of sophisticated quantitative credit
models. These models build on the extensive mathematical modelling of default
and loss probabilities for single obligors as well as correlations between obligors.
Several commercially available approaches have received a great deal of attention,
of which the five most prominent are:36

■ CreditMetrics (JP Morgan)

■ CreditMonitor (KMV)
■ RiskCalc (Moody’s)
■ CreditRisk+ (Credit Suisse Financial Products, now Credit Suisse First Boston)
■ CreditPortfolioView (McKinsey).

To the practitioner, the seemingly disparate collection of new approaches may be

confusing and should be interpreted as a warning sign of the early developmental
stage of the technology.37 Credit risk imposes new dimensions of complexity, as
return distributions are largely non-normally distributed and sufficient diversifica-
tion is reached only after including hundreds or thousands of positions in the
portfolio. While a variety of theoretical modelling approaches have been devel-
oped, the empirical testing of these models is still in its infancy.38
The goal of credit risk analysis is to calculate the probability distribution for credit
losses in order to define a ‘Credit VaR’ similar to ‘Market VaR’. But modelling credit
risk is significantly more complex than market risk, as there are numerous factors
driving the value of debt. Additionally, credit and market risk can be highly interre-
lated. One of the most critical issues in modelling credit risk is the data input (past
default rates, recovery rates, observed correlations between different obligors etc.).
The three main elements of credit risk are exposure, default probability (including
market sentiments about its possible changes) and loss following default, which is
expressed by recovery rates. Default probability is assessed by the credit rating (for

.... 189 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 190

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

debt rated by agencies) or quantitative models (e.g. the Merton model; see later).
Finally, exposure is defined by the investor or modelled through scenario simulations
(e.g. for OTC swaps where future exposure is not certain at present).
The expected loss and the distribution of losses can be estimated through prob-
ability distributions of default and the resulting loss following default. Loss
following default is composed of the net exposure to the firm and the forecasted
recovery rate in case of default. This is a function of the debt-issuing firm’s asset
value versus the face value of its liabilities. Research into recovery rates is a rather
undeveloped area in credit risk management and one where consensus in
approach is less clear.
While all three elements are important variables in evaluating credit risk, none
is more important than the default probability. Prior to default, there is no way to
distinguish unambiguously between firms that will default and those that will not.
At best the risk manager can make a probabilistic assessment of the likelihood of
default for individual firms.
Two general frameworks for calculating default probabilities have evolved, each
with different approaches modelling the ‘economics of default’. The first method,
which is referred to as ‘structural modelling’, was pioneered by Merton and describes
the firm’s liabilities as contingent claims issued against the firm’s underlying assets.39
According to Merton, default happens for an explicit reason – the value of the firm’s
assets falls below the face value of the debt (in which case the debt holder receives the
value of the remaining assets). Making rather simple assumptions as to the capital
structure of the company, assuming no liquidation costs and constant interest rates,
the payoff to a bank making a loan is isomorphic to writing a put option on the assets
of the borrowing firm (equivalently, equity of the firm can be seen as a call option on
the firm’s assets). Given this framework, it is not surprising that the Black–Scholes
(BS) option pricing formula is at the heart of all structural models. There are numer-
ous extensions of the basic model introducing more realistic capital structures.40
Three of the most popular credit risk models, KMV CreditMonitor, JP Morgan’s
CreditMetrics and Moody’s RiskCalc, fall within the extended class of Merton-based
structural models. And although McKinsey’s CreditPortfolioView and CSFB’s
CreditRisk+ do not rely on explicit modelling of the firm’s asset value, they can be
summarized under a similar generalized framework.41

.... 190 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 191

6 ■ Tools and Principles of Modern Financial Risk Management

A second class of models abstracts completely from the economics of default.

These models, summarized under the term ‘reduced form’ models, try to esti-
mate the risk neutral probability of default from prevailing credit spreads
quoted in the market. No reference to the cause of default is made. 42 This
approach relies on the availability of high-quality data for credit spreads, a pre-
requisite that is often not fulfilled. The deeper and more liquid the market, the
better the reduced-form models are expected to work. Reduced-form models
are not applicable to the obligations of small and micro cap firms with no
publicly traded debt.

Other risks
The development of risk management tools in recent years has mainly been
geared towards market and credit risk analysis. It remains more difficult to quan-
tify other risks. Pushed by regulators, the financial community is now turning to
operational risk, which in many cases has proven to be an important cause of
large financial losses (e.g. Sumitomo: $2.6 billion loss, Barings: $1.3 billion loss,
Daiwa: $1.1 billion loss, all three in 1997). Most of these losses can be attributed
to a combination of exposure to market and credit risk along with failures of con-
trols. Operational risk can be defined as the risk of losses due to people
(employees), processes (errors in internal systems or model, hedging errors), tech-
nology (IT programs or data), external factors (human factors outside the firm) or
physical events (weather, theft etc.). Besides the regulatory capital charges to
internationally active financial institutions for market and credit risk, which were
established in 1998 and 1996, the Basel Committee has now proposed the estab-
lishment of capital charges for operational risk.43 As a consequence, the
measurement of operational risk is currently the subject of intense discussions and
new models are under development.44 The various approaches can be broadly
classified into ‘top down’ and ‘bottom up’ models. While top down models meas-
ure operational risk at the broadest level (i.e. firm wide or even industry wide),
bottom up models start at the individual process level within business units.

.... 191 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 192

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The focus on operational risk is generally geared towards larger financial insti-
tutions, where processes and the interaction between different units can be rather
complex. Hedge funds are usually smaller and less complex from an operational
viewpoint. Nevertheless, operational risks are quite similar. Human error, model
risk and exposure to technology risk are equally present in most Hedge fund
operations. Some Hedge funds exploit market inefficiencies with very small price
margins. Here, frequent execution errors can quickly erase otherwise lucrative
returns. Methods that are used to measure operational risk include the following
quantitative and qualitative elements:45

■ Audit oversight: Review by an external auditor (or audit department of the

same firm in the case of large organizations).
■ Critical self-assessment: Each business unit or key person identifies the nature
and size of operational risk.
■ Key risk indicators: Audit scores, staff turnover, trade volume etc.
■ Earnings volatility: This includes business and macroeconomic risks and might
be less relevant for AIS managers.
■ Causal networks: Causal networks describe how losses can occur from a
cascade of different causes, which are linked to each other through
conditional probabilities. Most causal networks are based on Bayes’ theorem
and Bayesian statistics.46
■ Actuarial models: Actuarial models combine loss frequencies with loss severities
to generate losses distributions. Most actuarial models estimate the distribution
of losses, i.e. the loss frequency and loss severity distributions, from historical
data. Assuming that frequency and severity of losses are independent, these two
distributions can be combined into the distribution of aggregate losses through
what is called ‘convolution’ in mathematics. In practice, the calculation of
aggregate losses is often performed through numerical simulation. With a loss
distribution at hand, one can define an ‘Operational VaR’.
■ Extreme Value Methodology: Most operational loss distributions are non-
normally distributed and show fat tails. Some promising methods from the

.... 192 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 193

6 ■ Tools and Principles of Modern Financial Risk Management

Extreme Values Theory (EVT, see earlier) have lead to interesting applications
for operational risk analysis. Many of the developments and applications of
EVT in the field of finance originated within the insurance industry, where
due to the nature of the business the focus is much more directly on extreme
event risk.

Once operational risk is measured and quantified, it has to be controlled.

Operational risk management consists of various procedures and control mecha-
nisms. The following provides a list of some of the most important elements:

■ Separation of functions: This is one of the key risk management principles and
applies equally to large and small institutions. Responsibility for trading
should be separated from clearing, accounting and risk management functions.
■ Incentives: The compensation of the people responsible for accounting and risk
management should be independent from the success of trading or asset
management activities. Generally, the incentive structure within the firm should
encourage people to ask the right questions and express concerns openly.
■ Monitoring: Control policies and compliance rules should be in place to
ensure that objectives are carried out. Policies and procedures should spell out
expectations regarding the integrity and ethical behaviour of employees.
■ Risk-adjusted performance compensation: Traders and asset managers should
be compensated based on risk-adjusted performance, e.g. RAROC (‘risk-
adjusted return on capital’).
■ Involvement of senior management: Senior management should set clear risk
management guidelines and capital policies and enforce those efficiently.
■ Dual entries and reconciliation: Entries and final positions should be matched
from different sources. Each counter-party needs to confirm the trade tickets.
■ Price verification: Prices should be obtained from several external sources. The
institution should have the capability to value a transaction in house before
entering it.
■ Authorizations: All counter-parties should be provided with a list of personnel
authorized to trade.

.... 193 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 194

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Internal and external audits: Auditors provide useful information on potential

weaknesses.

Enterprise-wide risk management

Best risk management practices in the industry have evolved from the lessons of
the financial disasters in the 1990s. Financial institutions have realized that it is in
their best interest to promote a set of guidelines to forestall further regulatory
action. In 1993 the Group of Thirty (G-30) issued a report,47 which provides 24
sound risk management principles and became a milestone for risk management.
The failure of Barings Bank was followed by an in-depth report from the Bank of
England.48 Finally, the failure of Long-Term Capital Management (LTCM) led to
some very useful lessons, especially for the Hedge fund industry. In the wake of
this event, the Counter-party Risk Management Policy Group (CRMPG) was
established to strengthen financial risk management practices.49 Chapter 7 pro-
vides a further discussion of these reports.
The interaction of various types of risk requires particular attention, as differ-
ent sources of risks (market credit, credit risk, operational risk) are not acting
isolated from each other. The industry increasingly focuses on ‘integrated risk
management’ or ‘firm-wide risk management’. The major elements of risk man-
agement at the enterprise level are analysis, integration, data and reporting.50 The
focus on ‘firm-wide risk management’ is a recent development and the challenge
ahead is the extension of risk management beyond the sole monitoring of tradi-
tional credit or trading exposure. The establishment of feedback loops between
senior management and individual business and trading units allows for a proac-
tive response to risk assessment.
A general note: The ‘catastrophic risks’, i.e. the risks that can lead to failure of
the entire firm, are not the most complex to detect from an analytical standpoint.
They are usually due to failures in the organizational structure of the firm. Many
institutions put analytical efforts at the centre of risk management attention,
which leads to the focus being on providing answers to questions that have
already been answered with ever more detail and sophistication. But digging more
deeply into one spot does not ensure that one is digging at the right spot.

.... 194 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 195

6 ■ Tools and Principles of Modern Financial Risk Management

Active risk management

Much of the discussion about risk management centres on the question of how to
measure or analyze risk. However, a critical dimension to risk management is the
decision-making process after the analysis of risk. In order properly to control risk
and achieve the objective of maximizing risk-adjusted returns, risk management has to
be integrated into the investment decision-making process. Risk management entails
monitoring exposures and risks and using the risk measures to allocate risk optimally
among different assets and strategies with refinements and modifications undertaken
when necessary. The optimal allocation of risk among different AIS requires the
assessment of which risks should be borne for which level of expected return. The
allocator should define the acceptable risk level within the portfolio and judge current
risk as measured by VaR, stress tests or other risk characteristics against it.
The important initial step in effective risk management is setting up a diversified
portfolio. The theory of modern portfolio management (MPT) and the resulting
quantitative tools are well known by most portfolio managers. But the quantitative
optimization of AIS portfolios bears a number of pitfalls and significant qualitative
analysis is needed on top of the quantitative tools for the set-up of a well-diversified
portfolio (see Chapter 7 for more details). Risk limits on individual instruments, sec-
tors or managers/traders (‘risk budgeting’ or ‘risk bucketing’)51 serve the purpose of
avoiding excessive one-sided exposures in a portfolio. Stop-loss limits help to con-
trol losses due to sudden unexpected strong price moves.
Risk management requires action. Action must be taken in cases where: (a) risk
limits are exceeded; (b) it is decided that exposure to certain risk factors should be
reduced; or (c) external events demand cutting certain risky positions. This
requires reliable quantitative tools and a good understanding of the relationship
between risk levels and market developments. ‘Marginal risk’ analysis can be
useful to mitigate deteriorating effects in case of an unfavourable event.
One of the most important principles of risk management is the separation of
responsibilities between risk takers and risk managers. Risk management should
be an independent function that is implemented and directly supervised by a
senior management. The decision-making power for risk control should be put
into the hands of those who oversee the risk analytics and not those who take the

.... 195 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 196

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

risk. The principles that apply to large institutions apply equally to AIS managers,
even though they often operate within the structure of a small firm without inde-
pendent risk management, where risk manager and trader are often one and the
same person. Many Hedge fund managers implement independent risk manage-
ment through strict adherence to predefined risk principles such as stop losses,
risk limits, maximal exposure to single asset classes etc. This corresponds to how
many AIS managers interpret their success. Asked about the reasons for their suc-
cess many answer: ‘Investment discipline.’

Risk analysis systems in practice

The development of a fully operative risk analysis system that enables the user to per-
form the sort of calculations discussed (VaR calculations, stress tests, scenario analysis
etc.), is a very complex undertaking. The three critical issues are the build up and
maintenance of a database for historical volatility and correlation analysis, the defini-
tion of relevant risk factors and the construction of pricing models.
Many Hedge funds want access to enhanced risk analysis but lack the financial
and human resources to build a full risk analysis structure themselves or to run and
maintain an externally purchased system. A common solution for many AIS man-
agers is the complete outsourcing of their risk analysis to external service providers.
The demand for quantitative risk analysis tools and the increased availability of
computational power in recent years has led to a number of firms specializing in
providing risk services to investment managers. AIS managers have become an
important part of their client base. Similarly, most prime brokers have begun offer-
ing risk services to their AIS clients as a value-added service in return for obtaining
their attractive execution and clearing business.
Communication between service provider and investment manager happens
increasingly through the internet. With its emergence, many firms have web enabled
their software and run their services on an ASP (‘application service provider’)
basis.52 An ASP is a service that hosts and runs specific applications for a number of
different customers. While this occurs in most cases with little customization, some
AIS managers do request tailor-made analytics. An example for an ASP implementa-

.... 196 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 197

6 ■ Tools and Principles of Modern Financial Risk Management

tion of a risk analysis system offered by a prime broker is PrimeRisk at Credit Suisse
First Boston, which is based on Algorithmic’s RiskWatch.53
Outsourcing risk analysis has several advantages for the AIS manager:

■ He does not have to set up his own system for the risk analysis and valuation
computation of very complex instruments.
■ He does not have to gather all the data required to model risk appropriately.
■ He does not have to hire full-time risk management staff.
■ He gains credibility when an independent third party not involved in
marketing his products performs the risk analysis.
■ He gains access to risk analysis techniques that until recently only the big
investment firm could afford to operate.

But there are problems with outsourcing risk analysis. One might not know exactly the
nature and the origin of the risk analysis output. Like any number-crunching exercise,
it is difficult to interpret numerical results if one does not know precisely the calcula-
tions behind them. It is therefore essential for the AIS risk manager to have an
understanding of the risk systems, their underlying features and the models used for
the calculations.
Readers of financial magazines are confronted with a growing number of
advertisements for risk management services and increasingly these target particu-
larly the AIS manager. The seemingly disparate collection of risk service suppliers
and software vendors may be confusing for practitioners. In the following, I pro-
vide a list of some key players in this market (the URLs are included in
parentheses for further information). A description of the individual systems is,
however, beyond the scope of this book:

Algorithmics (http://www.algorithmics.com)
Askari (http://www.askari.com)
Barra (http://www.barra.com)
BlackRock Solutions (http://www.blackrocksolutions.com)
Financial Engineering Associates (http://www.fea.com)

.... 197 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 198

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

GlobeOp’s (http://www.globeop.com)
Imagine Trading System (http://www.imagine-sw.com)
Measurisk (http://www.measurisk.com)
Misys (formerly MKIRisk) (http://www.misys-ibs.com)
Kiodex (http://www.kiodex.com)
Kronos (http://www.krns.com)
Panalytix (http://www.panalytix.com)
PrimeRisk (Credit Suisse First Boston: http://www.primerisk.csfb.com)
Reech Capital (http://www.reech.com)
Reuters (http://about.reuters.com/risk)
RiskBox (http://www.riskbox.com)54
RiskMetrics (http://www.riskmetrics.com)
Summit Systems (http://www.summithq.com)
SunGard’s Opus, Infinity, Panorama (http://www.risk.sungard.com)
Ubitrade (http://www.ubitrade.com).
Wilshire (http//www.wilshire.com)

As the market for risk services is subject to constant change, the list of service
providers is necessarily incomplete (also some vendors might disappear from the
market). In January 2002, the firm Capital Market Risk Advisors (CMRA) com-
pleted a review of risk management systems for investment managers, Hedge
funds, funds of funds, plan sponsors, endowments and foundations.55 Risk
Magazine frequently publishes software and vendor surveys, which describe the
latest developments in this particular field of financial technology.56

Notes
1. Despite being a few years old, the original technical RiskMetrics documents
(‘RiskMetrics’, 1996 and ‘CreditMetrics’, 1997) provide an excellent discussion of
the fundamental market and credit risk principles and models. In 2001, the
company came out with a new edition of the document for market risk: ‘Return to

.... 198 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 199

6 ■ Tools and Principles of Modern Financial Risk Management

RiskMetrics’. A broader coverage of different risks and further references are

presented in The Professional’s Handbook of Financial Risk Management by M.
Lore and L. Borodovsky (eds) and the GARP Association (General Association of
Risk Professionals).
2. An excellent textbook designed for the candidates of the FRM (Financial Risk
manager) examination of the GARP association (http://www.garp.com) is The FRM
Handbook 2001/2002 by P. Jorion. This book introduces the main concepts of
modern risk management and provides numerous references.
3. See: Group of Thirty, Derivatives, ‘Practices and Principles’, New York; the
report is available on the IFCI web page risk.ifci.ch, maintained by the
International Finance and Commodities Institute (IFCI), a non-profit Swiss
organization.
4. The concept of duration is discussed in almost all textbooks about finance, e.g.
Investment Analysis and Portfolio Management by F. Reilly and K. Brown.
5. See, for example, Investment Analysis and Portfolio Management, by F. Reilly
and K. Brown.
6. For a good overview, see J. Hull, Options, Futures, and other Derivatives, 4th
edition, Prentice Hall, 1999, Chapter 13. The ARCH/GARCH approach is well
covered in the review paper by T. Bollerslev, R. Chou and K. Kroner, ‘Arch
Modelling in Finance’, Journal of Econometrics, 1992, 52, pp.5–59.
7. VaR is discussed in all textbooks about risk management, e.g. Mastering Value
at Risk: A Step-by-Step Guide to Understanding and Applying VaR by C. Butler.
8. There exist numerous publications covering this topic. I recommend the
discussion provided in the RiskMetrics Technical Document (the one from 1996 is
more detailed than the 2001 version). P. Best also provides a good overview of the
subject in Chapter 5 of Implementing Value at Risk, 1998.
9. The RiskMetrics Technical Document provides a discussion of the individual
risk factors. For commercially available risk management software, the technical
documents usually include a discussion of the risk factors used and the details of
the pricing and mapping procedure.
10. The construction of yield curves is discussed in detail in Chapter 2 of The
Professional’s Handbook of Financial Risk Management by M. Lore and

.... 199 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 200

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

L. Borodovsky (eds). The DataMetrics Technical Document by A. Malz also

provides a detailed explanation of how to construct a yield curve from a set of
bond and swap prices.
11. More details about both methods are described by P. Best in Implementing
Value at Risk, p.114.
12. For more information and a discussion of the quality of different volatility
and correlation forecasts, see the article ‘Volatility and Correlation Forecasting’ by
C. Alexander in The Handbook of Risk Management and Analysis by C. Alexander
(ed.), 1996.
13. See the article by T. Bollerslev et al., ‘Arch Modelling in Finance’, Journal of
Econometrics, 1992, 52, pp.5–59.
14. This method is often called the ‘RiskMetrics VaR’ method, as this was the
original method introduced by RiskMetrics in the early 1990s. Today, RiskMetrics
also offers other parametric VaR approaches as well as Monte Carlo-based and
historical simulation-based VaR.
15. In Beyond Value of Risk (Chapter 5) Kevin Dowd presents a good discussion
about the different aspects of Monte Carlo approaches.
16. See P. Embrechts et al., ‘Correlation and Dependency in Risk Management:
Properties and Pitfalls’.
17. An excellent paper about non-normal distributions is given by P. Embrechts et
al. in ‘Correlation and Dependency in Risk Management: Properties and Pitfalls’,
ETHZ Working Paper, 1998; see http://www.math.ethz.ch/~mcneil/pub_list.html.
A shorter version appeared in the journal Risk, May 1999, pp.69–71. Further, an
overview treatment of the issues involved with non-normality is presented in the
RiskMetrics document ‘Return to RiskMetrics, 2001, Appendix A.
18. A good overview of the different available methods of VaR testing and further
references is provided by K. Dowd in Beyond Value at Risk, Appendix to Ch. 2.
19. See, e.g., the article ‘Evaluating Value at Risk Models using Historical Data’
by D. Hendricks or the article by H. Hauksson et al. ‘Multivariate Extremes,
Aggregation, and Risk Estimation’.
20. There are many articles about this topic. Please refer to the article ‘Portfolio
Value at Risk with Heavy Tailed Risk Factors’ by P. Glassermann et al., 2000,

.... 200 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 201

6 ■ Tools and Principles of Modern Financial Risk Management

which also provides further references.

21. The concept of ‘coherent risk measure’ as well as a discussion on other
desirable theoretical properties of risk measure is discussed in the article by
P. Artzner et al., ‘Coherent Risk Measures’.
22. A good discussion of conditional VaR can be found in S. Uryasev and Tyrrall
Rockafellar, ‘Optimization of Conditional Value at Risk’, under
http://www.ise.ufl.edu/uryasev/roc.pdf.
23. For more details see the following articles: S. Uryasev and T. I. Rockafellar,
‘Optimization of Conditional Value at Risk’; P. Krokhmal, J. Palmquist and S. Uryasev,
‘Portfolio Optimization with Conditional Value-At-Risk Objective and Constraints’.
24. In early 2000, the Bank of International Settlement (BIS) initiated a survey of
stress tests scenarios employed by risk managers. Please refer to the article by I. Fender
and M. Gibson in Risk Magazine, May 2001, p.50, for the details and results.
25. An interesting discussion on the effect of correlation in stress tests is
presented in ‘A Stress Test to Incorporate Correlation Breakdown’ published in the
RiskMetrics Journal in May 2000.
26. Glassermann et al. present some interesting VaR calculations with the student-t
distribution in their article ‘Portfolio Value at Risk with Heavy Tailed Risk Factors’.
27. A good reference for the mathematical properties and theorems of extreme
value theory can be found in P. Embrechts et al., Modelling Extremal Events,
Springer, 1999. See also A. McNeil, ‘Extreme Value for Risk Managers’, in Internal
Modelling and CAD II, Risk Books, 1999, pp.93–113 and Extremes and Integrated
Risk Management by P. Embrecht (ed.).
28. A good presentation on the measurement and modelling of operational risk is
presented in ‘Modelling and Measuring Operational Risk’ by M. Cruz.
29. This generally applies to ‘elliptical distributions’, i.e. distribution with a density
function that is constant on ellipsoids (which include the normal and the student-t
distribution). For more details see the excellent discussion by P. Embrechts et al. in the
article ‘Correlation and Dependency in Risk Management: Properties and Pitfalls!’.
30. The following two articles provide good discussions of the subject and include
numerous references: P. Embrechts et al., ‘Correlation and Dependency in Risk
Management: Properties and Pitfalls’, and H. Hauksson et al., ‘Multivariate

.... 201 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 202

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Extremes, Aggregation, and Risk Estimation’.

31. For an introduction to copula theory, see An Introduction to Copulas by
R. Nelson.
32. For further reference, see ‘Understanding Relationships Using Copulas’ by
E. Frees and E. Valdez.
33. For an overview and further references for tail index estimation and
applications of EVT to financial risk analysis, see J. Danielsson and C. de Vries,
‘Value at Risk and Extreme Returns’, Tinbergen Institue discussion paper, TI, 98-
017/2, 1998. Downloadable from
http://www.fee.uva.nl/BIEB/TIDPs/TIDP98nr.htm. The paper proposes an
interesting semi-parametric approach, where smaller risks are captured by the non-
parametric empirical distribution function (historical method) and larger risks are
modelled parametrically by an estimation of tails exponents.
34. S. N. Neftci, ‘Value at Risk Calculations, Extreme Events, and Tail
Estimation’, The Journal of Derivatives, Spring 2000, p.23 and the paper by
J. Danielsson and C. de Vries, ‘Value at Risk and Extreme Returns’.
35. See the article by D. Cosandy in ‘Adjusting Value-at-Risk for Market
Liquidity’, Risk Magazine, October 2001, p.115 for an interesting discussion on
how to quantify liquidity risk.
36. A. Saunder presents an overview of the different approaches in Credit Risk
Measurement, Wiley Frontiers in Finance, 1999. The author introduces the
different models mentioned earlier, but also gives a rather extensive literature list
covering the most recent academic and non-academic research results.
37. A. Hickman and U. Koyuoglu demonstrated in an excellent research paper
that all these different models can be considered as quite similar once simplified to
one systematic risk factor. They are shown to effectively produce the same result.
This lead the Basel Commission to embrace credit risk modelling and base their
new framework for a revision of the 1988 Basel Accord on Internal Rating-Based
(IRB) capital requirements. See U. Koyluoglu and A. Hickmann, ‘Reconcilable
Differences’, Risk Magazine, October 1998.
38. An interesting survey on empirical studies can be found in J. Bohn, A Survey
of Contingent Claim Approaches to Risky Debt Valuation, Haas School of Business,

.... 202 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 203

6 ■ Tools and Principles of Modern Financial Risk Management

1999. See also by the same author: Empirical Assessment of a Simple Contingent-
Claims Model for the Valuation of Risky Debt, Haas School of Business, 1999.
39. R. Merton, ‘On the Price of Corporate Debt: The Risk Structure of Interest
Rates’, Journal of Finance, 29, p.440 (June 1974).
40. For further information and references, see M. Crouhy, D. Galai and R. Mark,
‘A Comparative Analysis of Current Credit Risk Models’, Journal of Banking and
Finance, 2000, 24, pp.59–117; M. Gordy, ‘A Comparative Anatomy of Credit Risk
Models’, Journal of Banking and Finance, 2000, 24, pp.119–49; and the work by the
Basel Committee on Banking Supervision, ‘Credit Risk Modelling: Current Practices
and Applications’, Basel, 1999, available on the BIS homepage: http://www.bis.org.
41. CreditRisk+ actually makes no assumptions about the economic reason for default
and for that reason might be better classified as a ‘reduced form’ model. But the model
does not require any credit spread data, as reduced models do. The general assessment
of the default rate distribution for an individual firm is quite similar to the one in
structural models, where CreditRisk+ assumes an C-distributed default rate distribution.
See also ‘CreditRisk+’, Credit Suisse Financial Products (now CSFB), 1997.
42. See F. Longstaff and R. Schwartz, ‘A Simple Approach to Valuing Risky Fixed
and Floating Rate Debt’, Journal of Finance, 1998, 50, pp.449–70; D. Duffie and
K. Singleton, ‘Modelling the Term Structure of Defaultable Bonds’, Stanford
University Paper, 1998; and Gregory Hayt, ‘How to Price Credit Risk’, Risk
Magazine, January 2000.
43. Discussions about this subject are ongoing. The propositions for operational
risk charges in the new amendment of the Basel Accord provide a good perspective
on the difficulties of quantifying operational risk. See the ‘Overview of the New
Basel Capital Accord’ by the Bank for International Settlement, January 2001 and
the discussion in Risk Magazine, March 2001.
44. See the articles by M. Crouhy et al., ‘Operational Risk’; D. Wilson,
‘Operational Risk’ in The Professional’s Handbook of Financial Risk Management
by M. Lore and L. Borodovsky (eds); A. Brewer, ‘Minimizing Operations Risk’, in
Derivatives Handbook by R. Schwartz and C. Smith (eds). The book Operational
Risk: Measurement and Modelling by J. King provides a wide coverage on the
subject of operational risk management.
45. Compare with P. Jorion in The FRM Handbook 2001/2002.

.... 203 ....

8409 Chapter 6 p171-204 11/4/02 1:14 PM Page 204

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

46. For an introduction to Bayesian statistics, see Bayesian Theory by J. Bernardo

and A. Smith.
47. Group of Thirty, ‘Derivatives: Practices and Principles’, New York, 1993,
available on the IFCI website http://risk.ifci.ch.
48. Bank of England, ‘Report of the Board of Banking Supervision Inquiry into
the Circumstances of the Collapse of Baring’, HMSO Publication, London, 1995,
available on the IFCI website http://risk.ifci.ch.
49. Counterparty Risk Management Policy Group, ‘Improving Counterparty Risk
Management Practices’, CRMPG, New York, 1999.
50. For more details see the article by D. Williams, ‘Selecting and Implementing
Enterprise Risk Management Technologies’ in The Professional’s Handbook of
Financial Risk Management by M. Lore and L. Borodovsky (eds).
51. A more detailed discussion about the risk budgeting methodology can be
found in Risk Budgeting: A New Approach to Investing by L. Rahl (ed.).
52. How Hedge funds increasingly use ASP-based risk management services is
discussed in the article ‘Hedge Funds look to Cultivate Technology Roots’, Risk
Magazine, August 2001. See also the article by A. Favell, ‘To ASP or not to ASP’ in
Risk Technology, October 2001.
53. See the article by A. Favell, ‘To ASP or not to ASP’ for more details about the
Prime Risk application.
54. RiskMetrics and Tremont have recently banded together to form a joint venture.
They announced the release of a ‘transparency product’ in the first half of 2002.
55. Highlights of this work were published in Risk Magazine’s Institutional
Investor Supplement in December 2001 (downloadable from the website
www.cmra.com). A detailed comparison of the systems is available at
www.risksystems.info.
56. See the January 2001 software survey by C. Davidson in Risk Magazine,
January 2001 or the 2000 and 2001 Vendor surveys in the August 2000 and
August 2001 editions, respectively.

.... 204 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 205

CHAPTER 7

Active Risk Management in

Multi-Manager AIS Portfolios

The discussion so far has centred on investment features, risk profiles and risk man-
agement principles of individual strategies (Chapters 2, 3 and 5), the benefits of AIS
in a traditional portfolio (Chapter 4), the generic risks of AIS investments (Chapter
5) and the variety of risk management tools available in today’s investment market
(Chapter 6). It is time to bring these different aspects together and discuss the risk
management principles that govern asset allocation, the monitoring and controlling
process and investment management. In this chapter I describe the elements of
sound AIS portfolio risk management practices starting with a discussion about the
importance of transparency, liquidity, leverage control and investment structure.
The remainder of the chapter is then dedicated to outlining an integrated invest-
ment and risk management approach for multi-manager AIS portfolios. The chapter
concludes with the description of active risk control in the AIS portfolio.

The AIS industry’s current best practices for risk management

As a result of various investment disasters, in particular the failure of Long-Term
Capital Management (LTCM) in September 1998, a lot of effort has in past years
been dedicated to defining sound risk management principles for the AIS industry.
The market crisis of 1998 created a heightened awareness among market participants

.... 205 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 206

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

of the need for better risk analysis and management tools including Monte Carlo
VaR, stress testing, leverage control and liquidity assessment. In April 1999 the
President’s Working Group on Financial Markets published its report entitled
‘Hedge Funds, Leverage and the Lessons from Long-Term Capital Management’.
The report requested that a group of Hedge fund professionals drafts and pub-
lishes a set of sound practices for risk management and internal controls in Hedge
funds. Consequently, five large Hedge fund managers1 developed and published a
document addressing this request.2 The report is a detailed description of AIS risk
management principles and provides a good reference for investors and AIS man-
agers. The reader is strongly recommended to examine the report. To summarize,
the report states that managers must understand the sources of returns and iden-
tify and quantify the types of associated risk. Hedge fund managers take
investment risks in order to earn commensurate returns and they are responsible
for setting, allocating and controlling the risk levels. The report emphasizes that
Hedge fund risk managers should closely monitor the interrelated exposure to
market risk (including asset liquidity), funding liquidity risk, credit risk (debt
investments, counter-party risk) and operational risk. An extended appendix is
dedicated to risk-monitoring practices. Some key risk management issues dis-
cussed are: Monitoring tools (VaR, scenario analysis, stress tests, back testing),
funding risk, counter-party risk and leverage.
In July 1993 the Group of Thirty published a guideline report entitled
‘Derivatives, Practices and Principles’.3 One of the main objectives of the report
was to establish ‘best practices’ in addressing the risks inherent in the use of deriv-
atives. The presented standards provide a description of important elements that
also apply to AIS risk management. Among the suggested principles are:

■ involvement of the highest level of management in determining the scope of

activities in derivatives trading
■ mark to market valuation of derivatives, if possible
■ performance of stress tests and forecasting cash and futures funding needs

.... 206 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 207

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

■ establishment of an independent risk management unit reporting directly to

senior management
■ installation of risk systems capable of measuring, reporting and managing risks
in an accurate and timely manner.

In October 2000 the Investor Risk Committee (IRC) published a set of consensus
guidelines for Hedge fund disclosure entitled ‘What is the Right Level of Disclosure
by Hedge Funds?’4 The IRC consists of institutional investors and individuals run-
ning Hedge fund managers. One goal of the report is to establish an industry-wide
standard for risk measures and calculation procedures. The report elaborates on a
number of risk indicators that investors in different AIS styles could find useful:
VaR, aggregate exposures to asset classes and to geographical regions, stress tests,
cash to equity ratio, tracking error, measures of optionality and key spread relation-
ships. According to the consensus, the three primary investor objectives in seeking
disclosure from managers are risk monitoring (ensuring that managers do not take
on risks beyond allowable investments and represented levels of notional exposures
and leverage), risk aggregation (ensuring the investors’ ability to aggregate risks
across all their investments in order to understand portfolio level implications) and
strategy drift monitoring (ensuring that managers are adhering to the stated invest-
ment strategy or style). The document endeavours to develop a set of disclosure
rules for the AIS industry and introduces four aspects of disclosure: content (type of
disclosure), granularity (the level of detail), frequency (how often disclosure is
given) and time delay (how current the disclosed information is).
The IRC members state that disclosure of position information should be car-
ried out in a way that minimizes ‘the possibility that it could adversely impact the
fund’. They go on to say that ‘full disclosure is not the solution’, expressing the
concern that detailed disclosure will put manager returns in jeopardy. Finally, the
IRC report indicates that the complexity of positions in an AIS portfolio may lead
to ‘operational difficulties associated with processing such vast quantities of data’.
Monthly disclosure is suggested as sufficient disclosure frequency, but a distinc-
tion is made between large and small funds. The discussion within the IRC group
is ongoing and further meetings in 2002 and later aim at elaborating on the
details of recommended disclosure norms for individual strategies.5

.... 207 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 208

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The Highly Leveraged Institutions Working Group (HLIWG), a joint working

group set up by the Basel Committee on Banking Supervision and the
International Organization of Securities Commissions for the purpose of review-
ing banks and securities firms dealing with ‘highly leveraged institutions’, issued a
report in March 2001.6 The report indicates that ‘highly leveraged institutions’
(i.e. Hedge funds and Managed Futures) have become greater in number, but gen-
erally less leveraged since 1998. It further indicates that the willingness of Hedge
funds to provide financial information to investors regarding their trading activi-
ties and risk exposures has improved, but progress has been inconsistent. Many
investors remain all too willing to accept a level of transparency that is insufficient
to enable them to make a full assessment of a manager’s soundness.
The discussion concerning transparency and the necessary disclosure for Hedge
funds is ongoing, but the trend is clear. The AIS industry is confronted with
stronger demand for investment transparency, especially from institutional
investors with fiduciary responsibilities. AIS allocators and portfolio managers
(fund of funds) focus increasingly on achieving transparency (e.g. through a man-
aged account) in order to perform frequent positions monitoring, detailed risk
analysis and continuous manager control.7

Multi-manager portfolios
The complexities of AIS risk management and the recognition of the importance
of diversification across managers have contributed to the rapid growth in AIS
multi-manager funds of funds. The emergence of funds of funds has made AIS
investing accessible to investors who do not have the necessary resources to
address the many challenges of AIS investment. At the same time, not all funds of
funds are created equal. There is a wide variability in the level of expertise and
rigour exercised by fund of funds managers and the dispersion of their returns has
clearly increased in recent years, primarily on the downside.8 One must ensure
that the additional fees associated with a fund of funds are justly earned.
The AIS industry itself is not informational efficient but rather opaque. In
other words, experienced fund of funds managers with a competitive advantage

.... 208 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 209

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

are able to add value through strategy and manager selection. With demand for
AIS exposure growing and the number of Hedge funds increasing the selection of
highest quality managers might become more difficult. The resulting imbalance
between demand for Hedge fund exposure and supply of quality managers is
likely to increase the added value of an experienced fund of funds manager.
In order to perform the complex tasks of strategy sector allocation, manager
due diligence and investment monitoring/risk management, fund of funds man-
agers have to be AIS specialists with an in-depth understanding of strategy sectors
and particular managers’ trading approaches. This includes the specific factors
contributing to risk and return. They must understand how numerous macroeco-
nomic factors influence the performance of the individual strategies. Further, one
of the main benefits of a fund of funds approach, sector and manager diversifica-
tion, can only be achieved if a fund of funds manager has a detailed understanding
of correlations between strategy sectors and asset classes in different market envi-
ronments. A diversified AIS portfolio should hold investments in strategy sectors
that have low correlations to one another and a diversity of return characteristics
in different market environments. Within each sector, risk that is specific to each
manager must be diversified across different managers.
If properly performed, a fund of funds can provide the security created by con-
tinuous monitoring of each manager’s trading activities for which the fund of funds
managers should have the necessary resources and infrastructure. The relationship
between the fund of funds manager and the single hedge fund he invests with must
provide sufficient transparency and liquidity to allow active risk management.
In summary, there is significant added value to an investor from a fund of
funds provided the fund of funds manager:

■ has a complete understanding of the individual strategies and their risks

■ has access to relevant information and the necessary data about AIS managers
(be inside the ‘information loop’ of high-calibre investment talent)
■ performs thorough due diligence of the managers and their respective ‘edges’

.... 209 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 210

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ diversifies among sufficiently different strategy sectors and individual managers

■ implements a system for continuous monitoring of open positions
■ performs active risk management of the investment exposure.

Funds of funds provide the further advantage of better access to more managers
for the investor. The minimum investment for a fund of funds is generally much
lower than for an individual Hedge fund. For a single Hedge fund an investment
of several million dollars is often required, making it very difficult for an individ-
ual to build a diversified portfolio. Further, while individual investors cannot
invest with managers who have closed their investment program to new investors,
fund of funds managers with good relationships in the industry are often able to
invest in strategies even if these are technically closed to new investments.
Pooling assets in a fund of funds provides the possibility of a more transparent
and more liquid investment platform than investing in single Hedge funds directly.
However, today most funds of funds have similar transparency problems to single
Hedge funds (see later). The redemption and subscription process for fund of
funds is usually similar to those of mutual funds, but generally comes with signifi-
cantly longer redemption periods. Monthly or quarterly periods are the norm.
Some (mostly closed-end) investment vehicles try to provide for better liquidity
through listings on an exchange (e.g. in Luxembourg, Ireland or Switzerland) or
trading OTC in a secondary market. In this case, dealing occurs on an intra-day
basis involving designated market makers. Due to the lack of a broad market,
these funds are generally not traded very actively, however.
The expertise and portfolio management skills of fund of funds managers come
at a price. Funds of funds carry an extra layer of fees on top of the usual manager
fees. This leads to lower returns compared to the mean of the individual man-
agers. On a risk-adjusted basis, however, reduced volatility due to diversification
tends to offset the lower absolute return of (see Chapter 4). Typical fees for funds
of funds include a 1–2% management fee and performance fee in the range of
10%, plus custodian and administration fees of about 0.5%.9 Some fund of funds
managers are able to mitigate the higher fees charged on the fund of funds level
by obtaining fee rebates from the trading managers, taking advantage of the
higher allocations they can make to single managers.

.... 210 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 211

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

Fund of funds managers usually do not invest across all possible strategy sec-
tors. Instead, they take a certain view on the attractiveness of individual strategies.
Some AIS allocators go as far as developing their own views on the future market
environment and structure their portfolio accordingly. A portfolio manager pre-
dicting rising equity markets, for example, will have a disproportionally high
investment in net long equity strategies (such as Long/Short Equity). There is dis-
agreement as to whether ‘strategy timing’, i.e. the systematic or discretionary
switching between different strategies, is a useful investment philosophy. Investors
should be aware of the inherent dangers underlying such an approach. Erroneous
market forecasts or the careless assumption that performance drivers will be the
same in the future as they have been in the past are among the main sources of
underperformance in AIS portfolios. During the equity bull market of the late
1990s, many funds of funds were heavily invested in Long/Short Equity strategies.
The strong market had enabled this strategy sector to capture impressive returns,
and many AIS portfolio managers believed the bull market was going to continue.
When Long/Short Equity strategies showed much weaker performance in falling
equity markets during the later part of 2000 and 2001, the over-allocation to
Long/Short Equity strategies caused many funds of funds to perform significantly
below their expectations (see Chapter 4 for more details).
A much more robust approach to generating steady portfolio returns is diversi-
fying systematically across those strategies that, by their nature, perform
differently in various market environments (see Chapter 5 for a more detailed
discussion on portfolio diversification). The consideration of ‘conditional
correlations’ (see Figures 4.4 and 4.5) is useful for this purpose. A good portfolio
is diversified across different performance and risk factors and not necessarily
across different managers within one strategy sector only.

The issue of transparency in AIS

Unfortunately, most Hedge funds confront direct investors as well as AIS portfolio
managers/allocators with numerous liquidity and transparency issues. The investor
faces lengthy redemption periods and is often not provided with the necessary

.... 211 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 212

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

degree of disclosure regarding the Hedge fund manager’s activities. This is one of
the most widely perceived disadvantages of AIS investments. The resulting risks are
severe and could be better controlled if regular disclosure of relevant information
about risk and trading exposure were provided. Furthermore, transparency can pro-
vide the peace of mind necessary for many new investors to enter the field of AIS.
The story of LTCM serves as an illustration. In September 1998 the failure of the
Hedge fund Long-Term Capital Management (LTCM) is said nearly to have brought
down the world financial system. The losses LCTM incurred were so large that the
Federal Reserve Bank took the unprecedented step of initiating the bailout of a pri-
vate investment vehicle, as the fear spread that forced liquidation would cause
global financial turmoil. Something very fundamental had gone wrong.10
The fund had earlier made very handsome returns following its core strategy,
which can be referred to as ‘Convergence Arbitrage’ and fell into the class of
Fixed Income Arbitrage. The managers at LTCM had placed a large amount of
money in trades involving the converging interest rates within the European
Monetary System in the wake of the euro, the European currency unit. The most
prominent example was buying Italian Government Bond (BTP) Futures and sell-
ing short German Bund contracts. The strategy was clearly defined and paid off
handsomely. Other examples of LTCM trades involved US, European and
Japanese yield curve and government bond swap spreads.
At the end of 1997 the fund back paid a significant amount of money to investors,
about $2.7 billion, with the NAV of the fund being at around $7.5 billion. The origi-
nal core strategy had clearly lost most of its edge. The yield spread between Italian
and German 10-year government bonds had narrowed from about 550bp in early
1993 to only about 20bp by the end of 1997. The fund managers were looking for
other opportunities and correspondingly found themselves engaged in a much wider
spectrum of strategies, e.g. Merger Arbitrage, Mortgage-Backed Securities Arbitrage
and Emerging Markets debt. Furthermore, in order to continue generating the attrac-
tive returns of the past, the fund increased its leverage substantially, from about 19 at
the end of 1997 to about 30 in early 1998 and 42 in the summer of 1998. Neither the
style drifts nor the increase in leverage had ever been communicated to investors. On
September 23, 1998 (the day of the bailout), the fund had lost 92% of its asset year to
date and the leverage had gone up to about 120.

.... 212 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 213

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

LTCM clearly shifted its investment practice in the course of the months before
the disaster. Investors had no knowledge of the strategy that LTCM was following
at the time. The excessive leverage employed by the fund managers (partly a result
of unrealised losses) remained undetected until the fund had already lost most of
its capital. LTCM’s investment process and the lack of disclosure to investors vio-
lated the core principles of diligent AIS investing, namely strategy understanding,
knowledge about the strategic edge of the manager and continuous monitoring
and risk management of the investment exposure.
Disclosure of the LTCM investment activities would quickly have made
investors aware of the style drifts and ‘bets’ performed by LTCM’s managers. It
would have further allowed the investor to understand the core strategy in more
detail, its requirements for earning its return and its behaviour in different
market circumstances (especially the unfavourable ones). This understanding
would likely have led to the detection of previously unknown (or unrecognized)
risks of the investment strategy, in this case, the disappearance of the strategy’s
core edge. Thorough understanding of market conditions that would result in
the edge of the manager disappearing would have enabled investors to exit the
strategy in time. Finally, investors would likely have reacted to the excessive
leverage employed by LTCM. Of course, all this would have required the
investor to look beyond the returns of the past and to be proactive in assessing
the strategy’s future outlook.
As already discussed, fund of funds managers specialize in finding the most
promising strategies and managers. Thereby, they have to diversify two main
sources of AIS risk: (1) the ‘market risk’ of each strategy, which is the a priori
known risk of potential market behaviours affecting the strategy (and its sector)
adversely and (2) the specific ‘credit risk’ (or manager risk) of each individual
Hedge fund or Managed Futures fund, which is the risk related to the specific
manager experiencing a large loss. Investment transparency is necessary for the
appropriate management of both, credit (manager) as well as market risk.
There are different views on how to achieve sufficient protection against man-
ager risk. Some AIS allocators follow the approach of selecting 50 to 100 different
managers, thus limiting the impact of a single trading advisor ‘blow up’ in their

.... 213 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 214

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

portfolio, and thereby achieve diversification comparable to managing a corporate

bond portfolio. This results in more required manager due diligences and less time
spent per manager and can quickly lead to over-diversification, which often results
in mediocre performance. A well-diversified portfolio can be built with 12 to 15
managers.11 Transparency and more detailed knowledge of the individual man-
agers and their strategies decreases manager risk and thus the number of managers
needed for its appropriate diversifications:

■ Access to regular information about the manager’s trading activities allows

style changes, undesired bets by the manager and leverage increases to be
quickly recognized and addressed.
■ Transparency allows leverage to be controlled.
■ Transparency enables the investor to examine all agreements the Hedge fund
is a party to such as prime brokerage agreement, fee agreements, past audit
reports, corporate registration and others.
■ Transparency allows for following a strategy in ‘real time’ and thus leads the
AIS allocator to a further understanding.
■ Transparency significantly decreases the probability of fraud.12

The market risk is best dealt with through appropriate diversification among strategy
sectors. This requires precise knowledge of the various strategy sectors including their
particular favourable and unfavourable market environments. An understanding of
how past returns were generated is imperative to successful asset allocation (this
includes an assessment of whether they were generated at all; often returns that look
too good to be true are not true). A diversified AIS portfolio contains strategies which
have their strengths and weaknesses in very different market environments (see
Chapter 5). Without transparency and the appropriate knowledge of the strategy,
diversification of a multi-manager portfolio remains largely a guessing game with the
tendency to allocate most money to the ‘stars of the past’. Transparency is important
for performance evaluation:
■ Transparency provides the possibility of detecting previously unknown (or
unrecognized) risks in the investment strategy. While one might obtain good
insights from past performance behaviour, certain risk factors may have yet to
be experienced.

.... 214 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 215

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

■ Transparency provides support for the challenging task of performance

measurement. Reported NAV often does not take into account the liquidity of
open positions, especially when a position is large compared to daily trading
volume of the asset or the asset is not actively traded at all (e.g. in Regulation
D13 and Distressed Debt strategies). In these cases, asset values remain
estimates until gains or losses are realized. Knowing the nature of illiquid
positions enables the investor to perform his own assessment of the manager’s
performance in consideration of the liquidity of open positions.

Despite growing investor awareness of the importance of transparency, there

remains a surprising degree of resistance against such in the AIS industry. The
‘black box’ investing approach is still widespread. Direct investors in Hedge
funds, AIS allocators and fund of funds investors often do not receive the appro-
priate amount of information or have the necessary knowledge of manager
activities to make educated investment choices. The following three arguments are
frequently used against transparency to the level of full position disclosure.
Trading managers often bring up the first two arguments and many AIS allocators
raise the third.

1 Confidential position information reaches the market place, potentially

causing the strategy to: (a) lose its edge, if more players adopt the strategy;
and (b) be actively traded against by other market participants.
2 Investors lack the skills necessary to evaluate the massive amount of
information provided by daily positions. This could lead to overload and/or
false comfort.
3 The best Hedge fund managers do not disclose their positions. The request for
transparency will thus eliminate the opportunity to invest in the most
promising investment strategies.

For the first argument, one must consider who actually poses a threat to AIS man-
agers. The threat of being copied or actively being traded against comes largely from
the dealer community and the investment banks’ proprietary trading desks rather than
from AIS allocators or direct investors. Once Hedge funds know who their investors

.... 215 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 216

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

are and what their intention with the disclosed information is, they can set up appro-
priate confidentiality agreements which keep the investor from proliferating any
relevant information. The necessary transparency can be provided to investors, while
details about the manager’s trading positions remain undisclosed to the wider public.
The second argument neglects the increasing expertise and capacity of fund of
funds managers. In combination with the appropriate level of strategy understand-
ing, information about the details of the trading activities is very useful to the
investor and AIS allocator. With the advent of information technology and power-
ful computers, the compilation of large amounts of data has become an easy
undertaking for professional investors and portfolio managers. Positions and
transactions can be downloaded and analyzed very efficiently and a large variety
of tools and risk management software packages are today commercially available
(see Chapter 6 for a more detailed discussion) to support AIS portfolio managers
in the task of monitoring managers’ trading behaviour and assumed risk levels.
Fund of funds managers often raise the third argument and claim that the best
managers are unwilling to provide transparency or insight into their trading
approach. They argue that AIS allocators that require transparency necessarily
have to invest with lower performing managers. There are indeed some managers
with excellent performance track records who refuse to provide transparency to
investors, but it is incorrect to link attractive returns with lack of transparency.
The belief that the best managers are necessarily non-transparent and that AIS
portfolio managers focusing on transparency are therefore left with ‘second tier’
managers finds no empirical support. Many high-quality managers with excellent
track records are willing to offer the desired transparency, if asked or required
(for an investment) to do so. Good performance has nothing to do with a lack of
transparency. Inversely, managers who refuse to provide transparency of their
investment activities are not necessarily the most preferable for the investor. It
should be noted that the more openly a trading manager discloses his strategy, the
more likely it is that he is able to present a clear edge. The lack of a manager’s
willingness to provide transparency can correspond to his inability to illustrate
why his approach makes money. A manager who is unwilling to explain his edge
may not have one and therefore hides behind a ‘black box’ approach.

.... 216 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 217

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

Many investors continue to regard AIS as an industry where returns are gener-
ated by mysterious means and judge successful managers exclusively on the basis
of their stellar past returns. Examples of secretive, non-transparent and, for cer-
tain periods of time, very successful strategies are LTCM, Quantum Fund, Tiger
and Niederhoffer, all of which failed spectacularly in the end. Investors failed to
understand that high past returns were connected to high risks, which eventually
led to failure.
The incorrect assumption that the best performing managers must operate in
secret is linked to persistent misperceptions about AIS. Many investors believe that
AIS returns are generated through the identification of unrecognized market ‘ineffi-
ciencies’. The reality of AIS investing is that most strategies systematically earn
premiums in return for assuming certain risks (see Chapter 2). Most managers follow
systematic investment strategies that, not surprisingly, perform better in certain
market environments than in others and that bear a range of risks. Transparency is
needed to allow for an adequate assessment of risks and returns. With some effort
most strategies are actually not very difficult to understand. As with all investing, past
performance is an insufficient indicator of a strategy’s future performance potential.
Investors must learn to look beyond past return and instead look at how and in
what market environment returns have been achieved. Even the most brilliant
investors of the past are not protected against losses and drawdowns and the con-
ditions for success are constantly changing. Recent studies have shown that there
is little convincing evidence that above average performance repeats itself in a way
that can be exploited by investors.14 The best way to predict future performance is
to understand the investment process, the underlying strategy and the manager’s
real ‘edge’ and to assess these factors in the context of the current and possible
future market environments. Strategy assessment is an ongoing process and
includes the effort of following the manager’s activities on a regular basis. It is not
enough to wait for the monthly arrival of NAV and then begin analyzing facts and
asking questions.
A note on the relationship between size and performance of a manager’s trad-
ing program: A good track record leads to asset growth and the largest players in
the AIS industry have shown impressive performance in the past. But many

.... 217 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 218

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

investors are not aware of the ‘survivorship bias’ inherent in relating size to future
performance outlooks. The following (extreme) example illustrates this:

EXAMPLE
Take 10,000 different ‘random’ strategies (i.e. positions are taken on a com-
pletely random basis) and let them run for a few years. Good performance is
rewarded with asset inflow, bad performance with asset outflow. Most of the
strategies have to close their programs due to sluggish performance. But
among the many strategies a few will turn out to be successful by pure coin-
cidence. These are the ones that survive and, after a few years, they will have
grown to be large.15 This does not mean, of course, that they are more trust-
worthy than the majority that failed.
Recent reports have actually shown that small and young programs have the best
future performance.16 With the ‘institutionalization’ of the AIS industry more and
more managers will be forced to disclose the details of their trading activities to
investors. If one considers the monitoring of investment activities inside financial
institutions such as banks, broker–dealers, corporate finance departments of large
firms, insurance companies, endowments, pension funds and foundations, it seems
difficult to imagine that external managers would be allowed to pursue com-
pletely unmonitored and largely unregulated investment activities. To make
matters worse, these strategies are exposed to a much larger variety of risk factors
and can be made riskier at the discretion of the manager.
There is much ongoing discussion about what the right level of disclosure to AIS
investors is. The discussion centres on the question whether aggregated ‘risk infor-
mation’ is sufficient for monitoring and risk management purposes or whether full
disclosure of all positions is needed. The belief that AIS risk can be adequately
monitored without obtaining the underlying positions is widespread. I strongly dis-
agree with this view. Along with the investor there is another party that is exposed
to the risk of financial losses as a consequence of a manager’s trading activities.
This is the prime broker as a lender of securities and provider of leverage. If suffi-

.... 218 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 219

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

cient risk and exposure monitoring without disclosure of positions was possible,
the industry would likely see according ‘best practices’ by the prime brokers. The
prime brokers would then monitor their exposures to Hedge funds based on
‘aggregated risk information’ without bothering with the tedious job of identifying
single positions. But, in reality, prime brokers have full insight into positions on a
continuous basis and so should the sophisticated investor. A high level of trans-
parency provides the very foundation for effective risk management. Without it,
reliable performance and risk measurement remain impossible.
The necessity for position disclosure and usefulness of full transparency varies
among strategies. For some strategies, disclosure of each individual position is not
always absolutely necessary for integrating exposure information into a risk man-
agement system. Neither do direct investors and allocators always need
transparency on a daily basis for each strategy to manage and measure risk effec-
tively. But in most cases and for most strategies frequent and detailed position
information remains the best way to ensure that the necessary information is
available to the portfolio and risk manager to fulfil his monitoring task. This
information should be disclosed as regularly and frequently as appropriate to the
strategy and, in many cases, daily disclosure is the appropriate frequency. It is the
fund of funds manager’s responsibility to define the necessary amount of position
information needed.
Transparency may take a variety of forms, from regular conversations with the
managers (the weakest form) to obtaining a daily download of all positions from the
manager’s prime broker (the strongest form). Model-based systematic strategies are
easier to monitor and understand on a daily basis than discretionary Long/Short
Equity, Macro and Short Selling strategies, where it is more difficult to follow the
rationale of positions on a standalone basis, i.e. without further information pro-
vided by the manager. For Relative Value strategies (Fixed Income Arbitrage,
Convertible Arbitrage, Equity Market Neutral) transparency is necessary, as leverage
and potential style drift are important risk factors that have to be controlled. This
also applies to Event Driven strategies (Merger Arbitrage, Distressed Securities,
Regulation D), but here a deeper analysis of positions in light of the underlying
strategy is necessary, which requires additional resources and capacities on the side

.... 219 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 220

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

of the portfolio manager. The optimal frequency of disclosure is also related to the
liquidity of traded instruments. For Distressed Securities and Regulation D strate-
gies, weekly or monthly disclosures can be sufficient, while for most Managed
Futures strategies, daily transparency is most appropriate.
To summarize then, the ‘transparency approach’ of funds of funds is characterized
by the following:

■ detailed understanding of the return sources and risks of all strategies

■ appropriate disclosure of positions by each manager
■ continuous position monitoring and risk measurement
■ high asset liquidity (note that most, but not all, AIS managers trade
instruments that are traded on public exchanges offering daily liquidity)
■ active risk management on a continuous basis.

Liquidity of AIS
Liquidity issues are among the most complex in AIS investing. A good understand-
ing of liquidity is critical to the investment process and risk management for AIS
portfolios and the lack thereof imposes risks that in the past have often received
insufficient attention. ‘Liquidity’ has two different aspects: the ability to sell an
investment without a price impact (instrument liquidity) and the availability of
financing for leverage (funding liquidity). The AIS allocator should understand the
liquidity of each of the instruments a strategy invests in and the sources and reliabil-
ity of leverage financing. An important question to ask is the ‘90/10’ question: ‘How
long would it take under normal market circumstances to liquidate 90% of the posi-
tions without a price impact?’ Further, it is important to note that the liquidity of
investments can change dramatically in times of market turbulence. Liquidity tends
to evaporate when most needed, i.e. just when investors most want it, it disappears.
An extreme example of previously unknown liquidity risk in a ‘flight to quality’
event led to the failure of LTCM described earlier.

.... 220 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 221

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

In order to avoid any asset–liability mismatch, there must be a realistic relation-

ship between the liquidity of the traded instruments and the fund’s redemption
provisions. This applies equally to the redemptions agreements of funds of funds
with the individual managers, which have to be in line with the fund of funds’ own
redemption policies. Note that most redemption policies can be overridden in
extreme market environments, as many managers include sections in their offering
memorandums that give them the right to suspend redemptions under certain
vaguely defined scenarios of market turmoil, regardless of the normal liquidity
policies of their funds. Obviously, the best liquidity indicator for a strategy is the
market liquidity of the traded instruments. Managed Futures, large cap Long/Short
Equity, Macro, Market Timing and Risk Arbitrage strategies involve liquid
exchange traded instruments, as do most Arbitrage and Relative Value strategies
(with some exceptions for Convertible and Fixed Income Arbitrage strategies).
Convertible bonds are usually traded OTC as are swaps and other fixed income
derivatives, but the liquidity of these instruments is often sufficiently high. On the
other side of the spectrum are Distressed Debt and Regulation D strategies, which
are generally extremely illiquid.
An important factor that requires an allocator’s attention when considering liq-
uidity risk is fund size. Large funds, even those operating in very liquid markets,
can easily encounter problems when shifting positions. There are excellent man-
agers with smaller investment programs who are able to provide better liquidity
simply because their position shifts do not impact market prices. Too often
investors and funds of funds avoid these trading advisors simply as a result of
their small size, despite research indicating that small and young programs tend to
show superior performances (see above).
The benefits of transparency in an AIS portfolio cannot be fully realized with-
out the individual managers providing sufficiently short redemption periods. If
redemption periods do not correspond to the frequency of disclosure, the fund of
funds manager is prevented from responding to time sensitive portfolio informa-
tion with immediate action (e.g. de-leveraging).
Many strategies employ leverage and thus rely on external funding. Funding
liquidity risk is largely determined by the prime brokers’ ‘haircut’ policies, which

.... 221 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 222

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

specify the leverage and margin requirements for a particular strategy. The AIS
allocator should consider the ‘haircut sensitivity’, i.e. the sensitivity to changes in
the haircut policy of the broker, when examining a strategy’s liquidity risk.
Managers can become victims of leverage in situations of financial turmoil, when
prime brokers refuse to continue to provide financing leverage because of the
increased risk of positions held. This can lead to forced liquidation of the posi-
tions at inopportune times and thus create large investment losses. The manager’s
relationship with his prime broker is therefore an important element in the imple-
mentation of his trading strategy.

Managed accounts
The most effective way to address the need for both liquidity and transparency is
via a ‘managed account’. Managed accounts can provide the AIS allocator with full
disclosure of the managers’ trading activities together with daily liquidity (assuming
the underlying instruments are sufficiently liquid). This creates the optimal basis
for proactive risk management. Managed accounts also drastically reduce the risk
of fraud. Figure 7.1 illustrates the set up of a ‘fund of managed accounts’. The indi-
vidual managers (in this context the word ‘trading advisor’ is more appropriate)
have no direct access to the investor’s money, which remains with the prime broker
chosen by the fund of funds manager. The trading advisors have special authority
to execute trades on behalf of the multi-manager fund in accounts that are specifi-
cally set up for that purpose. The fund of funds manager can download the
positions of each manager from the prime broker and monitor trading activities on
a daily basis. This results in maximal transparency. The fund of funds manager (or
better, the ‘fund of managed account manager’) is in a position to reallocate invest-
ments with each manager literally on a daily basis (assuming the underlying
investments are equally liquid). Ideally, he can also terminate the trading advisor’s
investment activities at any time. Finally, the fund of funds manager is in a position
to select all involved parties and transaction counter-parties including the prime
broker, custodian and auditor.

.... 222 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 223

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

Prime Broker and

other Executing Daily download:
Brokers Transactions and
Various Traders Positions
Bank as
Payment Agent

Bank’s Fund
Trader A Administration
• Daily NAV calculation
• Accounting

• Asset Allocation
• Monitoring
• Risk Management

FIGURE 7.1 ■ Structure of a ‘fund of managed accounts’

With the institutionalization of AIS and the fiduciary role that many funds of funds
must assume the popularity of managed accounts is likely to increase in the future.
However, a managed account is not always possible or practical. Not all AIS man-
agers agree to exercise their strategy via a managed account. Some investors are
constrained for tax or regulatory reasons and have to invest via the manager’s funds.
Managed accounts require a significant minimum investment ranging from $2 to
$25 million.

Fund Redemption Policies for multi-manager investment vehicles

The redemption period of a multi-manager fund has to be in line with the liquidity
offered by the individual AIS managers that the fund invests with. High liquidity on
the level of a fund of funds obviously places constraints on the universe of strategies
that can be included in the portfolio. Strategies whose returns are based on ineffi-
ciencies in smaller, less liquid markets and those that earn liquidity premiums (e.g.
Distessed Debt, Reg D) cannot be part of a fund of funds with short redemption
periods. But the AIS universe offers a large number of strategies that trade very
liquid instruments.

.... 223 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 224

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Ultimately, it is the preference of the investor that drives the requested liquidity
and return profile of a fund of funds. Some investors prefer highly liquid invest-
ments, while others are willing to invest in lower liquidity instruments in pursuit
of higher returns. With the current growth rate of the AIS industry and the spec-
trum of investors becoming more diverse, it is likely that the request for liquid
multi-manager AIS products will increase. But most Hedge fund managers – if not
trading in a managed account – offer their trading strategy to investors by setting
up their own fund, in which investors and AIS allocators face rather long redemp-
tion periods ranging from about one to six months. Four investment structures
can be broadly distinguished for multi-manager AIS products: open-ended funds;
closed-end funds; structured notes; and managed accounts (see also Chapter 5 for
the details about their legal set-up). With the exception of managed accounts, all
these structures confront those investors who search for high liquidity (i.e. short
redemption periods) with problems.
Open-ended funds of funds are among the most common structures, with
redemption periods ranging from one month to six months. The fund of funds
manager invests either directly into a manager’s fund or on the basis of a managed
account. Besides monthly or longer redemption periods, open-ended funds often
have lengthy settlement processes, increasing the time span between redemption
and receipt of the money beyond the official notice period. The redemption
process involves a number of tasks involving different parties: Net Asset Value cal-
culation by the administrator, money transfer by the custodian of the fund and
internal booking of the money at the investor’s bank.
Closed-end vehicles, wrapped as investment companies and listed on
exchanges, were established in part as an attempt to improve liquidity for the
investor. The fund manager of a closed-end investment company usually invests
directly into the Hedge fund manager’s funds. There are neither settlement prob-
lems nor redemption periods as with an open-ended fund, because the investment,
i.e. the investment company’s stock, can be bought and sold daily in the open
market. However, other problems render these closed-ended products unsuitable
for most investors. Due to the lack of a broad market, the stocks of the invest-
ment companies are not traded very actively and large investments cannot be sold
in timely fashion without a large impact on prices. This usually leads to a signifi-

.... 224 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 225

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

cantly lower market value compared to the NAV (discounts of up to 30% are fre-
quently observed). The volatility of the discount often greatly exceeds the NAV
volatility, which makes an investment in such an instrument very unattractive.
Structured notes are often designed to fulfil specific investment needs such as capital
protection, regulatory specifications, legal constraints and tax protection. Structured
notes usually have a defined investment period and cannot be redeemed for full value
before maturity. They are generally open ended and often traded in a secondary
market provided by the issuer. The issuer defines certain buyback terms, which include
significant discounts to NAV. The fund manager invests either directly into a manager’s
fund or on a managed account basis. In cases where the issuer provides principal pro-
tection, managed accounts are used for reasons of risk control. Structured notes usually
come with higher fees compared to other investment vehicles.
The fourth investment structure – based on a managed account – offers a solu-
tion to the liquidity limitations of other structures. The liquidity of investments
through managed accounts is limited only by the nature of the underlying securi-
ties, not by managers’ redemption policies or other factors. As long as there is a
ready market for the investments (as is the case with most securities in AIS funds,
the exceptions being mostly Distressed Debt and Regulation D), the positions can
be liquidated or adjusted on a daily basis. This, in combination with daily trans-
parency of positions and transactions, allows for real-time proactive, rather than
reactive, risk management. ‘Fund of managed accounts’ which pass the benefits of
short redemption periods and full transparency on to the investors are not yet
widely offered. But this structure raises a simple question: ‘Why can there not be
a multi-manager AIS fund with daily liquidity based on NAV that is as easy to buy
and sell as a traditional mutual stock fund?’

Leverage
Whereas leverage for companies is a function of balance sheet exposure,17 for AIS
investments, leverage describes the financing required to achieve a desired (gross)
exposure level that is in excess of net asset value (NAV). Many strategies achieve
leverage through external funding. For Futures strategies, leverage can be

.... 225 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 226

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

achieved without additional financing, as only a fraction of the exposure has to be

placed as margin. Note that US (SEC regulated) funds are limited in the amount
of leverage they can carry by Regulation T (maximum leverage 2:1).
Leverage is not always undesirable and, in certain instances, can be very helpful
for achieving the desired risk–return profile of a strategy. Leverage on a stand-
alone basis is not a very useful measure of risk and a high leverage factor does not
necessarily imply unacceptably high risk. It is important to understand the impli-
cations of leverage (rather than stick to rigid policies such as investing in strategies
with little or no leverage only), because of its impact on other risks such as
market, credit and liquidity risk.
In order to understand a single manager’s risk level as well as the risk in an AIS
portfolio, leverage has to be assessed diligently. One should distinguish between
gross and net leverage. Gross leverage is defined by (long positions + absolute
value of short positions)/(portfolio equity) and net leverage is (long positions –
absolute value of short positions)/(portfolio equity). The level of acceptable lever-
age depends strongly on the strategy. Gross and net leverage provide some insight
into hedge exposure when comparing similar strategies, but it is difficult to com-
pare leverage levels across strategies. For Fixed Income Arbitrage strategies,
leverage factors of higher than ten are not unusual, while Distressed Securities or
Regulation D strategies usually do not employ leverage at all. Defining a maxi-
mum accepted level of net and gross leverage for each strategy and monitoring
these limits closely is an important element of portfolio risk management. If a
manager exceeds predefined leverage limits, this should raise a red flag.
Leverage increases volatility by the employed leverage factor. But leverage also
bears further risks linked to liquidity and funding risk and it is important to ana-
lyze instrument liquidity as well as financing risk for each leveraged strategy. Asset
liquidity, leverage and financing risk are interrelated and cannot be seen in isola-
tion. When instrument liquidity evaporates (e.g. in market stress situations),
counter-parties, mostly prime brokers as providers of leverage financing, can
quickly withdraw funding. Prime brokers have policies as to how much a certain
type of strategy can borrow and the margins that have to be provided to each
strategy. These policies can be subject to sudden change. The manager’s relation-

.... 226 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 227

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

ship with prime brokers is therefore very important. It is essential to know loan
and collateral requirements, margin policies, the intricacies of the stock loan
market (e.g. how often a manager is subject to stock loan recalls) and how prime
brokers and other counter-parties behave in times of market stress.

Fee structure and other key administrative considerations

The fee scheme and other administrative aspects of the set-up of an investment
structure have a significant impact on its ability to meet performance goals. The
investment and allocation strategy should be supplemented with an efficient and
cost-sensitive administrative structure. Well-run AIS portfolios can suffer
mediocre performance when the costs of the investment activities are too high.
Funds can face problems that could have been avoided had fund set-up and the
selection of counter-parties been approached more carefully. The choices of the
prime broker, fund’s jurisdiction, legal counsel, administrator, auditor and
accountant have to be undertaken diligently by the fund of funds manager.

Fees
The fee structure should be chosen such that it balances fair compensation to the
investment managers (fund of funds as well as AIS managers) with the interests of
the investor. Fees that are too high can cause the net performance to the investor
to be below expectations. Fund of funds fees usually range between 1 and 2%
fixed fee and about 10% performance fee (fixed management fees above 3% can
be observed in some rare cases, which clearly exceed the range of a reasonable fee
structure). Besides the compensation structure for fund of funds managers (AIS
allocators), other fees borne by the fund include administration fees, auditing fees,
fees for fund set-up and possible sales fees. The AIS allocator should be sensitive
to the level of fees charged by each AIS manager and should actively negotiate
these fees, the spectrum of which is wide, ranging from 0 to 3% fixed and 10 to
30% performance fee.

.... 227 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 228

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Prime broker
Prime brokerage describes the function of providing custodian services, clearing, bro-
kerage, financing (providing leverage), reporting and stock lending to AIS managers.
These are essential services to any AIS fund (single and multi-manager) and the
chosen prime broker should have solid experience and sufficient competence in the
core areas of the business. Technical competence is equally important, as the internet
has become one of the most important communication tools between prime brokers
and their clients. Hedge funds and Managed Futures have become an important client
segment for investment banks and brokers, and prime brokers work often in close
relationship with the AIS managers as providers of a broader range of services.
Examples are economic research and risk analysis tools for complex multi-instrument
strategies. Furthermore, prime brokers are increasingly active as providers of Hedge
fund start-up assistance, distributors of the latest industry news and supporters of
managers’ marketing efforts. Many investors fail to realize that the selection of a
prime broker is one of the most important choices for the AIS manager. This applies
to the single Hedge fund as well as to the multi-manager fund18 (for the latter a solid
prime broker relationship is especially important, if managed accounts are employed).
Important considerations besides the prime broker’s ability to deliver high-quality
services in these areas are the global firm’s commitment to its brokerage unit (i.e. its
willingness to invest in technology and people), the way the firm handles conflicts of
interest with other parts of the firm, especially the proprietary trading units (set-up of
‘Chinese walls’), its connections in the AIS industry and its flexibility in accommodat-
ing the individual fund’s structuring and support needs. Finally, AIS investors and
allocators should be aware of the fees charged by prime brokers for their services. The
costs of prime brokerage services are often hidden in different layers of execution,
clearing and service fees, which can, in most cases, be negotiated.

Administrator
Most AIS funds are registered offshore and thus work with offshore fund adminis-
trators. An efficient administrator is critical, because he represents the investment
fund to the investors. He is responsible for independent preparation of the fund’s
books and records and carries out other important back-office operations such as

.... 228 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 229

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

NAV calculation, determination of fee attributions and maintenance of capital

account balances. Bad fund administration is more than just a headache, as mistakes
and significant time delays in NAV calculations or incorrect accounting can be very
harmful to the fund. The administrator should be experienced in the duty of admin-
istrating a Hedge fund or fund of funds, use appropriate technology and be globally
connected to provide the necessary services. He should be responsive to the fund’s
and investors’ needs, accurate in his calculations and accommodating to the chosen
investment structure. Further, he should have systems in places that are sophisti-
cated enough to deal with the pricing of exotic investments and the calculation of
complex performance fee structures. Important issues for the selection of the
administrator are the experience of the staff, the IT and telecommunication struc-
ture, global presence and the fee structure. Administrators’ fees are usually in the
range of 0.1 to 0.2% of assets under management, often with a minimum base fee.

The investment and risk management process for

AIS multi-manager portfolios
The performance characteristics of AIS are extremely diverse. While some strate-
gies are driven by similar factors as traditional bonds and stock performance, such
as corporate earnings, yield curves, credit spreads (they are often referred to as
‘return enhancers’), others are little affected by these factors and depend on other
market parameters (these are regarded as ‘diversifiers’).
The goal of investing in AIS is to earn an appropriate return for being exposed to
certain known and accepted risks. The different steps towards this goal include the
selection of the right strategy sectors and managers and the implementation of thor-
ough post-investment monitoring and risk management. In the remainder of this
chapter, I will discuss the core elements of the investment and risk management
process for AIS multi-manager portfolios. As described in Chapters 2, 3 and 5, Hedge
fund and Managed Futures strategies generally display more complex risk profiles
than traditional investments. Figure 7.2 summarizes these various risks of AIS.
Generally, AIS risk management should: (a) be integrated into the strategy and man-
ager allocation process prior to investment; and (b) include continuous monitoring
post investment. Effective AIS risk management is proactive rather than reactive and

.... 229 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 230

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Fraud
Operational Risk Legal
Periodical Professional
Risk manager visit Only registered external legal Risk
due diligence managers/ advice/expertise
on-site daily monitoring
Model Diversification
Ongoing Manager Selection Process Correlation
Risk Risk
model and Monitoring and volatility
analysis analysis
Internal Risk

External Risk
Corporate Liquidity
Event Risk Portfolio Allocation and Diversification Risk

Monitoring of Monitoring
Ratings/Quality market activity/
volatility

Exposure/VaR Models/ Market Credit Credit risk models

Monte Carlo/Stress Tests/ Risk Risk (KMV/CreditMetrics/
Leverage Control/ CreditRisk+
Correlation/Volatility

FIGURE 7.2 ■ The different sources of risk and how they can be approached

entails quantitative and qualitative analysis as well as a good amount of common sense.
Figure 7.2 distinguishes between ‘internal risks’ (or ‘manager risks’), which mainly
need to be addressed during the manager due diligence process and ‘external risks’
(market risks), which have to be diversified through appropriate (pre-investment) strat-
egy sector selection and then monitored regularly after investment initiation.
The most important features of both the investment and risk management
process for AIS multi-manager portfolios are illustrated in Figure 7.3 and can be
summarized by three core elements.
1 Strategy sector selection and allocation: Identification and selection of
appropriate AIS sectors.
2 Manager evaluation: Thorough analysis and due diligence for individual
trading managers.
3 Investment monitoring and control: Continuous supervision of the trading
managers’ activities, leverage and risk levels, taking corrective action when
necessary.

.... 230 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 231

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

I. Strategy allocation:
Sector portfolio set is key performance driver
(goal: High degree of diversification)

II. Manager evaluation: III. Continuous monitoring:

Qualitative and quantitative due diligence P&L and risk management
(‘pre-investment risk management’) (dynamic understanding of the strategy,
risk control)

FIGURE 7.3 ■ The three key dimensions of AIS investing

Top down strategy sector analysis and bottom up manager evaluation, although
principally pre-investment activities, must be periodically reassessed after the
investment initiation. As is the case for equity investments, sector allocation is the
most important determinant of performance in an AIS portfolio and should pre-
cede the selection of individual managers.19
‘Post-investment risk management’ is a frequently neglected part of the AIS
investment process and requires particular emphasis. Monitoring and risk meas-
urement are prerequisites for effective risk management. Armed with a
quantitative analysis of risk, the risk manager should: (a) continuously assess
whether the general risk level in the portfolio is too high or the losses under cer-
tain scenarios potentially too great; and (b) evaluate whether single managers are
operating outside their predefined parameters (which can, for example, be due to
style drifts, excess leverage or lack of diversification). The risk manager must take
appropriate steps to separate risks that are accepted as a consequence of the trad-
ing approach, from those that are not related to the strategy or accepted by the
investment profile. Appropriate actions range from discussing problems with man-
agers by telephone to reducing allocations or terminating the relationship with a
manager altogether. It is important to note that investment monitoring and active

.... 231 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 232

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

risk control do not mean interfering with the managers’ core strategy, but rather
monitoring the investment relative to the agreed investment principles and risk
parameters and taking corrective action, if deemed necessary.

Strategy sector selection (top down approach)

The risk and return properties of AIS portfolios are very sensitive to sector allocation
and the right selection of strategies is the key performance driver for a multi-manager
AIS portfolio. The goal of the sector allocation is to select those risk factors and levels
that match the investor’s risk profile and to reach effective diversification among
them. Diversification is the fundamental building block for portfolio sector allocation
and risk management. Effective diversification means that the targeted performance
can be reached in a variety of different market environments. Important determinants
of diversification are asset and strategy correlations20 (see Chapter 5 for a discussion
of AIS portfolio diversification) across the different sources of return. Style-pure sub-
indices of Hedge fund returns often provide more accurate estimates of strategy
performance characteristics and correlations than broad composite indices.
Portfolio allocation should take into account the changing dynamics of risk. A
statistical asset allocation can result in widely fluctuating portfolio risk. The AIS
allocator has to find a balance between a predefined and fixed strategy sector allo-
cation and an opportunistic switching between strategies. In the past few years the
concept of ‘risk budgeting’ (or ‘risk bucketing’) has gained considerable
attention.21 Risk budgeting is the process of allocating an acceptable level of meas-
urable risk to certain investment strategies, assessing frequently whether the risk
limits are exceeded and taking corrective measures, if deemed necessary. Hereby,
performance is always measured on a risk-adjusted basis. Usually, the risk budget-
ing approach requires the statistical characterization of each strategy based on
past returns, which has some pitfalls, as discussed in the following.
Determining the appropriate sector allocation for an AIS portfolio is a complex
undertaking. Unfortunately, for the purpose of AIS sector allocation, quantitative
analysis based on past performance and correlation data is of limited value. While
such tools can be quite useful in analyzing the general scope for efficiency enhance-
ment from adding AIS to a global traditional balanced portfolio (equity and bonds),

.... 232 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 233

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

they show their limitations when it comes to sector allocation in an AIS portfolio
itself. Statistical analysis and Markowitz-type portfolio optimizations are very sensi-
tive to the return and risk input used for the calculation. The statistical significance
and the predictive power of performance and correlation data for individual strate-
gies based on historical data are generally insufficient for that purpose:

■ Return and risk properties and the correlation characteristics of individual

strategies vary substantially over time as well as in different market
environments (see Figure 4.3 for an illustration of unstable correlation patterns).
The performance variability of managers within one strategy sector is not
reflected in the average performance numbers for the sector as given by data
providers. This dispersion of returns of AIS managers is much higher than the
dispersion in performance of traditional asset managers (measured by the
‘tracking error’) or the variability of different stocks in one industry sector.
■ The amount of available data is generally low. Performance data is usually
only provided on a monthly basis and does not go back far into the past. It is
also biased due to the exceptional equity bull market in the 1990s.
■ One of the main assumptions of the standard portfolio optimization process is
that investment risk can be completely described by the second moment
(variance) of returns. But the variance of returns (or its square root, the
standard deviation), does not describe the risk for AIS completely and is
therefore not very well suited for AIS portfolio optimization. As discussed in
different parts of this book, skewed and leptokurtic (i.e. fat-tailed) return
distributions are very general characteristics of many strategies. Differences in
historical returns as provided by different index data providers may lead to
large deviations in the obtained results for an optimized portfolio.
■ Hedge fund indices have some generic problems: (1) Not all AIS managers
disclose their performance to databases; (2) indices potentially suffer from the
‘survivorship bias’, which is due to managers terminating their (mostly poorly
performing) program and, as a consequence, disappearing from the index
completely; (3) the choice of a weighting scheme, i.e. whether managers are
weighted equally or asset based, has a significant impact on the AIS indices
(see discussion about strategy performance in Chapter 4).22

.... 233 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 234

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Some strategies include illiquid instruments, which cannot be reliably priced on

a regular basis. Performance tends to appear less volatile and correlations lower
for these strategies, as manager provided estimates leave space for performance
smoothing (‘stale price bias’). The less liquid the instruments traded, the more
knowledgeable and wary the investor has to be about the particular strategy.
■ Given the rapid and dynamic change within the industry and the
heterogeneity of Hedge fund investment strategies, optimization-based past
performances are not sufficiently representative.

Estimating correlations that are useful for the application of quantitative opti-
mization techniques thus requires more than collecting past return data. In his
seminal article ‘Portfolio Selection’, Markowitz notes the difficulty in determining
the correct inputs for portfolio optimization:

[We] must have procedures for finding reasonable [means] and [variances].
These procedures, I believe, should combine statistical techniques and the
judgement of practical men. My feeling is that the statistical computations
should be used to arrive at a tentative set of [means] and [variances].
Judgement should then be used in increasing or decreasing some of these
[means] and [variances] not taken into account by the formal computation.23

The detection of strategy performance drivers and the analysis of factor models for
AIS returns have been subject to much discussion and research in the academic as well
as the financial community in the past few years. But the statistical significance of AIS
factors models has, to date, been rather low. The heterogeneous nature of AIS strate-
gies requires more detailed examinations, including other than quantitative factors, to
render an understanding of their specific return generation possible.24
Two points require particular attention: First, certain strategies contain significant
‘short option’-type exposure, i.e. their risk reward structure is similar to selling
options. The investor receives a fixed premium for being exposed to the possibility of
large losses in some rare cases (which I refer to as the strategy’s ‘stress events’). This
helps to create impressive Sharpe ratios in ‘normal periods’ with no event of loss, but
can ultimately lead to large losses (or ‘blow ups’) following a ‘stress event’. For most of
these strategies, quoted risk-adjusted returns tend to be overestimated if only historical

.... 234 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 235

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

data is considered and quantitative mean variance-based portfolio optimization tends

to recommend inappropriately large allocations to these strategies. The risk inherent in
these strategies has usually not yet manifested itself during a manager’s brief perform-
ance history. A second matter of concern should be that the reported performance and
risk numbers of many strategies do not include certain risk factors such as low liquidity
and event risk. The use of historical data for the purpose of strategy and manager
selection in combination with conventional portfolio optimization techniques usually
results in portfolios with inappropriately high non-systematic risks and low liquidity
(e.g. large allocations to Regulation D and Distressed Debt securities).25
The preceding discussion makes it clear that the set-up of an AIS portfolio
requires more than just ‘number crunching’. The allocation process is less a quanti-
tative science than an art based on the allocator’s experience. Due to the limitations
of quantitative tools for portfolio allocation, qualitative understanding of AIS strate-
gies’ performance characteristics and good judgement are essential.26 Empirical
performance and correlation studies should serve as a general guideline for the AIS
allocator, but he should equally consider ‘semi-quantitative’ and qualitative elements
such as an assessment of economic sources of return and risk, distinction between
premium-based and skill-based returns and the determination of particularly hostile
market environments (which might not yet have tested the strategy). An AIS alloca-
tor must understand the performance drivers as well as the weaknesses of strategy
sectors in order to choose the appropriate combination of strategies in a portfolio.
A significant part of the qualitative understanding required for AIS portfolio
allocation is the assessment of the (economic) sources of return and correlation
features. In Chapters 2 and 3, I discussed the underlying economic functions of
the individual strategies. Many strategies earn a return by capturing a (risk) pre-
mium, which provides an inherent and permanent positive expected return. The
underlying return potential of ‘premium strategies’ does not disappear if other
investors spot its existence. The AIS allocator should generally invest in strategies
with an economic risk premium that he understands. Other AIS strategies benefit
from temporary mispricings and market inefficiencies. These usually disappear
quickly when spotted by other investors. The returns of these ‘pure skill’ strategies
depend primarily on the existence of such market inefficiencies and the manager’s
skill and research resources in detecting and quickly acting on them.

.... 235 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 236

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The following constitutes a summary of the key elements for successful sector
allocation within the AIS portfolio:

1 Analysis of general attractiveness of each strategy sector: The risk–return

profiles of the sectors should be examined and the performance drivers and
underlying general economic rationales assessed. Each chosen sector should
have a reason for being included in the portfolio. Selected sectors should
display attractive returns with appropriate risks (considering risks that do not
directly show in historical performance numbers). Understanding the
relationship (correlations) between different sectors is important.
2 Assessment of market environment: It is important for the AIS allocator to
understand the strategies’ behaviour in different market environments. He should
determine the specific favourable and unfavourable market conditions for each
sector, continuously follow the developments of financial markets and monitor
key economic and financial variables that drive returns. This is similar to what an
equity portfolio manager does with respect to different stock sectors, but extends
over a much wider universe of financial instruments (equities, foreign exchange,
fixed income, commodities). A helpful tool for the analysis of sensitivities to
specific market environments is the analysis of conditional correlations, which
show the behaviours of AIS in particular market environments (see Figures 4.4
and 4.5). The pitfalls of quantitative analysis must be kept in mind, however.
Common sense and solid economic reasoning should back statistics.
3 Assessment of risk factors: The particular risk factors (and risk premiums) of the
strategies should be understood. The portfolio/risk manager should try to measure
and quantify these risks and determine measures and procedures for controlling
them. Managers in the same strategy sector can have different focuses (e.g. sector,
country, credit quality) and therefore different exposure to particular risk factors.
The number of individual trading managers needed to reach sufficient
diversification within a strategy sector is an important consideration.

.... 236 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 237

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

4 Analysis of past and anticipation of future correlations: Correlations are the

key determinant of diversification and the optimal risk–return ratio within a
portfolio. The correlations between strategies as well as each strategy’s
correlation to other, traditional, asset classes (i.e. bonds and stocks) should be
analyzed and understood (see also the discussion in Chapter 5). Strategies that
behave in complementary ways in particular market environments are
especially interesting from the viewpoint of portfolio diversification.

It should be noted that the sector allocation process does not end with the initial
investment, but requires ongoing market monitoring and portfolio reassessment.
Changes in the economic environment may make portfolio modifications neces-
sary. Furthermore, the attractiveness of particular strategies can change or even
vanish over time, due to increased market efficiency, regulatory constraints, lim-
ited capacity or other reasons. This might require a sector reallocation within the
AIS portfolio. Finally, portfolio considerations have to take into account the (pos-
sibly changing) liquidity requirements of the client, legal constraints (concerning
e.g. shorting stocks, instrument constraints), home currency and tax status.

Manager evaluation (bottom up approach)

Each investment involves certain ‘manager risks’ specific to trading styles and
employed investment techniques such as leverage, investment instruments and
Short Selling. Manager-specific risks also include operational risks, legal risk, key
person risk, model risk and fraud risk. Two measures are essential in minimizing
these risks: proper due diligence and manager diversification. Thorough due dili-
gence should, without exception, be a requirement for an investment with a
particular AIS manager and should lead to a detailed understanding of the invest-
ment strategy. Managers within the same strategy sector can differ with regards to
strategy implementation, instrument diversification, hedging, use of derivatives,
short selling and the degree of leverage.
There are qualitative and quantitative aspects to the process of AIS manager due
diligence. Quantitative analysis consists mostly of ‘number crunching’ and should

.... 237 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 238

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

include the analysis of past returns, volatility, drawdowns, Sharpe ratio and other
performance measures,27 as well as correlation studies, benchmark comparisons
and peer group analysis. It should be noted that quantitative performance evalua-
tion is easier to perform for ‘premium-based’ strategies, as benchmarks are more
uniform and peer group comparison more informative. In comparison, it is more
difficult to evaluate the performance of ‘pure skill’ strategy managers. There is usu-
ally no clear benchmark to compare the single managers to and it is more difficult
to distinguish past performance as a result of skill from pure luck. Thus, due dili-
gence of the particular approach and the manager becomes even more important.
Unfortunately, many investors still rely too much on past performance when
selecting managers, without thorough qualitative analysis. Qualitative aspects of
the due diligence involve knowing the manager’s ‘edge for returns’, the invest-
ment style and attitude of the manager, details of the investment decision-making
processes, the organization and structure of the manager’s operations, the trading
facilities and the character, quality and background of key people.
The precise process of manager selection varies among AIS allocators, but there
are some due diligence rules that one should keep in mind.

Concerning the track record

■ A good track record is a necessary but certainly not a sufficient condition for a
high-quality trading strategy. Poor strategies can come with good records obtained
in a favourable market environment (e.g. equity bull market) or simply because of
luck. Database analysis often defines the starting point of manager selection, but
the AIS allocator cannot rely on quantitative analysis alone for manager selection.
Understanding the strategy and its sources of returns and the strategy’s
weaknesses and risks is crucial. Strategies can be vulnerable to certain types of
market stress, which might not be apparent from historical return numbers (‘short
option’ type of strategies).

.... 238 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 239

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

About meeting the manager

■ It is important to meet many managers and build up experience. Managers
have different focuses and comparisons between managers in one peer group
can be helpful in understanding important nuances.
■ The AIS allocator should prepare a due diligence questionnaire which should
be filled out by the manager at the beginning of the due diligence process. The
questionnaire should be designed thoroughly, as it can be one of the most
important sources of information for desk research.
■ Manager due diligence has a strong subjective component. It is useful to visit
and evaluate managers in a team, as several people observe more than one
person by himself. References are a good starting point for a due diligence and a
good industry network can be helpful for discussing and evaluating managers.
■ It is important to talk to the key people (trader, fund manager, CEO) directly
instead of the marketing staff. Generally, the latter do not know the strategy in
sufficient detail. Onsite visits are an important element of the due diligence.

Necessary background checks

■ Valuable information can be obtained from the marketing and legal material
(offering memorandum, past letters to investors, audited track record, annual
reports and any marketing material). This rather broadly composed material is
usually of limited value in respect of the details of the investment process, but
contains important information about fees, NAV calculation, biographies of
key people, legal set-up, redemption periods, restrictions on leverage, counter-
parties and disclosure policies.
■ Independent checks of the background of key personnel are a necessity. Many
cases of fraud could have been detected early on had investors performed
sufficient background checks.28 Sources of information are ex-employers, auditors,
administrators, prime brokers, other investors and government agencies.

.... 239 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 240

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The steps of the due diligence process are illustrated in Figure 7.4. Due diligence
usually progresses through the following stages: initial screening; first contact;
quantitative analysis; desk research; onsite meetings; final evaluation. The out-
come of the due diligence process is a decision as to whether or not to invest with
the manager. Initial screening usually requires database analysis to reduce the
global manager universe to a manageable size. Other sources of information
include personal references, conferences, manager sales visit, prime brokers and
industry hearsay. One of the edges of a fund of funds manager is to be inside the
‘information loop’ of high-calibre managers. The next step is quantitative analysis,
which includes numerical analysis of performance (return, volatility, drawdowns,
Sharpe ratio etc.) and correlations to other managers in the peer group/strategy
sector and in the AIS portfolio as well as to other asset classes. The following desk
research is the beginning of the qualitative analysis and includes reviewing legal
documents, the fee structure, assets under management, investment capacity, audi-
tor statements, reference checks and, possibly, a completed due diligence
questionnaire. An onsite visit to the manager completes the qualitative due dili-
gence process and should include the following: a discussion with key personnel
about the investment approach, strategy implementation and details of risk man-
agement techniques; a review of operations and trade executions; an examination
of back-office structures; an assessment of the firm’s working atmosphere.
Some AIS allocators try to formalize the due diligence process by creating a ‘point
system’, where the manager is given points for each area of investigation according to
a predefined system. Whether one prefers a point system or another approach, the
allocator should cover the following key issues during his due diligence.

.... 240 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 241

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

Sector appropriate Yes Reviewing Peer Group Initial Meeting

New HF Manager Questionnaire
for Portfolio? Info-Material Analysis or Conf Call

• HF Database • Marketing • Categorization • Strategy Explanation

• HF Journals No • Performance • Statistical Analysis • Discussion with HFM
• Introduction • Other Info and Data • Solve 1st questions
• Recommendation • Managed Account
• Conference • Min. size
• Cold Call from HFM
Files

Desk Research Onsite Due Diligence Internal HFM Rating

Report Meeting Report Presentation By Committee

• Preparation for • Verification • Presentation of • Strong Buy

onsite Meeting • Operation HFM by analyst • Buy
• Quant Analysis • People • Discussion within • Watch List
• Return Assessment • Trading/Execution Investment • Not recommended
• Risk Catalogue • Final discussion Committee
• References • Open questions
• 8 P’s

FIGURE 7.4 ■ The process of Hedge fund manager (HFM) due diligence

Manager due diligence: The ‘eight P analysis’

PEOPLE

The allocator should examine the background, experience, integrity, attitude and
lifestyle of key people in the manager’s firm. Information sources include prior employ-
ers, other investors, auditors, administrators and prime brokers. He should also check
whether any key person or director has been subject to an investigation by a govern-
ment regulatory agency. It is advisable to invest only with managers that are registered
and regulated by regulatory authorities (e.g. with CFTC, NFA, SEC, SFA etc.).

PRODUCT

The manager’s investment strategy (his ‘product’) is defined by certain character-

istics such as target return and risk (e.g. volatility, drawdowns); investment
instruments; level of diversification; position entry and exit criteria; hedging

.... 241 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 242

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

tools; time horizon; asset liquidity; leverage; use of derivatives; short selling. The
investor should familiarize himself with the details of these characteristics.
Additionally, he should assess the manager’s fee structure and obtain information
about targeted investors and existing investor base.

PROCESS

A key element of the due diligence is to evaluate how investment decisions are
made. The allocator should know the key steps and people involved in making
and executing decisions. Where investment decisions are based on a model, the
model developer is an important person to talk to. The allocator should familiar-
ize himself with the details of the model development and testing and assess the
potential danger of model over-optimization/curve fitting. Performance reports
should distinguish between in-sample and out-of-sample testing (in-sample testing
refers to the optimization of the model’s parameters on historical data, while out-
of-sample testing means examining model performance on historical data that was
not used during the optimization process).

PERFORMANCE

Quantitative analysis of the track record is an important part of the performance

evaluation process. The AIS allocator should request audited performance numbers
of all relevant trading accounts of the manager. Performance presentation standards
can vary among different managers. The calculation assumptions and methods used
by each manager should be clear.29 Performance analysis should include the exami-
nation of returns, risk and correlations. Returns should be measured on a
risk-adjusted basis. Measures for risk are standard deviation, maximal drawdown,
downside deviation, semi-variance and leverage factors. Risk-adjusted performance
measures are Sharpe ratios, Sortino ratios, Information ratios, Park ratios, Treynor’s
measure, Risk-Adjusted Return On Capital (RAROC), ‘effective return’ and others.30
As emphasized throughout this book, besides examining the manager’s historical
returns, the allocator should ask about the sources of the manager’s returns. He
should understand the economic reasons for the strategy’s returns and risk and
familiarize himself with the trading manager’s ‘edge’. Trading discipline and solid
risk management are important edges for ‘premium-based’ strategies. For strategies

.... 242 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 243

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

based primarily on forecasting skills (‘pure skill strategies’), edges can be of a differ-
ent nature. Informational edges can be better or faster access to relevant
information, superior research capabilities or simply better knowledge about partic-
ular sectors/countries and the relevance of information or the behaviour patterns of
investors. Statistical edges are based on recognizing market features such as trends,
price patterns, seasonal price developments or conditional (auto-) correlations.31 It
must be emphasized that statistical and informational edges can disappear over time,
e.g. when other market participants copy and adopt a proven strategy.
The intricate trade secrets and details about models are, for the most part, not
too important for the AIS allocator. Non-confidential information is generally
much more informative than the details the manager is unwilling to disclose.
Relevant information can be obtained by discussing the general approach, philoso-
phy and economic edge of the strategy, which the manager should be more than
comfortable discussing.

PARTNERSHIP

The AIS allocator has to know the firm’s legal and ownership structure. Incentive and
compensation schemes within the firm can give important hints as to the commitment
of its employees, especially the dedication of key people. The geographic location of
the firm matters for attracting investment talent. The allocator should make an assess-
ment of the firm’s business risk. Hedge funds sometimes fail for pure business reasons
such as employee turnover, partner disputes, poor accounting or others. Finally, he
should know whether the manager is investing his own money into the strategy and
assess the principle’s personal commitment to, and belief in, the strategy.

PORTFOLIO

Diversification and portfolio management are important considerations for the alloca-
tor. Systematic model-based trading should occur over a broad universe of different
instruments and possibly asset classes, while discretionary approaches are more likely
to be successful in a limited range of instruments. Limits on positions in certain instru-
ments and asset classes, if any, provide a good idea about possible concentration risk.

.... 243 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 244

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

PEERS

The allocator should examine the uniqueness of the strategy and its return correla-
tions with peer strategies. One should differentiate between the inherent return of the
strategy sector (see Chapter 3) and the edge of a particular manager, i.e. his ‘alpha’.
Earning a general risk premium does not necessarily constitute an edge. The manager
should be able to distinguish himself clearly from other managers within the sector.

POTENTIAL

Most managers have capacity limits, i.e. their strategies are constrained in the
amount of assets they can manage without diluting performance. The indicated
dollar capacity should be compared to what can reasonably be expected for the
strategy. Managers sometimes accept more money than they can efficiently invest.
It is also important to follow up on this issue, as managers often change their view
on capacity once the initially indicated limit is reached. This should raise a warn-
ing flag. Often managers who have reached maximal capacity for their core
strategy offer other investment programs in order to benefit from their positive
reputation in the market and further increase their assets under management. The
allocator should be careful about being persuaded to invest in another program,
when the original program of his choice is closed for further investments.

The manager’s risk control

There are important risk management procedures on the single strategy level (see
Chapter 5). The allocator should pay particular attention to the following points:

1 He should know the details of the manager’s internal risk management

procedures such as stop-loss limits, maximal allocations to particular
instruments and asset classes, portfolio diversification, hedging tools, risk
monitoring, risk budgeting etc.
2 He should ask the manager how large he expects a maximal drawdown to be.
Further, he should assess at what point the manager would lose confidence in his
approach and stop trading. It is also interesting to know what level of positive
performance would significantly exceed the manager’s expectations. Outstanding

.... 244 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 245

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

performance during a certain period can be a sign of increased risk and can
quickly turn into severe losses when market conditions change.
3 He should address the manager’s possible past mistakes and discuss the lessons
learnt from these mistakes. Often internal risk management procedures are the
result of those past mistakes. A good manager has nothing to hide and should
openly discuss this issue.
4 The details about the strategy’s risk factors should be listed in a risk catalogue.

Market environment analysis

The performance of each strategy strongly depends on the market environment. Each
strategy has particular weaknesses and strengths, which show at different moments. It
is the AIS allocator’s responsibility to structure a well-diversified portfolio, where the
weaknesses and strengths that single strategies display in different market environ-
ments balance out to yield consistent and smooth overall portfolio returns. For this
purpose, he should consider three issues during the manager due diligences:

1 He should assess the favourable and unfavourable market environments for each
manager. Again, this requires more than evaluating historical returns. The
allocator should be aware of the inherent ‘survivorship bias’ of AIS performance
numbers. Managers have often never experienced a particularly painful period
and therefore looking at past returns does not reveal all possible weaknesses of
their trading strategy. The first occurrence of a ‘stress period’ can easily lead to
the end of a trading program, such that survivors tend to be ‘untested’
managers. Common sense might yield a better understanding of critical market
environments for a strategy than pure number crunching.
2 The AIS allocator should know the key performance drivers (and risk factors)
for the trading strategies included in the portfolio. These can be equity market
direction (e.g. for most Long/Short Equity, Short Selling and Equity Market
Timing strategies), the volatility of equity markets (most Risk Arbitrage and
Convertible Arbitrage strategies), yield curve configurations (e.g. Fixed Income
Arbitrage), the volatility of commodity markets (Managed Futures strategies)

.... 245 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 246

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

and others. A significant part of this assessment should take place on the level
of sector allocation, but the market environment analysis has to consider
specific manager characteristics.
3 Hedge funds and Managed Futures are ‘absolute return’ investments. Their
goal is not to beat a particular index, but to yield absolute returns
independent of the broad equity and bond markets. Nevertheless, certain
manager benchmarks can be helpful for the purpose of performance
evaluation. This applies in particular to ‘premium strategies’, i.e. strategies
based on capturing specific (risk) premiums. There are a number of sector
indices offered by various data providers (see appendix to Chapter 4), most of
which come with conceptual problems. Commonly used AIS benchmarks are
the returns of the median (or mean) manager in a certain pool of managers. It
should be noted that benchmarks defined in this way can be flawed.32

Final questions of the due diligence process should centre on ‘soft issues’ such as
fee agreements with brokers,33 the manager’s personal trading and investments in
the fund and elsewhere, fee rebates and special fee arrangements for other
investors, business targets and other possible concerns.

Investment monitoring: ‘Post-investment’ risk management

Diligent pre-investment analysis is essential for making successful AIS investments.
But the job of an AIS allocator is far from complete after strategy and manager allo-
cation. The allocator and AIS portfolio manager should by no means neglect the
importance of ‘post-investment risk management’, i.e. ongoing investment monitor-
ing and active risk control after investment with the chosen managers. While one
might learn a great deal during due diligence, it is almost certain that risk factors not
encountered before will emerge. In dealing with the current and future risks in the
portfolio, risk management should thus be proactive rather than solely reactive to
reported performance. Manager monitoring and risk control is at least as time con-
suming and challenging as sector selection and manager due diligence.
A key element of managing AIS risk is continuous and independent monitoring
of positions, P&L and risk levels. This can be seen as an ongoing post-investment

.... 246 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 247

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

manager due diligence. Investment monitoring in combination with good strategy

knowledge enables the allocator to obtain a ‘dynamic understanding’ of the strat-
egy and the manager’s behaviour in various market circumstances, detect
previously unknown (or unrecognized) strategy risks, quickly recognize style
changes, undesired ‘bets’ and increases of leverage employed by the manager, per-
form independent performance evaluation and decrease the probability of fraud
risk. Continuous monitoring is a prerequisite for controlling the risks of the man-
agers’ trading activities, performing active risk management (e.g. de-leverage) and
reallocating investments within the portfolio when circumstances change.
At a minimum, post-investment risk management should involve reading the
monthly reports of managers with great care and speaking with each of them on a
regular basis in order to receive first-hand information about the market situation,
the background of performance numbers and changes in the strategy such as modifi-
cations in the investment decision process, adjustments to the manager’s operating
facilities and possible staff turnover. The AIS allocator should further observe the
relevant financial markets and with his understanding of key performance drivers
develop his own views on the current market environment and the performance of
his managers. For strategies that involve credit risk (e.g. Regulation D or Distressed
Securities) he should monitor the credit markets and the ratings of the relevant com-
panies and possibly perform independent research on default and loss probabilities.
Periodically, he should check volatility patterns and correlation features in order to
assess whether the manager has changed anything relative to his original approach.
Changing performance and correlation behaviour can be an indication of long-term
style drifts. Information about the size of assets under management, the nature of
new investors and possible redemptions should be obtained on a regular basis.
Ideally, post-investment risk management should involve frequent – in best
cases daily – monitoring and risk analysis as discussed in the following.

Independent pricing and performance analysis and performance attribution

Considering the numerous instruments and asset classes traded by AIS, pricing, NAV
calculation and performance measurement is a non-trivial task. Assigned prices may
fail to recognize position size, low market liquidity, instrument complexity (e.g.
derivatives), flaws in the valuation model and possible circular reference to ‘inde-

.... 247 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 248

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

pendent’ price sources. The portfolio and risk manager must be critical of the
integrity of the pricing or mark to market valuations provided by the manager and
obtain market prices from independent sources, preferably directly from prime bro-
kers and market makers. Prices should be discounted in the case of low liquidity.
The analysis of performance attribution (possibly by asset class, sector and
instrument) can be very helpful for a further understanding of the trading strategy
and an evaluation whether the manager followed his indicated investment strategy
(examination of possible style drifts). Performance attribution is twofold:

1 Breakdown of performance to individual asset classes and sectors.

2 Analysis of performance with respect to different investment styles.

The latter is an important part of traditional asset management (e.g. ‘value vs.
growth’) and often requires an underlying factor model.

Exposure analysis
Besides aggregating exposure on the level of the global portfolio, the analysis should
include a breakdown of gross exposure into different aggregation levels such as asset
class, sector and instrument type. Without this ‘drilldown’ the risk manager might miss
possible exposure concentrations in the portfolio. A seemingly small net equity position
on the level of the total portfolio, for example, can still result in large gross exposures to
certain sectors. Figure 7.5 shows an example for the analysis of equity exposure in an
AIS portfolio. If a manager tends to take large positions in single sectors or instruments,
these should be monitored with special care and checked for possible violations of pre-
defined maximum position sizes. An important part of the exposure analysis is the
monitoring of leverage. Leverage analysis should be considered on a gross and net basis.

Correlation and volatility analysis

The AIS allocator and risk manager should observe volatilities and correlations
between strategies and with stock and bond markets and analyze those on the
level of the global portfolio as well as the individual managers. Short-term shifts
as well as persistent changes of volatility and correlation patterns can be indica-
tions of possible problems of a strategy or of manager style drifts.

.... 248 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 249

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

Equity Exposure

10,000,000
Interpretation (Example):
8,000,000 Market exposure in Asian
equities was USD 6 million
6,000,000
on April 4, 2000
Exposure in USD

4,000,000

2,000,000

–2,000,000 March 29/00

March 30/00
–4,000,000
March 31/00
–6,000,000 April 03/00
USA April 04/00
Europe
Asia
Category

FIGURE 7.5 ■ Example for the equity exposure analysis in an AIS portfolio. The development of
exposure is displayed over one week

Value-at-Risk (VaR)
VaR is a volatility-based portfolio risk measure that describes the maximal loss in
a portfolio or sub-portfolio for a specified confidence level (e.g. 95% or 99%)
over a certain trading horizon (e.g. 1 day or 5 day). It provides a measure that
enables the risk manager to quantify and compare portfolio risk across different
instrument and asset classes uniformly. One drawback of VaR is that it does not
explicitly consider the extreme part of the return distribution (see discussion in
Chapter 6). It is also often criticized that the measurement of volatility (and VaR)
includes upside deviations, i.e. volatility caused by excessive gains. But the pru-
dent risk manager is actually well advised to include upside deviations in his risk
measurement. Often excessive gains are due to higher risk and upside deviation
can quickly turn into downside deviation, when markets turn against a strategy.
Not all managers are skilled at reducing position sizes just at the right time to
avoid the downward move when market conditions change.

.... 249 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 250

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The selection of time horizon and confidence intervals depends on the objective
of the analysis. For daily monitoring, a one-day time horizon is most appropriate. A
95 or 99% confidence interval is usually chosen. The risk manager also has to make
a choice concerning the employed calculation method (see Chapter 6). Because of
the non-linear and complex derivative positions present in most AIS portfolios, the
variance-based method can be quite misleading and should not be used exclusively.
Historical simulations are attractive, because they make no distributional assump-
tions, but the results depend strongly on the historical period chosen for the
analysis. The most reliable is the Monte Carlo method, but it is computationally also
the most intensive. Further, as long as normally distributed draws are used, the
Monte Carlo VaR does not address the issue of non-normally distributed asset
returns (for more details, the reader is referred to Chapter 6).
VaR can be analyzed along two different dimensions:

1 ‘VaR tracking’ is the measurement of risk as a function of time. This provides

an indication of how high current risk is compared to risk levels experienced
in the past. Analyzing VaR over time can provide valuable insights for the
assessment of how often single managers push their risk level to certain limits
and whether over time risks are in line with expectations. Tracking VaR is
another tool for the diagnosis of ‘style drifts’.
2 ‘Risk drilldown’ is the analysis of risk at different aggregation levels. Risk
drilldown helps the risk manager to break down risk with respect to different
sectors or trading managers. This answers the question about where the
overall portfolio risk originates.

Figure 7.6 shows an example of AIS portfolio risk analysis with VaR tracking and
drilldown of VaR.
A concept related to VaR is Expected Shortfall (also called ‘Conditional VaR’;
see Chapter 6), which describes the mean portfolio value conditional on the port-
folio loss exceeding a certain threshold (which is usually chosen to be the 95 or
99% confidence limit VaR). Expected Shortfall supplements VaR by providing
more insight into the tail of the return distribution.

.... 250 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 251

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

VaR 99% Confidence Interval: VaR Tracking

Interpretation:
1.00% There is a 1% chance (99%
0.80% confidence interval) that a daily
0.60% return will fall outside of the VaR
0.40%
Realized P&L band. The actual profit an loss
0.20% realized on the following day is
VaR

–0.00% also displayed

02.11.
09.11.

16.11.

23.11.

30.11.

07.12.

14.12.

21.12.

28.12.

04.01.

11.01.

18.01.

25.01.

01.02.

08.02.
–0.20%
–0.40%
–0.60%
–0.80% VaR band
–1.00%
Date

VaR 99% Confidence interval: Drilldown to asset classes

0.60%

0.50%

Interpretation: 0.40%

On February 11 2002, there was a 1% chance 0.30%

(99% confidence interval) to loose more than 0.55%
of the overall fund’s NAV in equity positions 0.20%
05.Feb
0.10% 06.Feb
07.Feb
0.00% 08.Feb
11.Feb-02
s
itie

e
om

FX
od

y
uit
Inc
m

Eq
m

ed
Co

Fix

FIGURE 7.6 ■ VaR tracking and ‘Drilldown’ of VaR in an AIS portfolio

Marginal and incremental VaR analysis

Marginal VaR expresses the amount by which a portfolio’s VaR is changed on
inclusion of a particular position or sub-portfolio. This measure can be used to
evaluate how much an individual trading manager or position affects the risk
structure of the entire portfolio. An asset or a particular trading strategy with a
negative correlation to the rest of the portfolio gives a favourable contribution to
the portfolio VaR, i.e. it decreases the total risk in the portfolio if included.

.... 251 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 252

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Stress tests and scenario analysis

Stress tests provide the risk manager with insights about the portfolio behaviour under
extreme, but plausible market conditions. They can be performed by considering cor-
responding scenarios, e.g. a 5+ standard deviation move of single risk factors (such as
an 8% drop in the S&P500), a correlation ‘breakdown’ (i.e. the scenario of certain
assets displaying a correlation close to one) or longer liquidation periods (i.e. applying
longer time horizon VaR calculations). Scenario analysis consists of applying the prede-
termined changes, the so-called ‘what-if ’ scenarios, to the risk factors and calculating
the resulting effect on the portfolio value. This analysis can be applied independently
to each risk factor (which affects only those instruments that directly depend on it) or
in consideration of the correlation structure between all risk factors. A scenario analy-
sis can also be performed with respect to external factors such as policy changes,
election outcomes or weather conditions. The effects on security values must then be
‘guesstimated’. An example of a stress test report is presented in Figure 7.7.

Liquidity analysis
The risk manager should examine the liquidity of individual positions in the port-
folio by assessing how quickly and at what price they could be liquidated. This is
especially important in times of financial distress. Important indicators for instru-
ment liquidity are trading volume, bid–ask spreads and market capitalization (in
relation to position size).
Computational technology plays a more and more important role in the analysis
and measurement of risk. AIS investors and portfolio allocators can today choose from
many different off-the-shelf risk analysis tools which are based on increasingly sophisti-
cated software running on faster, comparably cheaper computers. Today’s computers
can handle highly complex calculations and simulations (e.g. Monte Carlo simulation,
stress tests) in a very short time frame and at high frequency (e.g. every day or even
intra-day). Prime brokers offering risk analysis as a value-added service to clients are
often at the frontier of technical development for the calculation and reporting of risk.
With the use of modern information technology, risk managers are in a position to
gather all relevant positions and transactions, extract the relevant information and per-

.... 252 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 253

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

form the necessary risk analytics (VaR, scenario analysis, stress tests etc.) without any
significant time delay. While not having to develop risk analysis systems themselves,
risk managers must be equipped with the skill base and experience necessary to apply
the tools, interpret their outcome and act when necessary.

Simulated Scenario –10% –5% 5% 10%

S&P 500 –1.94% –0.80% 1.98% 5.91%
Euro-Stoxx-50 –0.98% –0.39% 1.01% 3.04%
Nikkei 225 –0.31% 0.02% 0.91% 2.87%
80 bp down 40 bp down 40 bp up 80 bp up
US government curve 2.16% 0.65% –1.34% –3.14%
30 bp down 15 bp down 15 bp up 30 bp up
Corporate single A yield spread 2.01% 1.01% –2.02% –1.04%
–8% –4% –4% 8%
EUR-USD 2.38% 1.19% –1.04% –2.01%
USD-JPY –1.62% –0.34% 0.56% 1.98%
–10% –5% 5% 10%
GS Commodity Index 3.13% 1.11% –0.71% –3.01%
Historical Scenario
Stock market crash 1987 –4.91%
Bond market crash 1994 –4.22%
Russian default / LTCM crisis 1998 –5.48%
Brasil crisis 1999 –1.01%
Sept 11, 2001 –2.79%

FIGURE 7.7 ■ Stress tests risk report

Evidently, each strategy has particular features and appropriate risk monitoring
requires different tools for different strategies. The quantification of risk is generally
more difficult for Merger Arbitrage, Distressed Securities, Regulation D, Convertible
Arbitrage and many Long/Short Equity strategies. VaR analysis based on historical
volatilities and correlations is not very insightful for Event Driven strategies in most
cases, while it provides a useful tool for the analysis of risk for most opportunistic
Hedge fund and Managed Futures strategies. Nevertheless, continuous exposure and
leverage analysis, monitoring of large positions and checking whether the manager is

.... 253 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 254

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

adhering to the predefined guidelines and focuses of his strategy remain very impor-
tant elements of risk management for all strategies. A Merger Arbitrage or Long/Short
Equity manager who defines his edge in the banking industry should not have large
concentrations in utility or chemical stocks and a Convertible Arbitrage manager
focusing on in-the-money convertibles should not have positions in convertibles
priced near their bond value. What the monitoring process boils down to is checking
whether the manager’s activities correspond to what the allocator expects.
Finally, risk management requires appropriate risk reporting schemes.
Investment decision makers (i.e. the AIS allocators) and risk managers should
work closely together and communicate efficiently. The most important informa-
tion in risk reports is often buried in a vast amount of data. It is important to
select and report the right measures of risk to enable decision makers to identify
quickly the necessary actions to take. Reports should aggregate risk on the appro-
priate level highlighting the critical exposures, so that decision makers can focus
their attention directly on key issues. The illustration and visualization of risk
through diagrams and charts can be very helpful in the communication of risk to
parties who are not directly involved in the analysis process.34

Active risk management

As mentioned throughout the book, risk management is both an art and a science.
While the latter refers to the more quantitative elements of risk measurement, the
‘art’ of risk management corresponds to the wisdom and good judgement that
the risk manager develops over time. The financial industry has seen an astonish-
ing proliferation of risk analysis tools in recent years. Most of these tools provide
the risk manager with the ability properly to measure and analyze risk. Risk
measurement is indeed an important element of risk management, but in itself it
remains a passive activity. In contrast, risk management requires action and goes
beyond the pure measuring of risk. Often, quantitative reports provide necessary
data, but not sufficient information. While risk measurement aims at providing
an objective assessment of how much risk is present in the portfolio, risk manage-
ment can be based on a variety of other (sometimes rather subjectively
determined) factors, which, among other determinants, depend on the investor’s

.... 254 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 255

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

risk profile. Active risk management, (a pleonasm, as ‘risk management’ should

always be ‘active’), entails the dynamic and optimal allocation of risk among dif-
ferent assets and managers. It consists of defining and enforcing risk limits,
performing dynamic portfolio allocation according to – possibly changing – risk
parameters and accepting certain risk factors, but eliminating others that are
deemed unacceptable.
As discussed throughout this book, the performance drivers in AIS are risk pre-
miums and the investment skills of single managers. The very fact that AIS managers
take risks is surely not undesirable. However, risk that is unintended, misunder-
stood, mismanaged or mispriced should not be accepted and it should be the risk
manager’s goal to avoid these risks. The individual AIS manager is responsible for
executing his particular trading strategy. If he executes his strategy along the invest-
ment guidelines and the AIS allocator accepts the known underlying risks, there
should be no reason for concern on the part of the risk manager. The less the man-
ager’s activity has to be interfered with the better. Ideally, the manager follows his
strategy and shows performance as expected in given market conditions.
But the risk manager needs to able to take proactive steps, once this ideal does
not apply any longer, in order to remedy critical situations quickly. He should be
concerned when, in given market environments, strategies deviate from their per-
formance expectations. Deviations from expected performance are indications for
strategy style shifts, which are among the main reasons for sudden unexpected
losses. Performance deviations can be mediocre performance in favourable market
conditions as well as better than expected performance in unfavourable market
environments. When a crisis has arrived, it may be too late to make adjustments.
Within a framework of active AIS risk management there are various ways to
manage risk and exercise control over trading managers:

■ Regular conversations. The AIS allocator should frequently talk with managers
about the current and potential future risks of their trading strategy. He
should know how the individual manager judges his strategy’s current risk
profile and, based on his understanding of the strategy, develop his own view
about its vulnerability in the prevailing market conditions. The allocator
should identify changes in a strategy’s risk structure at the earliest possible
stage, if possible before they occur.

.... 255 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 256

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■ Risk or exposure limits. The risk manager should distinguish between risk levels
and sources which are acceptable and those which are not. He can define certain
‘risk budgets’ on the level of the global portfolio, single asset class or strategy
sector. These are ceilings on the amount of acceptable risk as indicated by
measures such as VaR, notional exposure or possible stress test losses. The
definition of specific values for these limits is critical and depends on the
investor’s investment objectives and risk tolerances. They should be dynamic
rather than fixed.35 Once predefined limits are exceeded, the risk manager must
undertake action such as re-allocation or de-leveraging.
■ Imposing leverage controls and margin requirements. Each manager should be
restricted as to how much leverage he is allowed to employ at any time. The
maximal leverage should be set in agreement with the manager before trading
starts. A violation of the allowed leverage limits should trigger immediate
action ranging from information requests to closure of positions or complete
exclusion of the strategy from the portfolio.
■ Definition of stop-loss limits. The portfolio manager can set limits on each
manager’s maximal drawdown. The definition of such a limit should depend
on the strategy and should be made in consultation with the manager before
investment. Once a drawdown exceeds the maximal loss limit, the
corresponding manager should be excluded from the portfolio.
■ Request for explanations about unusual positions. Unexpected and
unexplained high exposures to certain securities or asset classes or unusual
leverage factors should be immediately addressed. They are warning signs for
upcoming problems. If the manager does not provide a convincing
explanation for such exposures, the allocator should force him to close the
positions and consider excluding him from the portfolio.
■ Firing of managers with ‘issues’. Managers who have breached certain
agreements, are convicted of illegal or unethical behaviour by regulatory
authorities or take excessive single bets that do not correspond to defined
trading strategies should be quickly excluded from the portfolio.

.... 256 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 257

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

It has to be noted that risk management does not, by itself, lead to positive perform-
ance neither does it prevent losses under all circumstances. But it may help to decrease
the severity of losses. Risk monitoring and active risk management help to detect man-
agers who are deviating from their strategies and thus prevent accidents and
investment disasters from happening. Performance generation is a matter of sector and
manager allocation. For each strategy there exist hostile market environments that will
lead to the strategy showing weak performance or losses. In a well-diversified AIS port-
folio, lacklustre performance of single strategies should be balanced by above average
performance of others for which prevailing market conditions are more favourable.
Finally, a word about the tragic events of September 11, 2001. It is clear that
Hedge funds, like all other financial institutions, are exposed to operational risks of
different severity. Some of these risks might be perceived as ‘acts of God and other
things you cannot plan for’. While banks and large financial firms have offsite opera-
tions centres and back-up facilities, smaller firms cannot afford the ‘luxury’ of
thorough contingency planning. While the direct impact of the attack on the World
Trade Center on Hedge fund and Futures managers was very limited, AIS firms might
nonetheless have to revisit operational and other risk management policies in light of
the events on September 11 and put corresponding emergency measures in place. But
the key principles discussed throughout the book remain applicable even in the most
extreme of circumstances. Managers should stick to the risk management basics by
remaining diversified and understanding their exposures.

Notes
1. Caxton Corporation, Kingdom Capital Management, Moore Capital
Management, Soros Fund Management LLC, Tudor Investment Corporation.
2. ‘Sound Practices for Hedge Fund Managers’, report issued by the group of five
Hedge fund managers, February 2000.
3. Group of Thirty, ‘Derivatives, Practices and Principles for Dealers and End
Users’, New York; the report is available on the IFCI web page http://risk.ifci.ch,
maintained by the International Finance and Commodities Institute (IFCI), a non-
profit Swiss organization.

.... 257 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 258

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

4. Investor Risk Committee (IRC), ‘Hedge Fund Disclosure for Institutional

Investors’, available on the IAFE web page http://www.iafe.org or on
http://www.cmra.com. The report was amended in July 2001. The discussion within
the IRC is ongoing. I am referring to the document published on July 27, 2001.
5. See discussion in ‘Hedge Funds to Clear the Information Hurdle’ in Risk
Management, October 2001, p.8.
6. See http://www.bis.org/publ/bcbs79.htm for more details.
7. See also the results of a survey by the firm Capital Market Risk Advisors
(CMRA) in early 2002, available on http://www.cmra.com.
8. For a detailed discussion of the value added of fund of funds and some
interesting statistics on their performance, see the contribution by A. Ineichen, ‘The
Search for Alpha Continues – Do Fund of Hedge Fund Managers Add Value?’.
9. See A. Ineichen in The Search for Alpha Continues – Do Fund of Hedge Fund
Managers Add Value? for more details about the fee structure of funds of funds.
10. For a detailed discussion of the LTCM bankruptcy see Ph. Jorion, ‘Risk
Management Lessons from Long-Term Capital Management’, downloadable from
http://www.gsm.uci.edu/~jorion/research.htm; also ‘Hedge Funds, Leverage, and
the Lessons of Long-Term Capital Management’, report of the President’s Working
Group on Financial Markets, April 1999, on http://risk.ifci.ch/146530.htm.
11. Interesting studies about how many managers are needed to reach sufficient
‘manager diversification’ in an AIS portfolio can be found in ‘Fund of Fund
Diversification: How Much is Enough?’ by J. Park and J. Strum.
12. The other side of the coin of the Hedge fund boom is a boom in Hedge fund
fraud. A good article about Hedge fund fraud cases is presented by D. Kramer in
the article ‘Hedge Fund Disasters: Avoiding the Next Catastrophe’ in the
Alternative Investment Quarterly, October 2001; see also an interesting article by
T. Fedorek, ‘Is Fraud Flourishing At Your Hedge Fund?’, Pension & Investments,
March 19, 2001, p.14.
13. Regulation D managers have been among those with the highest risk-adjusted
performance in recent years, as measured e.g. by the Sharpe ratio. As the
underlying investments are very illiquid and cannot be accurately priced, reported
volatility is usually much lower than the real risk of the investment.

.... 258 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 259

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

14. See the following articles: M. Peskin et al., ‘Why Hedge Funds make Sense’,
Quantitative Strategies, Morgan Stanley Dean Witter, November 2000; V. Agarwal
and Y. Narayan, ‘Multi-Period Performance Persistence Analysis of Hedge Funds’;
F. Edwards and M. Caglayan, ‘Hedge Fund Performance and Manager Skill’.
15. N. Taleb illustrates this example in the chapter ‘Monkeys on Typewriters’ in his
book Fooled by Randomness: The Hidden Role of Chance in the Markets and in Life.
16. See the following articles: M. Peskin et al., ‘Why Hedge Funds make Sense’,
Quantitative Strategies, Morgan Stanley Dean Witter, November 2000; Crossborder
Capital, ‘The Young Ones’, Absolute Return Fund Research, April 2001; and
M. Howell in ‘Fund Age and Performance’, Journal of Alternative Investment, fall 2001.
A more general audience is addressed in the following article: ‘The Big, the Bold, and
the Nimble’, The Economist, February 24 2001, p.87. See also the discussion in
T. Schneeweiss in ‘Understanding Hedge Fund Performance: Research Results and Rules
of Thumb for the Institutional Investor’, Lehman Brothers Publications, December
2001. Here, evidence is given that smaller funds tend to overperform larger funds but
also have higher risk. Results differ among the various substrategies, though.
17. In corporate finance, leverage describes the ratio of assets to equity and
accounts for the difference between return on equity (ROE) and return on assets
(ROA). ROA is the return on all assets of a company including debt, while ROE
describes the company’s profit regardless of debt levels.
18. See the article ‘Prime Brokers can make and break Hedge Funds’, Pension &
Investments, September 3, 2001. This article refers to the importance of the prime
brokerage relationship, particularly in crisis situations.
19. See also the study by T. Schneeweiss et al., ‘Understanding Hedge Fund
Performance: Research Results and Rules of Thumb for the Institutional Investor’,
Lehman Brothers Publications, December 2001.
20. For a discussion of diversification within a multi-strategy portfolio, see J. Park
and J. Strum, ‘Fund of Fund Diversification: How Much is Enough?’, The Journal
of Alternative Investments, Winter 1999. See also R. McFall Lamm, ‘Portfolios of
Alternative Assets: Why not 100% Hedge Funds?’, The Journal of Investing,
Winter 1999, pp.87–97.
21. For a coverage of the different aspects of ‘risk budgeting’, see the collection
of articles edited by L. Rahl, Risk Budgeting: A New Approach to Investing, 2000.

.... 259 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 260

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

22. See also the article ‘The Benchmark Bane’ in The Economist from August 31, 2001.
23. H. Markowitz, ‘Portfolio Selection’, Journal of Finance, March 1952, p.91.
24. See the following articles for a further discussion: T. Schneeweiss et al.,
‘Understanding Hedge Fund Performance: Research Results and Rules of Thumb for the
Institutional Investor’, Lehman Brothers Publications, December 2001;
T. Schneeweiss and G. Spurgin, ‘Multifactor Analysis of Hedge Funds, Managed
Futures, and Mutual Fund Returns and Risk Characteristics’, Journal of Alternative
Investments, Fall 1998; B. Liang, ‘On the Performance of Hedge Funds’, Financial
Analysts Journal, July 1999; S. Fung and D. Hsieh, ‘Empirical Characteristics of
Dynamic Trading Strategies: The Case of Hedge Funds’, The Review of Financial
Studies, 1997, 10, 2.
25. An interesting discussion of the optimization problem, some illustrations of
the pitfalls of relying on historical data only and the introduction of an alternative
method for Hedge fund performance and risk evaluation is provided by
A. Weismann and J. Abernathy in the article ‘The Dangers of Historical Hedge
Fund Data’, in Risk Budgeting: A New Approach to Investing, 2000.
26. H. Kazemi and T. Schneeweiss discuss the effects of non-normality in Hedge
fund returns for the quantitative portfolio allocation process in their paper
‘Traditional Asset and Alternative Asset Allocation’. They show some possibilities
to deal with the problem through various constraints.
27. See also Chapter 27 in Investment Analysis and Portfolio Management by F. Reilly
and K. Brown for a discussion of performance measures in traditional portfolios.
28. See the article, ‘Hedge Fund Disasters: Avoiding the Next Catastrophe’ by
D. Kramer in The Alternative Investment Quarterly, October 2001.
29. The Association for Investment Management and Research (AIMR) defined
the AIMR-PPS standards aiming to define a well-accepted industry standard for
performance presentation of money managers; see also
http://www.aimr.org/standards/pps.
30. See the paper by M. Dacorogna et al., ‘Effective Return, Risk Aversion and
Drawdowns’ for a discussion of risk-adjusted performance measures. See also Chapter
27 in Investment Analysis and Portfolio Management by F. Reilly and K. Brown.

.... 260 ....

8409 Chapter 7 p205-261 11/4/02 1:14 PM Page 261

7 ■ Active Risk Management in Multi-Manager AIS Portfolios

31. The presence of any of these ‘edges’ contradicts the well-known efficient market
hypothesis (EMH). An abnormal (risk-adjusted) return based on superior information
is a violation of the semi-strong form of EMH, while a statistical edge contradicts the
‘weak form of EMH’. For more information on market efficiency theoriews, see the
seminal papers by Eugen F. Fama, ‘Efficient Capital Markets: A Review of Theory and
Empirical Work’, 1970 and ‘Efficient Capital Markets: II’, 1991.
32. See the discussion by J. Bailey, ‘Are Manager Universes Acceptable
Performance Benchmarks?’.
33. An example is ‘soft dollar’, which refers to an agreement between manager
and broker to pay higher brokerage fees in exchange for added value service to the
manger, such as access to research or risk management tools.
34. A number of risk firms have developed visualization tools recently; see the
article by G. Polyn, ‘Getting a Better Risk Picture’, Risk Management, August 2001.
35. For a detailed discussion of risk budgeting, see Risk Budgeting: A New
Approach to Investing by L. Rahl (ed.).

.... 261 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 262

CHAPTER 8

Conclusion and Outlook

The AIS investment process, risk management,

transparency and liquidity
AIS have had excellent performance records on a standalone basis in recent years. The
case for AIS becomes even more compelling when considered in the context of a global
portfolio, where their low correlations to traditional investments result in significantly
more attractive risk–return properties for the
■■■■■■■■■■■■
overall investment portfolio. At the same time AIS
More and more have been subject to extensive criticism and
investors recognize investor concern after several widely publicized
investment disasters. More and more investors
that evaluating the
recognize that evaluating the ‘risk dimension’ is
‘risk dimension’ is
critical for achieving the benefits of AIS.
critical for achieving the With the institutionalization of the AIS indus-
benefits of AIS try that followed the 1998 Hedge fund crisis has
■■■■■■■■■■■■
come a heated debate over the appropriate
approach to transparency, manager control and
risk management. The different views range from managing risk through extremely
broad diversification, such that a limited number of unavoidable ‘blow ups’ have a

.... 262 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 263

8 ■ Conclusion and Outlook

minimal effect on the portfolio, to a fully transparent, risk-controlled investment

approach including detailed understanding of strategies and managers’ trading edges,
continuous independent monitoring of manager investment activities, and active risk
management. This book has focused on the merits of the second approach, which I
referred to as ‘the transparency paradigm’. The integration of transparency and risk
control into the investment process through the three-pillar framework of strategy
sector allocation, manager due diligence and continuous post-investment monitoring
was outlined. These three elements of an integrated investment and risk management
approach are effective means for preventing investment ‘accidents’.
With detailed knowledge of the various AIS sectors and a good understanding of
individual managers’ trading edges, combined with the appropriate portfolio risk
management investment in AIS allows for the ‘fine tuning of risk’. The skilled investor
can choose from a universe of risk factors and can select those he is willing to accept
and avoid others he does not want to be exposed to. The task of an AIS allocator and
portfolio manager is to ensure that the right types and levels of risk are taken and
continuously monitored. Expected returns should be in line with the risk levels.
Risk management starts with strategy assessment. Investment strategies typically
fulfil various economic functions, which are in particular capital formation, risk
transfer, creation of efficient markets and liquidity generation. The prudent AIS
investor should invest in strategies with sources of return and economic rationales
(risk premiums) he understands, avoid strategies with strong elements of specula-
tion and monitor risk regularly. It is useful to distinguish two sources of AIS
returns: risk premiums and manager skill. The returns of many managers are the
result of premiums for taking particular risks. Other strategies rely more strongly
on the manager’s ability to identify mispricings and forecast future price develop-
ments. The goal of the AIS asset allocation is to be in the right sector during the
right time and avoid too high exposure to the individual sectors’ cycles.
It should be clear that active risk management does not necessarily mean positive
investment performance. Each strategy is exposed to risks and possible hostile
market environments. That is why they earn positive expected returns; return comes
with risk. The essence of risk management is distinguishing ‘good’ from ‘bad’ risks,
controlling the amount of risk and achieving an appropriate level of diversification.

.... 263 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 264

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

The active risk management approach outlined in this book has an important
prerequisite: transparency. Considering the additional types of risk an AIS
investor is exposed to as compared to traditional equity and bond investments,
the ‘black box approach’, i.e. investing in non-transparent and illiquid funds, has
to be truly questioned. A surprising degree of resistance towards transparency
remains in the AIS industry, but the trend in investors’ attitude from accepting
(‘trust me’) to requesting (‘show me’) is clearly observable. Transparency provides
the path along which efficient monitoring and risk management can be imple-
mented. Without transparency, risk analysis has to be performed as guesswork in
an information vacuum. I disagree with the belief that sufficient transparency can
be provided without disclosing underlying positions. Prime brokers seem to share
this view in that they require full position disclosure from their AIS clients (i.e.
Hedge funds and Managed Futures clients).
Further, transparency is not nearly as useful without the appropriate level of
liquidity. Ideally, if investments are of sufficient size, the AIS allocator (fund of
funds manager) should strive to have managers operate on a managed account
basis. Investing via a managed account is the best route to transparency and pro-
vides maximal liquidity. This enables the allocator to exercise control, perform
continuous monitoring and risk analysis and act in timely fashion once problems
are identified. This liquidity can be passed on to the final investor, which creates
the opportunity to structure multi-manager AIS investment products that are as
easy to buy and sell as mutual stock funds (even on a daily basis).
It is sometimes argued that fund of funds managers asking for a high level of
transparency are left with ‘second tier’ managers, as, it is argued, the best per-
forming managers are unwilling to provide full position disclosure. The reality is
that many of the top AIS managers offer managed accounts to investors if the
investment size is sufficiently large. The incorrect belief that non-transparent
funds are the most attractive is the result of persistent misperceptions about AIS.
Many investors regard AIS as an industry where returns are generated by mysteri-
ous means. In fact, most Hedge fund strategies are no mystery and can be
understood very well if studied sufficiently.

.... 264 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 265

8 ■ Conclusion and Outlook

Past performance is far from being a suffi- ■■■■■■■■■■■■

cient indicator for future performance. The
Past performance is far
best way to predict future performance is by
from being a sufficient
understanding the source of generated returns
of the underlying strategy, the strategy’s risks indicator of future
and the environments that are most favourable performance
and unfavourable for the strategy.
■■■■■■■■■■■■
The subject of AIS risk is complex, as is AIS
risk management. Some risk factors in AIS bear no similarity to risks in traditional
bond and equity investment. Proper risk management requires extensive experi-
ence in the field of AIS and a proper appreciation and accounting for the
complexity of their risks. Generally, risk cannot be quantified into a single number
(e.g. VaR) and, for some risk types, it is questionable whether they can be quanti-
fied at all. AIS risk management has quantitative and qualitative aspects. It does
not stop with the sole measurement of risk. Active risk management consists of
numerous guidelines and actions designed to distinguish desired risks from
unwanted risks and to prevent accidents and unacceptable losses. It is both an art
and a science. While the ‘scientific’ aspects refer to the quantitative elements of
risk measurement, the ‘art’ of risk management lies in the understanding of the
complexity of AIS risk and the experience and good judgement necessary to rec-
ognize when corrective actions are necessary.
Both the AIS industry and the standards for the measurement and management
of risk in the financial community are subject to rapid change. Investors in AIS as
well as financial risk managers are confronted with new developments every year.
The combination of these two dynamic fields makes AIS risk management particu-
larly challenging. Keeping up with the latest developments in both AIS and
financial risk methods is key to mastering this challenge.
Many institutional investors have to date remained reluctant to invest more
extensively in AIS instruments, as low levels of transparency and liquidity caused
them a great amount of concern. With the fiduciary role that many current and
potential future investors in AIS take, I predict a general development towards
transparency and active risk management which will enable the industry to continue

.... 265 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 266

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

its rapid growth of recent years. Institutional and private investors are likely to invest
more in AIS once they are convinced that the risks can be systematically addressed.

Regulating AIS?
An important note concerns the issue of regulation. There are increasing calls for
the establishment of a Hedge fund regulation scheme that would be enforced by
international monetary agencies. My argument for transparency is not a request
for AIS regulation. Rather, better transparency and disclosure to investors might
actually diminish the push for regulation. An efficient risk management and con-
trol structure established by the industry itself is surely preferable to a rigid set of
regulations imposed by national or international regulatory authorities as a result
of a political struggle.

The Hedge fund bubble?

The strong growth of the AIS industry has also led to other criticisms. Besides the
lack of transparency, many investors raise concerns about the industry’s capacity.
Many AIS strategies are indeed limited in the amount of money they can manage
without deteriorating returns. This applies in particular to Relative Value
(Convertible Arbitrage, Statistical Arbitrage, Fixed Income Arbitrage) and Merger
Arbitrage strategies, where investors earn specific (risk) premiums, the size of
which depends on the number of other people searching for the same return. But
capacity is also limited for other strategies. Sceptical market participants and
investors compare the recent development of Hedge funds with the technology
bubble in the late 1990s that eventually burst in 2000/2001,1 and they predict
that the current AIS euphoria will similarly end in tears.
Individual AIS sectors could indeed face comparable problems, as AIS investors
are not free from herding behaviour. Many investors focus particularly on those
strategies which have shown good returns in the very recent past (Chapter 4 dis-
cussed some empirical evidence of that). A ‘boom and bust’ cycle was experienced
by Fixed Income Arbitrage strategies in the period from 1995–98, which ended

.... 266 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 267

8 ■ Conclusion and Outlook

with the collapse of LTCM. Another example is the huge allocations to Long/Short
Equity strategies during the equity bull market in the late 1990s. In the 19 months
from March 2000 to September 2001 Long/Short Equity strategies returned –13%
according to the CSFB/Tremont index (–6% for the HFR index). As a third exam-
ple, Convertible Arbitrage and Merger Arbitrage strategies, among the top
performing strategies in the period from 1999–2000, saw huge asset increases in
the same period. Concerns about capacity problems were justified when the merger
market dried up in the wake of a US recession in 2001 and convertible issues began
being priced more aggressively in early 2001. Merger Arbitrage strategies had per-
formance of around zero from March to September 2001 (about –1% according to
the HFR index, +4% for the Tremont (‘Event Driven’) Index, with September
accounting for most of the losses). Convertible Arbitrage strategies yielded lower
returns than before, but continued to have positive performance (around 6%
according to Tremont and 7% according to HFR from March to September 2001).
For the AIS industry as a whole there is not much ground for predicting the phe-
nomenon of a bubble.2 The industry in its entirety is sufficiently well diversified to
deal with extreme market circumstances.
■■■■■■■■■■■■
The AIS industry currently accounts for about
1% of the capitalization of global financial A bubble builds up
markets (equities and bonds are estimated at when expectations
around $50 trillion), so its overall size is still
skyrocket, valuations
comparably small. A bubble builds up when
shift away from
expectations skyrocket, valuations shift away
fundamentals and
from fundamentals and everyone does the same
thing at the same time (homogeneity of expecta- everyone does the same
tions and behaviour). It is important to realize thing at the same time
that opportunities and challenges in AIS differ ■■■■■■■■■■■■
strongly across strategies. The industry per se is
extremely heterogeneous; strategies cover a very broad range of asset classes;
favourable and unfavourable market environments deviate strongly across sectors.
Many Hedge fund strategies are cyclical in their performance. Economic develop-

.... 267 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 268

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

ments, political events and changes in the market environment create and destroy
profit opportunities and some strategies do well in exactly those markets where
others perform poorly.
The downturn of Fixed Income Arbitrage in 1998 was followed by the rise of
other strategies (Long/Short Equity, Convertible Arbitrage, Risk Arbitrage) and the
problems of Macro managers in the late 1990s occurred simultaneously to Fixed
Income Arbitrage strategies showing impressive performance. The industry as a whole
survived the ‘Nasdaq crash’ well and enjoyed massive inflows during this period (it
did, however, show below average return in 2000 and 2001 as compared to historical
performance). Some strategies, in particular the generation of big Global Macro play-
ers, suffered declines and scaled down, but a new generation of successful and highly
skilled Hedge fund managers emerged in the first years of the new millennium.
Finally, investors in diversified AIS fund of funds products were well protected
against large losses in the market turmoil following the tragic events of September
11, 2001, which were a strong stress test for global financial markets in general and
the AIS industry in particular. The S&P500 lost –8.17%, the MSCI World –8.92%
in the month of September 2001. The HFR fund of funds index (net of fees to fund
of fund managers) and CSFB/Tremont Hedge Fund index (gross of fees to fund of
fund managers) displayed performances of –1.61% and –0.83% respectively. While
there were certainly managers on both ends of the spectrum with sharp gains as well
as steep losses, most were able to keep portfolios relatively stable during the highly
volatile month of September 2001. Long/Short Equity strategies lost the most with
–3.45% (–1.57%) according to HFR (CSFB/Tremont), while Futures strategies
gained 3.40% (2.39%) according to CSFB/Tremont (Zurich/MAR).
The opposite view of the ‘bubble-bursting scenario’ for Hedge funds is the view
that Hedge funds have introduced a new paradigm in asset management. The main
underlying argument for this view is that AIS have strong absolute returns and low
correlations to traditional asset classes. But perhaps these benefits are being oversold,
creating unrealistic expectations. A few years ago investing in emerging markets was
proposed as a new way to decrease overall portfolio risk.3 Experiences in the 1990s
have aligned hype with reality. The diversification benefits of AIS are perhaps also
being overestimated, as the AIS industry as a whole has a long equity bias and it is

.... 268 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 269

8 ■ Conclusion and Outlook

debatable whether the AIS industry can develop independently from global equity
trends. Further, given the strong inflows into Hedge funds one has to seriously ask
whether return expectations are decoupling from reality. The lower absolute Hedge
fund performance achieved in 2000 and 2001 may help gradually to align expecta-
tions with reality. I believe that AIS investing will increasingly require strong skill on
the side of the allocator (fund of funds manager, direct investor) to unravel the bene-
fits of Hedge funds and Managed Futures while avoiding excess risk.

Outlook for AIS

Quo vadis, AIS? The ‘bursting of the bubble’ is an unlikely scenario for the AIS
industry. A more realistic, but nevertheless unfavourable, development would be
that investors simply lose interest in AIS if (absolute) returns flatten out. The
downturn of the equity market in 2000/2001 coincided with a period of Hedge
fund underperformance (compared to their his-
torical returns; compared to global equity ■■■■■■■■■■■■
markets AIS showed very strong outperfor-
Quo vadis, AIS?
mance in 2000/2001). The Tremont Composite
index showed performance of only 1.14% for ■■■■■■■■■■■■
the period from March 2000 to October 2001
and the HFR Composite index actually displayed negative performance (–0.75%)
in this period. Considering these numbers the ‘no correlation to equities’ argu-
ment has to be critically reassessed. Again, returns varied widely among the
individual sectors; Relative Value and Futures strategies proved to be good
‘hedges’ in the 2000/2001 equity bear market, while most Long/Short Equity
strategies showed performance significantly below average in this period.
Investors should generally realize that long-term return expectations exceeding
15% per annum (net of all fees) for a diversified AIS portfolio are unrealistic.
Periods of low performance are part of AIS investing as for any other investment.
AIS investing is one of the most dynamic and fastest growing areas in modern
finance. It is interesting to observe that some universities are now incorporating
Hedge funds into their educational programs. Hedge funds are gaining acceptance

.... 269 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 270

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

■■■■■■■■■■■■ as legitimate investments by academics. One

example is the London Business School, which
Steady growth is
is now running a Hedge fund research insti-
much more likely
tute. Change is so rapid that it is extremely
than a ‘boom and difficult to predict what the field will look like
bust’ scenario in just a few years. However, for a number of
reasons steady growth is much more likely than
■■■■■■■■■■■■
a ‘boom and bust’ scenario.
First, the investor base has changed. The AIS industry has developed and will
continue to develop towards satisfying the needs of institutional investors (private
and public pension funds, endowments, family offices, university trusts etc.)
which have started to realize how well AIS fit into their portfolios.
Along with this has come a change in the industry’s attitude about transparency
and investment liquidity as more investors will question the perception that Hedge
funds must, by nature, be non-transparent and secretive. Better investor understand-
ing and a higher degree of transparency will add to the credibility of AIS.
Third, the spectrum of AIS has expanded dramatically. New strategies have
joined the most prominent strategies of the past, Global Macro and Long/Short
Equity. The heterogeneity of AIS provides internal diversification and protection for
the industry from a fallout due to sudden and unforeseen events. The downturn of
equity markets in the early 1970s led to the end of the first ‘Golden Hedge fund’
era, while in sharp contrast during the 2000/2001 equity crisis, which is comparable
in magnitude, investors’ interest in AIS investments increased remarkably.
Fourth, increased interest on the part of (comparably risk averse) institutional
investors together with some well-publicized ‘investment accidents’ has led to the
recognition that AIS risks have to be more systematically addressed. As a result, risk
management techniques have become much more sophisticated. The old ‘cowboy
mentality’ of Hedge funds has given way to new investment managers who are better
and more professional risk managers and thus more credible in the eyes of the investor.
The greatest hurdle for AIS strategies’ further growth is image. The well-
documented problems of LTCM as well as the events surrounding the Manhattan
Investment Fund, where investors were defrauded of $350 million, prompted gen-

.... 270 ....

8409 Chapter 8 p262-271 11/4/02 1:15 PM Page 271

8 ■ Conclusion and Outlook

uine concern about unregulated and uncon- ■■■■■■■■■■■■

trolled Hedge funds potentially putting the
Effective risk
global financial system at risk and causing
management is
investors to lose their entire investments. A fur-
ther focus on increased liquidity, higher essential to
transparency and more sophisticated risk man- continued rapid
agement will surely help to improve investor growth in AIS and to
perception about AIS investing and prevent their unambiguous
further investment disasters that might under- recognition as
mine the future of AIS. Effective risk
a legitimate
management is essential to continued rapid
investment class
growth in AIS and to their unambiguous recog-
nition as a legitimate investment class. ■■■■■■■■■■■■

Notes
1. See, for example, ‘Hedge Funds – The Latest Bubble?’, The Economist,
September 1, 2001; ‘The $500 billion Hedge Fund Folly’, Forbes Magazine, August
6, 2001; ‘The Hedge Fund Bubble’, Financial Times, July 9, 2001. These articles
express a great deal of scepticism towards hedge funds. Unfortunately, their focus
is mainly on polemics rather than discussing the risks and merits of Hedge funds
adequately. It is quite unfortunate that it is mostly non-AIS experts who lead the
discussion about a possible ‘Hedge fund bubble’ and who refuse to account for
some of the basic characteristics of AIS.
2. See L. Jaeger, ‘Is There a Speculative Bubble in Hedge Funds?’, Risk &
Reward, April 2002.
3. For an interesting (unfortunately pre-1997) study on emerging markets, see
the report by C. Barry, J. Peavy, and M. Rodriguez, ‘Emerging Stock Market: Risk,
Return, and Performance’.

.... 271 ....

8409 Glossary p272-278 11/4/02 1:15 PM Page 272

Glossary

Active Discretionary Futures The manager’s Arbitrage Profiting from differences in

approach is primarily fundamental in price when the same or a similar security,
nature. In this regard, trading decisions currency or commodity is traded on two
are based largely on the study of external or more markets.
factors that affect the supply and demand backwardation Pricing structure in
of a market (commodities in particular). commodities in which deliveries at
By monitoring relevant supply and present or in the near future have a
demand factors, a state of dis-equilibrium higher price than those made later on.
of conditions may be identified that has The inverted situation is called contango.
yet to be reflected in the price of the com-
broker Agent who effects securities
modity. Such factors may include weather,
transactions on behalf of and for the
the economics of a particular commodity,
account of the trust or the manager
government policies, domestic and foreign
(securities dealer).
political and economic events and chang-
ing trade product. Capital Asset Pricing Model (CAPM)
Asset pricing model developed by W.
AIS Alternative Investment Strategy, i.e.
Sharpe relating the return of individual
Hedge funds and Managed Futures.
securities to the return of the broad
AIS manager The individual or the firm market. According to CAPM, each
operating a Hedge fund or Managed security’s value depends on a return
Futures strategy. linked to the systematic market risk
alpha Difference between a manager’s (determined by the security’s specific
return and the return appropriate for the ‘beta’) and an unsystematic
given level of risk (as measured by a (idiosyncratic) risk term (which can
‘benchmark portfolio’). be diversified away in a portfolio).

.... 272 ....

8409 Glossary p272-278 11/4/02 1:15 PM Page 273

Glossary

carry Charge for carrying the actual Distressed Security Strategy Involves
commodity instead of a futures investing in debt, equity or trade claims
contract, including interest, storage, and of companies in financial distress.
insurance costs. drawdown Peak to bottom loss in the
CFTC Commodity Futures Trading performance curve of an investment.
Commission of the United States. due diligence Thorough analysis proce-
contango Pricing situation in which dure including assessment, evaluation,
Futures prices are progressively higher as selection of managers for inclusion in an
maturities get longer. The increase asset allocation.
reflects financing, storage (carrying), Equity Market Neutral Involves investing in
insurance and other costs. The inverted securities both long and short, attempting
situation is called backwardation. on average to have a very low net market
convenience yield A certain portion of the exposure. Generally attempts to select long
difference between the spot price and positions that are undervalued and short
the Futures price of a commodity that positions that are overvalued.
cannot be explained by interest rates or Equity Market Timing Involves allocating
storage costs. The convenience yield is assets among investments by switching
simply the price of having a certain into investments that appear to be
commodity available now instead of later beginning an up trend and switching out
(can also be negative). of investments that appear to be starting
Convertible Arbitrage An effort to capital- a downtrend. This primarily consists of
ize on relative pricing inefficiencies, by switching between mutual funds and
purchasing long positions in convertible money markets.
securities, generally convertible bonds, Event Driven Strategy Invests in companies
convertible preferred stock or warrants with ‘special situations’, e.g. mergers,
and hedging a portion of the equity risk restructuring, distress.
by selling short the underlying common Extreme Value Approach An approach to
stock. risk measurement based on the use of
correlation A (linear) measure of the statistical field of Extreme Value
comovement of different assets. Theory (EVT).
CTA (Commodity Trading Advisor) fat tails Extreme areas of the return prob-
Trading advisor of Managed Futures ability distribution that are larger than
strategies, regulated with the CFTC. those of the normal distribution.

.... 273 ....

8409 Glossary p272-278 11/4/02 1:15 PM Page 274

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Fixed Income Arbitrage A market neutral private limited partnerships the SEC
hedging strategy that seeks to profit by limits Hedge funds to sophisticated
exploiting mispricings in fixed income accredited investors.
instruments utilizing a variety of high watermark (also called ‘loss
strategies, including cash vs cash, carryforward’) Method of calculation of
butterflies, basis trading, Treasury vs the performance fee whereby the man-
Eurodollar (TED) spreads, cash vs ager, having incurred a loss for the
Futures. A different type of Fixed investor, is only entitled to further profit
Income Arbitrage is forecasting the sharing by way of the performance fee
change of the shape of the yield curve once this loss has been recuperated.
(duration trade) or profiting from yield
Incremental VaR The change in VaR
differential of issuers with a different
resulting from the inclusion of a particular
quality (credit quality).
investment instrument or sub-portfolio.
fund of funds A multi-manager fund
Jones model The method of investing
which invests in other (AIS) funds.
originally used by Alfred W. Jones in
Futures Derivative taking the form 1949. His method was to invest in US
of a standardized forward exchange stocks, both long and short, to reduce
contract based on currencies, financials, exposure to the broad market and focus
commodities etc. on stock selection.
Global Macro A strategy that employs an leverage Effect achieved when assets of
opportunistic, ‘top down’ approach, the investor are pledged in order to
following major changes in global increase investment exposure. Use of
economies and expecting to profit from derivatives may create the same effect.
significant shifts in the global economy.
long position Ownership of securities,
haircut A reduction from market value in which are not hedged in any way.
computing the value of assets deposited
Long/Short Equity Combines core long
as collateral or margin.
holdings of equities with short sales of
Hedge fund A fund (often not regulated) stock or stock index options. The portfo-
that engages in non-conventional invest- lio may be anywhere from net long to
ment techniques and strategies, making net short depending on market condi-
use of derivatives, selling short positions tions. The manager generally increases
and borrowing and thereby achieving a net long exposure in bull markets and
leverage effect. Because Hedge funds are decreases net long exposure or is even

.... 274 ....

8409 Glossary p272-278 11/4/02 1:15 PM Page 275

Glossary

net short in bear markets. Generally, the Master feeder fund Involves a master
short exposure is intended to generate an trading vehicle domiciled offshore with
ongoing positive return in addition to two investors: another offshore fund and
acting as a hedge against a general stock a US (usually Delaware) Limited
market decline. Stock index put options Partnership (feeder funds). A typical
are also often used as a hedge against structure for a Hedge fund.
market risk. Profits are made when long Mean Variance Approach An approach to
positions appreciate and stocks sold short portfolio analysis based on the premise
depreciate. Conversely, losses are that the determination of the optimal
incurred when long positions depreciate portfolio is possible using only
and/or the value of stocks sold short information about means, variances
appreciates. Long/Short Equity managers’ and co-variances of returns (i.e. no
source of return is similar to that of information about higher moments of
traditional stock pickers on the upside, the return distribution).
but they use short selling and hedging Modern Portfolio Theory (MPT) Theory
in an attempt to limit the risk on the of investment decision making that
downside. permits investors to classify, estimate
managed account Trading account held and control both the kind and amount
with a broker and wholly owned by the of expected risk and return. The core of
fund of funds or directly by the investor. MPT is the assumption that investors
The managers employ their strategies must be compensated appropriately
through executing in the trading account for taking risks. MPT departs from
on behalf of the investor. traditional security analysis as it
describes risk in the context of the
Managed Futures An investment strategy
investor’s entire portfolio rather than on
mainly involving trading on derivatives
the level of the individual security. In
and forward markets. This primarily
MPT, the correlations between the
comprises Futures, as well as other
different instruments in the portfolio are
derivatives in the equity, fixed income
the key components of portfolio risk.
currency or commodity sectors.
net asset value (NAV) The net asset value is
manager see ‘AIS manager’
the aggregate value of all of a fund’s
Margin to Equity The ratio between the investments determined on a middle
margin required to be deposited with the market basis less the aggregate amount of
broker to overall exposure. its liabilities and accrued expenses.

.... 275 ....

8409 Glossary p272-278 11/4/02 1:15 PM Page 276

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

NFA National Futures Association of the Regulation D An Event Driven strategy

United States. where investment usually takes place in
Opportunistic Description for a trading the form of a convertible debenture with
strategy that acts mostly according to an exercise price that floats or is subject
presented opportunities. Opportunistic to a look-back provision. Investment
strategies are usually quite aggressive occurs according the Regulation D US
in nature. Securities Act 1933.

option Derivative embodying the right to Relative Value strategies Seek to generate
buy (in the case of a call option) or sell profits by investing simultaneously in
(in the case of a put option) a fixed related instruments that are, in the eye
number of a particular underlying asset of the manager, mispriced. The goal is
within a specified period and at a to capture the difference between the
predetermined price. current market price and the fair value.

OTC transaction Financial transaction Risk Arbitrage Also known as Merger

that does not take place in a regulated Arbitrage; involves investing in securities
market (‘over the counter’). of companies that are subject to some
form of extraordinary corporate
over-fitting (‘curve fitting’) The systematic
transaction, including acquisition or
optimization of a model on historical
merger proposals, exchange offers, cash
data and the subsequent failure of the
tender offers and leveraged buyouts.
model to perform on untested data.
These transactions will generally involve
performance fee (also called ‘incentive fee’)
the exchange of securities for cash, other
Fee paid to the manager for achieving a
securities or a combination of cash and
certain investment result.
other securities. Typically, a manager
portable alpha The alpha of a manager is purchases the stock of a company being
transported to a target index, e.g. acquired or merging with another
swapping total returns of an equity index company and sells short the stock of the
with the risk free rate. acquiring company. A manager engaged
prime broker Institution that provides the in Merger Arbitrage transactions will
global settlement, clearing and execution derive profit (or loss) by realizing the
of the trades done by an AIS manager. It price differential between the price of
furthermore provides credit lines for the securities purchased and the value
leverage and short selling. ultimately realized when the deal is

.... 276 ....

8409 Glossary p272-278 11/4/02 1:15 PM Page 277

Glossary

consummated. The success of this Statistical Arbitrage Strategy that tries to

strategy is usually dependent on the capture momentary pricing aberrations
proposed merger, tender offer or in the stock market, often by employing
exchange offer being actually quantitative statistical models.
consummated. Stress Testing The process of determining
Scenario Analysis A form of stress testing the vulnerability of the portfolio to an
focusing on the impact of only one or a extreme hypothetical event (scenario).
few specified scenarios. survivorship bias Performance bias in a
SEC Security Exchange Commission of data sample as a result of disappearing
the United States. managers dropping out of the data set.

SFA The Securities and Futures Authority Systematic Futures – Passive Commercial
(United Kingdom). The SFA is the hedging strategies mirroring the inherent
regulatory organization established under returns in the commercial futures markets
the Financial Services Act 1986 with by systematically assuming the opposite
responsibility for regulating members of position to commercial hedgers. These
the organized City investment markets. indices take long or short positions in
different markets depending on the
Sharpe ratio A risk-adjusted performance
supply and demand balance between pro-
measure originally introduced by W.
ducers (e.g. oil company) and consumers
Sharpe consisting of the ratio of return
(e.g. airline). They capture these inherent
minus risk free rate and the standard
returns in the commodity, foreign
deviation of returns. A related measure is
exchange and interest rate futures markets
the Sortino ratio, which uses the down-
by assuming the price risk commercial
side deviation in the denominator and a
hedgers seek to transfer.
minimum required return in the numera-
tor (the ‘Minimal Accepted Return’, Systematic Futures – Technical Model-
MAR), which is subtracted from the based trading in Futures markets at any
time horizon (short, medium, or long
return (instead of the risk free interest).
term). For long-term models (longer
short position Sale of securities, which the
than one month), trend following is
seller does not own.
typically the main strategy. For managers
short rebate The portion of the interest on with a short-term time frame, various
the cash proceed from selling short a statistical tools including momentum
stock that the short seller earns. and countertrend techniques are used.

.... 277 ....

8409 Glossary p272-278 11/4/02 1:15 PM Page 278

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

trading advisor/manager Trading depends on a specific calculation proce-

organization that trades and invests on dure. One distinguishes between the
behalf of the investor via a fund or a variance/co-variance method, the Monte
managed account. Carlo method, and the Historical method.
Value-at-Risk (VaR) The maximum likely volatility Statistical measure for the degree
loss over some specified holding period at of price fluctuations (usually expressed
a particular confidence level. VaR by standard deviation of the returns).

.... 278 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 279

Bibliography

Ackermann, C., McEnally, D. and Ravenscraft, R., ‘The Performance of Hedge Funds:
Risk, Return and Incentives’, Journal of Finance, 2 (1999)
Agarwal, V. and Naik, N., ‘Multi-Period Performance Persistence Analysis of Hedge
Funds’, Journal of Financial and Quantitative Analysis, 35, 2 (2000)
Agarwal, V. and Naik, N., ‘Performance Evaluation of Hedge Funds with Option-based
and Buy-and-Hold Strategies’, Working Paper (September 2000)
Alexander, C., Risk Management and Analysis: Measuring and Modelling Financial Risk,
John Wiley & Sons (1999)
Alexander, C. (ed.) The Handbook of Risk Management and Analysis, John Wiley &
Sons (1996)
Alexander, G., ‘Efficient Sets, Short-selling, and Estimation Risk’, Journal of Portfolio
Management (winter 1995)
Amin, G. and Kat, H., ‘Hedge Fund Performance: 1990–2000: Do the Money Machines
Really Add Value?’, ISMA Centre Working Paper 2001–5 (January 2001)[download
able from www.ssrn.com]
Amin, G. and Kat, H. ‘Welcome to the Dark Side: Hedge Fund Attrition and Survivorship
Bias over the Period 1994–2001’, ISMA Centre Working Paper (December 2001)
Arnott, R. and Pham, T., ‘Tactical Currency Allocation’, Financial Analysts Journal, p.47
(May 1993)
Artzner, P., Delbaen, F., Eber, J. and Heath, D., ‘Coherent Risk Measures’, Mathematical
Finance, 9, 3, p.203 (1999)
Asness, C., Krail, R. and Liew, J., ‘Do Hedge Funds Hedge?’, forthcoming in the US
Journal of Portfolio Management [downloadable from www.aqrcapital.com]

.... 279 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 280

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Bank for International Settlement, ‘Overview of the New Basel Capital Accord’ (January 2001)
Bank of England, ‘Report of the Board of Banking Supervision Inquiry into the
Circumstances of the Collapse of Baring’, HMSO (1995) [downloadable from
http://risk.ifci.ch]
Barry, C., Peavy, J., and Rodriguez, M., ‘Emerging Stock Market: Risk, Return, and
Performance’, The Research Foundation of The Institute of Chartered Financial
Analysts (1997)
Basel Committee on Banking Supervision, ‘Credit Risk Modelling: Current Practices and
Applications’, Basel Committee Publications (1999) [downloadable from www.bis.org]
Basel Committee on Banking Supervision, ‘Review of Issues relating to Highly Leveraged
Institutions’, Basel Committee Publications No. 79 (March 2001) [downloadable from
www.bis.org/publ/bcbs79.htm]
Bernardo, J. and Smith, A., Bayesian Theory, John Wiley & Sons (2001)
Best, P., Implementing Value at Risk, John Wiley & Sons (1998)
Bohn, J., A Survey of Contingent Claim Approaches to Risky Debt Valuation, Haas School
of Business (1999)
Bohn, J., Empirical Assessment of a Simple Contingent-Claims Model for the Valuation of
Risky Debt, Haas School of Business (1999)
Bollerslev, T., Chou, R. and Kroner, K., ‘Arch Modelling in Finance’, Journal of Econometrics,
52, pp.5–59 (1992)
Brock, W., Lakonishok, J. and LeBaron, B., ‘Simple Technical Trading Rules and the
Stochastic Properties of Stock Returns’, Journal of Finance, 5 (December 1992)
Butler, C., Mastering Value at Risk: A Step-by-Step Guide to Understanding and Applying
VaR, Financial Times Prentice Hall (1999)
Caglayan, M. and Edwards, F., ‘Hedge Fund and Commodity Fund Investment Styles in
Bull and Bear Markets’, Journal of Portfolio Management, 27, 4, p.97 (summer 2001)
Capital Market Risk Advisors, ‘NAV/Fair Value Practices Survey Results’ (July 2001)
[downloadable from www.cmra.com]
Cass, D., ‘Hedge Funds to Clear the Information Hurdle’, Risk Magazine, p.8 (October 2001)

.... 280 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 281

Bibliography

Caxton Corporation, Kingdom Capital Management, Moore Capital Management, Soros

Fund Management LLC and Tudor Investment Corporation, ‘Sound Practices for Hedge
Fund Managers’, report issued by a group of five Hedge fund managers (February 2000)
Chalmers, J., Edelen, R. and Kadlec, G., The Wildcard Option in Transacting Mutual-Fund
Shares, Wharton School of Finance (2000)
Clair, C., ‘Hedge Funds Suffer without a Benchmark’, Pension & Investments, p.1
(June 11, 2001)
Clair, C., ‘Prime Brokers can Make and Break Hedge Funds’, Pension & Investments, p.3
(September 2001)
Connolly, K., Pricing Convertible Bonds, John Wiley & Sons (2000)
Cosandy, D., ‘Adjusting Value-at-Risk for Market Liquidity’, Risk Magazine, p.115
(October 2001)
Cottier, P., Hedge Funds and Managed Futures, Verlag Paul Haupt, Bern (1997) [down
loadable from www.aima.org/aimasite/indexfrm.htm]
Counterparty Risk Management Policy Group, ‘Improving Counterparty Risk Management
Practices’, CRMPG (1999)
Credit Suisse Financial Products, ‘CreditRisk+’ (1997)
Crossborder Capital, ‘TSS(II)-Tactical Style Selection: Integrating Hedge Funds into the
Asset Allocation Framework’, Hedge Fund Research (August 2000)
Crossborder Capital, ‘The Young Ones’, Absolute Return Fund Research (April 2001)
Crouhy, M., Galai, D. and Mark, R., ‘A Comparative Analysis of Current Credit Risk
Models’, Journal of Banking and Finance, 24, pp.59–117 (2000)
Crowder, G. and Hennessee, L., ‘Hedge Fund Indices’, Journal of Alternative Investments
(summer 2001)
Cruz, M., ‘Modelling and Measuring Operational Risk’, Journal of Risk, 1, 1 (1998)
Dacorogna M., Gencay, R., Müller, U. and Pictet O., ‘Effective Return, Risk Aversion and
Drawdowns’, Physica A, 289, pp.229–48 (January 2001)
Danielsson, D. and de Vries, C., ‘Value at Risk and Extreme Returns’, Tinbergen Institute
Discussion Paper, TI 98-017/2 (1998) [downloadable from
www.fee.uva.nl/BIEB/TIDPs/TIDP98nr.htm]
Davidson, C., ‘Shopping in the E-market’, Risk Magazine (August 2000)

.... 281 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 282

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Davidson, C., ‘Software Survey 2001’, Risk Magazine (January 2001)

Dowd, K., Beyond Value at Risk, Wiley Frontiers in Finance (1998)
Duffie, D. and Singleton, K., ‘Modelling the Term Structure of Defaultable Bonds’,
Stanford University Paper (1998)
Dunbar, N., ‘French Fight over Hedge Fund Products’, Risk Magazine (February 2001)
The Economist, ‘The Big, the Bold, and the Nimble’, p.87 (February 24, 2001)
The Economist, ‘The Benchmarking Bane’, p.67 (September 1, 2001)
The Economist, ‘The Latest Bubble’, p.59 (September 1, 2001)
Edwards, F. and Caglayan, M., ‘Hedge Fund Performance and Manager Skill’, Working
Paper (May 2001)
Eichengreen, B. and Mathieson, D., ‘Hedge Funds and Financial Markets Dynamics’,
International Monetary Fund Occasional Paper (1998)
Embrechts, P., Klüppelberg, C. and Mikosch, T., Modelling Extremal Events for Insurance
and Finance, Springer (1997)
Embrechts, P., McNeil, A. and Straumann, D., ‘Correlation and Dependence in Risk
Management: Properties and Pitfalls’, Risk Management: Value at Risk and Beyond,
Dempster, M. (ed.) Cambridge University Press, (2002) pp.176–223 [download
able from www.math.ethz.ch/~mcneil/pub_list.html]
Embrechts, P. (ed.) Extremes and Integrated Risk Management, Risk Books (2001)
Fabozzi, F., Fixed Income Analysis, Frank J. Fabozzi Associates, New Hope, Pennsylvania (2000)
Fama, E., ‘Efficient Capital Markets: A Review of Theory and Empirical Work’, Journal of
Finance, 25, 2, p.383 (May 1970)
Fama, E., ‘Efficient Capital Markets: II’, Journal of Finance, 46, 5, p.1575 (December 1991)
Fama, E. and French, K., ‘The Cross Section of Expected Stock Returns’, Journal of
Finance, 47, 2, p.427 (June 1992)
Fama, E. and French, K., ‘Size and Book-To-Market Factors in Earnings and Returns’,
Journal of Finance, 50, 1, p.131 (March 1995)
Fama, E. and French, K., ‘Multifactor explanations of Asset Pricing Anomalies’, Journal of
Finance, 51, p.55 (1996)
Favell, A., ‘To ASP or not to ASP’, Risk Technology, p.14 (October 2001)

.... 282 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 283

Bibliography

Fedorek, T., ‘Is Fraud Flourishing At Your Hedge Fund?’, Pension & Investments, p.14
(March 2001)
Fender, I. and Gibson, M., ‘The BIS census on stress tests’, Risk Magazine, p.50 (May 2001)
Financial Times, ‘The Hedge Fund Bubble’ (July 9, 2001)
Forbes Magazine, ‘The $500 Billion Hedge Fund Folly’ (August 6, 2001)
Frees, E. and Valdez, E., ‘Understanding Relationships Using Copulas’, North American
Actuarial Journal, 2,1, p.1 (1998)
Frey, R., McNeill, A. and Nyfeler, M., ‘Copulas and Credit Models’, Risk Magazine, p.111
(October 2001)
Fung, W. and Hsieh, D., ‘Empirical Characteristics of Dynamic Trading Strategies: The
Case of Hedge Funds’, Review of Financial Studies, 10, 2 (1997)
Fung, W. and Hsieh, D., ‘A Primer on Hedge Funds’, Journal of Empirical Finance, p.309 (1999)
Fung, W. and Hsieh, D., ‘Measuring the Market Impact of Hedge Funds’, Journal of
Empirical Finance, 7, pp.1–36 (2000) [downloadable from
www.faculty.fuqua.duke.edu/~dah7/vitae.htm]
Fung, W. and Hsieh, D., ‘Performance Characteristics of Hedge Funds and Commodity Funds:
Natural versus Spurious Biases’, Journal of Financial and Quantitative Analysis (2000)
Fung, W. and Hsieh, D., ‘The Risk in Hedge Fund Strategies: Theory and Evidence from
Trend-Followers’, Review of Financial Studies, 14, 2, p.313 (summer 2001)
Fung, W. and Hsieh, D., ‘Benchmarks of Hedge Fund Performance: Information Content
and Measurement Biases’, Financial Analyst Journal (2001)
Glassermann, P., Heidelberger, P. and Shahabuddin, P., ‘Portfolio Value at Risk with Heavy
Tailed Risk Factors’, Working Paper, Columbia Business School and Paine Webber
(2000) [downloadable from www.columbia.edu/cu/business/wp]
Goldmann, Sachs & Co. and Financial Risk Management Ltd., ‘The Hedge Fund
“Industry” and Absolute Return Funds’, Journal of Alternative Investments (1999)
Golin/Harris Ludgate, ‘The Future Role of Hedge Funds in European Institutional Asset
Management 2001’, Golin/Harris Ludgate (2001)
Gordy, M., ‘A Comparative Anatomy of Credit Risk Models’, Journal of Banking and
Finance, 24, pp.119–49 (2000)

.... 283 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 284

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Group of Thirty, ‘Derivatives: Practices and Principles’, New York (1993) [downloadable
from http://risk.ifci.ch]
Hauksson, H., Dacorogna, M., Domenig, T., Müller, U. and Samorodnitsky, G., ‘Multivariate
Extremes, Aggregation, and Risk Estimation’, Quantitative Finance, 1, p.79 (2001)
Hayt, G., ‘How to Price Credit Risk’, Risk Magazine (January 2000)
Hedge Fund Research, ‘Market Neutral and Hedged Strategies’ (August 2000)
Hendricks, D., ‘Evaluating Value at Risk Models using Historical Data’, Federal Reserve
Bank of New York, Economic Policy Review (April 1996)
Hull, J., Options, Futures, and Other Derivatives (4th edn) Prentice Hall (1999)
Ineichen, A., ‘In Search of Alpha – Investing in Hedge Funds’, UBS Warburg, London
(October 2000)
Ineichen, A., ‘The Search for Alpha Continues – Do Fund of Hedge Fund Managers Add
Value?’, UBS Warburg, London (September 2001)
Investor Risk Committee (IRC), ‘Hedge Fund Disclosure for Institutional Investors’,
[downloadable from www.iafe.org or from www.cmra.com]
ISI Publications, ‘Starting a Hedge Fund – A European Perspective’ (2001)
ISI Publications, ‘Starting a Hedge Fund – A US Perspective’ (2001)
ISI Publications, ‘The Capital Guide to Alternative Investments’ (2001)
Jaeger, L., ‘Risk Management for Multi-Manager Portfolios of Alternative Investment
Strategies’, AIMA Newsletter (April 2001)
Jaeger, L., ‘The Benefits of Alternative Investment Strategies in the Institutional Portfolio’,
saisGroup (Partners Group) Research Publication (November 2001) [available on
http://www.saisgroup.com]
Jaeger, L., ‘Risk Management for Multi-Manager portfolios of Alternative Investment
Strategies – Part I & II’, ISI Publications, Alternative Investment Quarterly (October
2001, January 2002)
Jaeger, L., ‘Hedge Funds in Marktturbulenze’, Neue Zürcher Zeitung (8 January 2002)
Jaegar, L., ‘Is There a Speculative Bubble in Hedge Funds?’, Risk & Reward (April 2002).
Jaeger, L., Jacquemai, M. and Cittadini, P., ‘The saisGroup Futures Index (sGFI) – A New
Passive Futures Investment Strategy’, saisGroup Research Paper 2000, forthcoming in the
Journal of Alternative Investments [downloadable from www.saisgroup.com]

.... 284 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 285

Bibliography

Jorion, P., ‘Risk Management Lessons from Long-Term Capital Management’, European
Financial Management, 6, p.277 (2000) [downloadable from
www.gsm.uci.edu/~jorion/research.htm]
Jorion, P., The FRM Handbook 2001/2002, John Wiley & Sons, The GARP Risk
Management Library (2001)
Kazemi, H. and Schneeweiss, T., ‘Traditional Asset and Alternative Asset Allocation’,
CISDM/SOM Working Paper, University of Massachusetts (2001)
Keynes, J. M., A Treatise on Money Volume II, Macmillan & Co. (1930)
Kim, J. and Finger, C., ‘A Stress Test to Incorporate Correlation Breakdown’, RiskMetrics
Journal (May 2000)
King, J., Operational Risk: Measurement and Modelling, John Wiley & Sons (2001)
Koyluoglu, U. and Hickmann, A. ‘Reconcilable Differences’, Risk Magazine (October 1998)
Kramer, D., ‘Hedge Fund Disasters: Avoiding the Next Catastrophe’, ISI Publication,
Alternative Investment Quarterly, p.5 (October 2001)
Krokhmal. P., Palmquist, J. and Uryasev, S., ‘Portfolio Optimization with Conditional
Value-At-Risk Objective and Constraints’, Journal of Risk, 4, 2 (2002)
Lake, R., Evaluating and Implementing Hedge Fund Strategies, Euromoney Books (1996)
LeBaron, B., ‘Forecast Improvements using a Volatility Index’, Journal of Applied
Econometrics, 7, p.137 (1992)
LeBaron, B., ‘Some Relation between Volatility and Serial Correlations in Stock Market
Returns’, Journal of Business, 65, 2, p.199 (1992)
Lederman, J. and Klein, R. Hedge Funds: Investment and Portfolio Strategies for the
Institutional Investor, Irwin Professional Publishing (1995)
Lefevre, E., Reminiscences of a Stock Operator, John Wiley & Sons (1994; first published 1923)
Liang, B., ‘Hedge Funds: On the Performance of Hedge Funds’, Financial Analysts Journal
(July 1999)
Longstaff, F. and Schwartz, F., ‘A Simple Approach to Valuing Risky Fixed and Floating
Rate Debt’, Journal of Finance, 50, pp. 449–70 (1998)
Loomis, C., ‘The Jones Nobody Keeps up With’, Fortune, p.237 (April 1966)
Lore, M. and Borodovsky, L. (eds) The Professional’s Handbook of Financial Risk Management,
General Association of Risk Professionals (GARP), Butterworth-Heinemann (2000)

.... 285 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 286

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

McFall Lamm Jr., R., ‘Portfolios of Alternative Assets: Why not 100% Hedge Funds?’,
Journal of Investing (Winter 1998)
McNeil, A. ‘Extreme Value for Risk Managers’, in Internal Modelling and CAD II, Risk
Books, pp.93–113 (1999) [downloadable from www.math.ethz.ch/~mcneil/pub_list.html]
Malz, A., ‘DataMetrics Technical Document’, RiskMetrics Group Publications (2001)
Markowitz, H., ‘Portfolio Selection’, Journal of Finance, 7, 1, p.77 (March 1952)
Markowitz, H., Portfolio Selection – Efficient Diversification of Investment, John Wiley &
Sons (1959)
Merton, R., ‘On the Price of Corporate Debt: The Risk Structure of Interest Rates’,
Journal of Finance, 29, pp.440–70 (June 1974)
Moore, K.M., Risk Arbitrage: An Investor’s Guide, John Wiley & Sons (1999)
Neftci, S., ‘Value at Risk Calculations, Extreme Events, and Tail Estimation’, Journal of
Derivatives, p.23 (spring 2000)
Nelson, R., An Introduction to Copulas, Springer (1999)
Park. J. and Strum, J., ‘Fund of Fund Diversification: How Much is Enough?’, Journal of
Alternative Investments (winter 1999)
Parker, V. R. (ed.) Managing Hedge Fund Risk, Risk Books (2001)
Patel, N., ‘Courting the Hedge Funds’, Risk Magazine, p.31 (November 2001)
Peskin, M., Urias, M., Anjilvel, S. and Boudreau, B., ‘Why Hedge Funds make Sense’,
Quantitative Strategies, Morgan Stanley Dean Witter (November 2000)
Pictet, O., Dacorogna, M., Müller, U., Olsen, R. and Ward, R., ‘Real-time Trading Models
for Foreign Exchange Rates’, Neural Networks World, 26, p.713 (1992)
Pitaro, R., Deals, Deals, and More Deals, Gabelli University Press (1999)
Polyn, G., ‘Value-at-Risk for Merger Arbs’, Risk Magazine, p.6 (April 2001)
Polyn, G., ‘Hedge Funds placed under the Microscope’, Risk Magazine, p.12 (August 2001)
Polyn, P., ‘Getting a Better Risk Picture’, Risk Magazine, p.12 (August 2001)
President’s Working Group on Financial Markets, ‘Hedge Funds, Leverage, and the
Lessons of Long-Term Capital Management’, Report of the President’s Working Group
on Financial Markets (April 1999) [downloadable from http://risk.ifci.ch/146530.htm]

.... 286 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 287

Bibliography

PricewaterhouseCooper, ‘European Private Banking/Wealth Management Survey

2000/2001’, presented at the GAIM Conference, Geneva (June 2001)
Rahl, L. (ed.) Risk Budgeting: A New Approach to Investing, Risk Books (2000)
Reilly, F. and Brown, K., Investment Analysis and Portfolio Management, Dryden Press (1997)
Riley, B., ‘Hedge Funds come in from the Cold’, Financial Times (June 12, 2000)
Risk Magazine, ‘Hedge Funds look to Cultivate Technology Roots’, Risk Magazine, p.S24
(August 2001)
RiskMetrics, ‘RiskMetrics – Technical Document’, (1996) [downloadable from
www.riskmetrics.com]
RiskMetrics, ‘CreditMetrics – Technical Document’, (1997) [downloadable from
www.riskmetrics.com]
RiskMetrics, ‘Return To RiskMetrics – The Evolution of a Standard’, (2001)
[downloadable from www.riskmetrics.com]
Rohrer, J., ‘The Red Hot World of Julian Robertson’, Institutional Investors p.86 (May 1986)
Rosenberg, B., Reid, K. and French, K, ‘Persuasive Evidence on Market Inefficiencies’,
Journal of Portfolio Management, 11, 3, p.9 (Spring 1985)
Ross, S., ‘The Arbitrage Theory of Capital Asset Pricing’, Journal of Economic Theory, 13,
2, p.341 (December 1976)
Saunder, A., Credit Risk Measurement, Wiley Frontiers in Finance (1999)
Schenk, C., ‘Convertible Arb Funds turn to Default Swaps’, Risk Magazine, p.14 (July 2001)
Schneeweiss Partners, ‘A Review of Alternative Hedge Fund Indices’ (2001)
Schneeweiss, T. and Kazemi, H., ‘The Creation of Alternative Tracking Portfolios for Hedge
Fund Strategies’, CISDM/SOM Working Paper, University of Massachusetts (2001)
Schneeweiss, T. and Martin, G., The Benefits of Hedge Funds, Lehman Brothers Publications
(August 2000)
Schneeweiss, T. and Pescatore, J., ‘Alternative Asset Returns; Theoretical Bases and
Empirical Evidence’, Handbook of Alternative Investment Strategies, Institutional
Investors Inc., New York (1999)
Schneeweiss, T. and Spurgin, R., ‘Multifactor Analysis of Hedge Funds, Managed Futures, and
Mutual Fund Returns and Risk Characteristics’, Journal of Alternative Investments (1998)

.... 287 ....

8409 Bibliography p279-288 11/4/02 1:16 PM Page 288

MANAGING RISK IN ALTERNATIVE INVESTMENT STRATEGIES

Schneeweiss, T. and Spurgin, R., ‘Trading Factors and Location Factors in Hedge Fund Return
Estimation’, CISDM/SOM, University of Massachusetts (2001)
Schneeweiss, T., Kazemi, H. and Martin, G., Understanding Hedge Fund Performance:
Research Results and Rules of Thumb for the Institutional Investor, Lehman Brothers
Publications (December 2001)
Schwartz, R. and Smith, C. (eds) Derivatives Handbook, John Wiley & Sons (1997)
Sharpe, W., ‘Capital Asset Prices: A Theory of Market Equilibrium Under Conditions of
Risk’, Journal of Finance, 19, 3, p.425 (September 1964)
Sharpe, W., ‘Asset Allocation: Management Style and Performance Measurement’, Journal
of Portfolio Management, 18, 2 (winter 1999)
Shefren, H., Beyond Greed and Fear: Understanding Behavioural Finance, Harvard Press,
Boston (1999)
Shiller, R., ‘Human Behavior and the Efficiency of the Financial System’, in Handbook of
Macroeconomics, Taylor, J. and Woodford, M. (eds), North-Holland (1999)
Stratman, M., ‘Behavioural Finance: Past Battles and Future Engagements’, Financial
Analysts Journal (1996)
Taleb, N., Fooled by Randomness: The Hidden Role of Chance in the Markets and in Life,
Texere (2001)
Tavakoli, J., Credit Derivatives: A Guide to Instruments and Applications, 2nd edn,
Wiley & Sons (2001)
Till, H., ‘Life at Sharpe’s End’, Risk and Reward, p.39 (September 2001)
Till, H., ‘Measure for Measure’, Risk and Reward, p.33 (October 2001)
Tremont, Partners Inc. and TASS Investment Research Inc., ‘The Case for Hedge Funds’ (1999)
Uryasev, S. and Rockafellar, T., ‘Optimization of Conditional Value-At-Risk’, Journal of
Risk, 2, 3, pp.21–41 (2000) [downloadable from www.ise.ufl.edu/uryasev/roc.pdf]
Watson Wyatt Partners, ‘Alternative Investment Review relating to the Continental
European (the United States, the United Kingdom) Marketplace’, (fall 2000)
Watterson, P., ‘Offering Private Investment Funds in the Capital Markets’, Alternative
Investment Quarterly, p.43 (October 2001)
Weismann, A.and Abernathy, J., ‘The Dangers of Historical Hedge Fund Data’ in Risk
Budgeting: A New Approach to Investing, Rahl, L. (ed.) Risk Books (2000)

.... 288 ....

8409 Index p289-306 11/4/02 1:16 PM Page 289

Index

absolute return strategies 17 manager evaluation (bottom up

account value 143–4 approach) 237–46
active risk management 195–6, 254–7 multi-manager portfolios 208–211
firing managers with ‘issues’ 256 risk management best practices 205–8
leverage controls and margin strategy sector selection (top down
requirements 256 approach) 232–7
regular conversations 255 transparency in AIS 211–20
risk or exposure limits 256 actuarial models 192
risk management 254 administrative risk 142
risk measurement 254
administrator 32, 228–9
stop-loss limits definitioin 256
AIS see Alternative Investment
unusual position explanations 256
Strategies
active risk management in multi-
Algorithmics 197–8
manager portfolios 205–261
alpha return 24, 136
active risk management 254–7
Alternative Investment Strategies (AIS) 1
fee structure and key administrative
considerations 227–9 disclosure 2
Fund Redemption Policies 223–5 investment approach and fund of
investment monitoring: post- funds 8–9
investment risk management investor demand 3–5
246–54 liquidity 13–14
investment/risk management process LTCM 2, 5–6
229–32 new investment paradigm 7–8
leverage 225–7 regulation 266
liquidity of AIS 220–22 risk management 6–7
managed accounts 222–3 risk management challenges 14–15

.... 289 ....

8409 Index p289-306 11/4/02 1:16 PM Page 290

Index

(Alternative Investment Strategies cont’d) MSCI Europe Index 129

risk-reward characteristics 1 MSCI World Index 129
transparency 9–13 S&P500 Index 129
see also universe of AIS sGFI 129
Altvest, New York 131 Zurich Managed Futures Index 129
Arbitrage Pricing Theory (APT) 26, 55 Berger, John 140, 142
ARCH models 177 beta 176
Asian crisis 1997 126–7 return 24
Askari 197 risk 151
asset allocation by strategy sphere 23–4 bets 139, 144, 213
asset-liability mismatch risk 140 black box 1, 3, 145, 215–16
audit oversight 192 BlackRock Solutions 197
Bloomberg 161
authorizations 193
blow ups 4, 213
availability risk, Short Selling 94
bond crash 1994 126–7
borrowing risks, Short Selling 94
back testing and model verification 180
Brazil crisis 126
Barings Bank 191, 194
bubbles 22
BARRA 55, 156, 197
Basel Committee on Banking
calculation of VaR 177–9
Supervision 191, 208
distributional assumptions 178
behaviour in extreme market conditions
historical simulation 179
126–8 Monte Carlo simulations 178–9
Asian crisis 1997 126–7 parametric approach 178
bond crash 1994 126–7 security valuation 177
NASDAQ crash 2000–1 126–8 Cambridge Fund 140
returns 128 capacity constraints 150
Russian default crisis 1998 126–7 capacity risk 140
September 11, 2001 126, 128 Capital Asset Pricing Model (CAPM)
benchmarks 140–41 24–7, 29, 176
benefits in traditional portfolio 129–30 assumptions 25–6
HFR Composite Index 129 leptohurtic (fat tails) return
HFR Fund of Funds (FoF) Index 129 distributions 26
Lehman Brothers US Bond Index 129 non-symmetric return distributions 26

.... 290 ....

8409 Index p289-306 11/4/02 1:16 PM Page 291

Index

capital formation 27 liquidity 168

Equity Market Timing 167 risk management 158–9
Long/Short Equity 167 convexity risk, Fixed Income Arbitrage
Capital Market Risk Advisors (CMRA) 53
198 copula theory 187
Capital Markets, EACM 131–2 corporate event risk 137
Capital Z Investments 132 Convertible Arbitrage 46
catastrophic risk 194 Short Selling 94
causal networks 192 corporate event risk premium 28
central limit theory 186 corporate risk, Distressed Securities 72
Chapter 11, US bankruptcy law 72–3 correlation attributes 111, 173
Reorganizations 70 correlation breakdown risk, Fixed
closed-end vehicles 13, 33, 224 Income Arbitrage 52
Commodity Futures Trading Commission correlation and volatility analysis 248–9
(CFTC) 18, 21, 95, 241 counter-party risk 139
Commodity Pool Operator (CPO) 18, 95 Counter-party Risk Management Policy
Commodity Trading Advisor (CTA) Group (CRMPG) 194
18, 95 country risk, Long/Short Equity 85
funds 21 covariance matrix calculation 177
common factor risk 137 credit risk 137, 147, 189–91
comparison with equities and bonds 119 Black-Scholes (BS) option pricing
complexity premiums 28 formula 190
conditional correlation properties Convertible Arbitrage 46
123–6, 211 Convertible Debenture Arbitrage
confidentiality 10 (Reg D) 77
Convergence Arbitrage 5–6, 212 CreditMetrics (JP Morgan) 189–90
Convertible Arbitrage 40–48 CreditMonitor (KMV) 189–90
diversification 153 CreditPortfolioView (McKinsey)
flight to quality 153 189, 191
Relative Value Arbitrage 168 CreditRisk+ (CSFB) 189, 191
risk management 152–3 Distressed Securities 72
strategies 20, 41–4 Fixed Income Arbitrage 53
Convertible Debenture Arbitrage Merton model 190
(Reg D) 73–8 premiums 29

.... 291 ....

8409 Index p289-306 11/4/02 1:16 PM Page 292

Index

(credit risk cont’d) diversification risk, Futures Passive 105

RiskCalc (Moody’s) 189–90 Donchian, Richard 21
Value-at-Risk (VaR) 172 downside risk management 136
Credit Suisse First Boston drawdowns 111
(CSFB)/Tremont 113, 132 dual entries and reconciliation 193
critical self-assessment 192 Dunn and Hagitt 21
crowd behaviour 98 duration risk
CSFB see Credit Suisse First Boston Convertible Arbitrage 46
CTA funds 21 Convertible Debenture Arbitrage
custodian 33 (Reg D) 78
Distressed Securities 73
Daiwa 191 Fixed Income Arbitrage 53
data providers for AIS 131–4 premiums 29
data risk 140
Dutch tulip mania 22
deal flow risk, Risk Arbitrage 67
deal risk, Risk Arbitrage 66
earnings volatility 192
default risk, Convertible Debenture
economic functions and risk premiums
Arbitrage (Reg D) 77
27–31
delta risk, Convertible Arbitrage 45
Capital Asset Pricing Model (CAPM)
development of AIS 20–24
29
asset allocation by strategy sphere
capital formation 27
23–4
complexity premiums 28
Convertible Arbitrage strategies 20
CTA funds 21 corporate event risk premium 28
Global Macro strategy 20–21 credit risk premiums 29
Hedge funds 20 duration risk premiums 29
Long/Short Equity 20 efficiency premiums 28
Managed Futures 21 equity Market risk premiums 28
directional market risk 146 foreign exchange risk premiums 29
disclosure 2, 218 liquidity 29
Discretionary 100–102 liquidity premiums 29
Distressed Securities 69–73 liquidity provision 27
liquidity 168 manager forecasting 30
risk management 157–8 market efficiency and price
distributional assumptions 178 transparency 29

.... 292 ....

8409 Index p289-306 11/4/02 1:16 PM Page 293

Index

price transparency and efficiency 27 Equity Market Neutral 53–60

Relative Value and Arbitrage Relative Value Arbitrage 168
strategies 28 risk management 155–6
risk premium strategies 27 equity Market risk premiums 28
risk transfer 27 Equity Market Timing 87–90
risk transfer premium 28 capital formation 167
style factors 31 risk management 161
effective return 242 equity risk, Convertible Arbitrage 45
efficiency market hypothesis (EMH) 25 Evaluation Associates 132
efficiency premiums 28 event risk 137
empirical properties of AIS 111–35 Discretionary 102
behaviour in extreme market Distressed Securities 72
conditions 126–8 Equity Market Neutral 58
benefits in traditional portfolio Fixed Income Arbitrage 52
129–30 Global Macro 80
comparison with equities and bonds Long/Short Equity 85
119 execution brokers 33
conditional correlation properties execution risk
123–6 Discretionary 102
data providers for AIS 131–4 Equity Market Neutral 59
performance measurement challenges Equity Market Timing 90
112–13 Fixed Income Arbitrage 53
risk and return 113–19 Futures Passive 105
summary of properties 130–31 Global Macro 81
unconditional correlation properties Short Selling 94
120–23 Systematic Technical 99
enterprise-wide risk management expected shortfall (conditional) VaR 183
172–3, 194–5 exposure analysis 171, 248
equities and bonds comparison 119 external risks 230
MSCI Europe Index 119 extreme market conditions 126–8
MSCI World Index 119 Extreme Value Methodology 192–3
S&P500 119 Extreme Value Theory (EVT) 185–8
Salomon Smith Barney World central limit theory 186
Government Bond Ten Market copula theory 187
Index 119 fat tails 185–6

.... 293 ....

8409 Index p289-306 11/4/02 1:16 PM Page 294

Index

(Extreme Value Theory cont’d) foreign currency risk, Distressed

Fisher-Tippett theorem 186 Securities 72–3
Frechet distribution 186 foreign exchange risk
Gaussian distribution 186 Convertible Arbitrage 47
Generalized Extreme Value (GEV) Global Macro 81
distributions 186 Long/Short Equity 85
Gumbel distribution 186 premiums 29
Weibull distribution 186 Franklin Templeton 88
fraud 144–5
fat tails 111, 185–6 black box 145
Federal Communications Commission, due diligence 144
Risk Arbitrage 63 kickbacks 144
Federal Energy Regulatory manipulated performance statement
Commission, Risk Arbitrage 63
140, 144
Federal Reserve Bank 5, 212
monitoring 144
Risk Arbitrage 63
fraud risk 140
Federal Trade Commission (FTC), Risk
Convertible Debenture Arbitrage
Arbitrage 63
(Reg D) 78
fee structure and key administrative
Risk Arbitrage 69
considerations 227–9
Frechet distribution 186
administrator 228–9
fund of funds 31–4
fees 210, 227
prime broker 228 managers 11
Financial Engineering Associates 197 see also investment approach and
Financial Risk Management 133 fund of funds
firm-wide risk see enterprise-wide risk fund of managed account manager 222
Fisher-Tippett theorem 186 Fund Redemption Policies 223–5
Fixed Income Arbitrage 48–53 closed ended funds 224
liquidity 168 managed accounts 224–5
quantitive tools 154 open ended funds 224
Relative Value Arbitrage 168 structured notes 224–5
risk management 154 funding risk, Fixed Income Arbitrage
flight to quality 220 52
Convertible Arbitrage 153 funds of funds 208–211, 220

.... 294 ....

8409 Index p289-306 11/4/02 1:16 PM Page 295

Index

gamma risk, Convertible Arbitrage 45 risk factors 90

GARCH (General ARCH) models 177 sources of return 89–90
Gaussian distribution 186 strategy 87–9
Generalized Extreme Value (GEV) Global Macro 78–81
distributions 186 risk factors 80–81
Global Macro 78–81 sources of return 80
risk management 159 strategy 78–80
strategy 20–21 Long/Short Equity 82–6
GlobeOp’s 198 risk factors 85–6
Group of 30 (G–30) 194, 206 sources of return 84–5
Gulf War 126 strategy 82–4
Gumbel distribution 186 Short Selling 91–5
example 92
haircut policies 221–2
risk factors 94–5
Hedge Fund Research 132
sources of return 92–3
Hedge funds: Event Driven 38–9
strategy 91–2
Convertible Debenture Arbitrage
Hedge funds: Relative Value 38–9
(Reg D) 73–8
Convertible Arbitrage 40–48
example 75–6
example 43–4
risk factors 77–8
risk factors 44–8
sources of return 76
sources of return 44
strategy 73–6
Distressed Securities 69–73 strategy 40–44
example 71 Equity Market Neutral 53–60
risk factors 72–3 example 56–78
sources of return 71 risk factors 58–60
strategy 69–71 sources of return 57–8
Risk Arbitrage 60–69 strategy 53–7
examples 64–6 Fixed Income Arbitrage 48–53
risk factors 66–9 risk factors 52–3
sources of return 66 sources of return 51–2
strategy 60–66 strategy 48–51
Hedge funds: Opportunistic 38–9 Hedge funds 1, 18, 20
Equity Market Timing 87–90 bubble 266–9
example 88–9 failures 2

.... 295 ....

8409 Index p289-306 11/4/02 1:16 PM Page 296

Index

(Hedge funds cont’d) advisors/managers 32

LCTM 2, 5–6 bubble 22
Manhattan Hedge Fund 2 liquidity 4
Hedge funds and Managed Futures underperformance 143
38–110 investment approach and fund of funds
Hedge funds 38–81 8–9
Managed Futures 82–106 manager evaluation 8–9
Hedgefund.net 132 monitoring/risk assessment 9
Hennessee 132 post-investment risk management 9
HFR Composite Index 129 pre-investment risk management 9
HFR Fund of Funds (FoF) Index 129 sector allocation 8
HFR (Hedge Fund Research) indices 113 see also fund of funds
high investment liquidity 7
investment monitoring: post-investment
Highly Leveraged Institutions Working
risk management 246–54
Group (HLIWG) 208
correlation and volatility analysis
historical simulation 179
248–9
exposure analysis 248
idiosyncratic equity risk 149
independent pricing and performance
Imagine Trading System 198
analysis/attribution 247–8
inappropriate strategy sector allocation
liquidity analysis 252–4
143
marginal and incremantal VaR analysis
incentives 193
independent pricing and performance 251
analysis/attribution 247–8 stress tests and scenario analysis 252
Information ratios 242 Value-at-Risk (VaR) 249–51
institutional investors 4 investment monitoring and control 230
interest rate risk investment/risk management process
Convertible Arbitrage 46 229–32
Distressed Securities 73 external risks 230
Risk Arbitrage 68 internal risks 230
internal and external audits 194 investment monitoring and control
internal risks 230 230
International Organization of Securities key dimensions 231
Commissions 208 manager evaluation 230
investment 262–6 performance characteristics 229

.... 296 ....

8409 Index p289-306 11/4/02 1:16 PM Page 297

Index

risk sources 230–32 leptohurtic return distributions 26

strategy sector selection/allocation 230 leverage 225–7
investor, information management skill risk 139, 150
10 Links Holdings 132
investor demand 1, 3–5 Links Securities LLC 132
black box investing 3 liquidation risk, Fixed Income
blow ups 4 Arbitrage 52
institutional investors 4 liquidity 13–14, 29, 168
investment liquidity 4 of AIS 220–22
risk management 4 analysis 252–4
transparency 4 closed-end investment vehicles 13
Investor Risk Committee (IRC) 12, 207 Distressed Securities 168
InvestorForce 131 Fixed Income Arbitrage 168
funding liquidity 220
Japan stock market crash 1998 126, 137 haircut policies 221–2
Jones, A. W. 20–21 instrument liquidity 220
Jones strategy 82 low 1
J.P. Morgan 172 open-ended funds 12
premiums 29
Kemper 88 provision 27
key people risk, Discretionary 102 Regulation D 168
key risk indicators 192 structured notes 13
kickbacks 144 liquidity risk 137, 148, 188
Kiodex 198 Convertible Arbitrage 47
Kronos 198 Convertible Debenture Arbitrage
kurtosis 111–12 (Reg D) 77
Distressed Securities 72
lack of transparency 149 Fixed Income Arbitrage 52
risk 138 Futures Passive 105
legal risk 142 Global Macro 81
Convertible Arbitrage 47 Long/Short Equity 86
Convertible Debenture Arbitrage Risk Arbitrage 67
(Reg D) 78 Short Selling 94
Distressed Securities 72 Value-at-Risk (VaR) 172

.... 297 ....

8409 Index p289-306 11/4/02 1:16 PM Page 298

Index

LJH Global Investments 132–3 Managed Futures 1, 18, 21

Long/Short Equity 20, 82–6 and Currency strategies 95–6
capital formation 167 model development 163–4
liquidity filters 160 risk management 162–4
risk management 160–61 risk transfer 168
lower partial moments (LPM) 183–4 management company/fund of funds
LTCM (Long-Term Capital investment manager 32
Management) 2, 5–6, 11, 21, manager
194, 205, 212–13, 270 disclosure 10
Convergence Arbitrage 5–6, 212 due diligence 32, 144
Federal Reserve Bank 5, 212 evaluation 8–9, 230, 237–46
systemic risk 143 forecasting 30
transparency 217 risk 138–9, 213–14
strategies and risks 7
managed accounts 33, 222–5 manager due diligence: eight P analysis
fund of managed account manager 241–4
222 partnership 243
fund of managed accounts 223 peers 244
trading advisors 222 people 241
Managed Futures: Active performance 242–3
Discretionary 100–102 portfolio 243
example 101 potential 244
risk factors 102 process 242
sources of return 101–2 product 241–2
strategy 100–101 manager evaluation 8–9, 230, 237–46
Systematic Technical 97–9 background checks 239
examples 97 due diligence 32, 144, 237, 240
risk factors 98–9 eight P analysis 241–4
sources of return 98 market environment analysis 245–6
Managed Futures: Passive 102–6 meeting managers 239
examples 104 risk control 244–5
risk factors 105–6 risks 237
sources of return 104–5 track record 238
strategy 102–4 Manhattan Fund 2, 140, 142, 270

.... 298 ....

8409 Index p289-306 11/4/02 1:16 PM Page 299

Index

manipulated performance statement Equity Market Neutral 59

144 Fixed Income Arbitrage 53
MAR Futures 133 Futures Passive 105
marginal and incremental VaR 183, 251 Global Macro 81
mark to market risk, Distressed Systematic Technical 99
Securities 72 modern portfolio theory (MPT) 164,
market capitalization risk, Long/Short 173–4, 195–6
Equity 85 monitoring 8–9, 144, 193
market efficiency and price transparency Monte Carlo simulations 178–9, 185
29 Morgan Stanley Capital 133
market risk 137, 214–15 MPT see modern portfolio theory
Convertible Debenture Arbitrage
MSCI Europe Index 119, 129
(Reg D) 78
MSCI World Index 119, 129
Discretionary 102
multi-manager funds structure 31–4
Distressed Securities 73
administrator 32
Equity Market Timing 90
closed funds 33
Futures Passive 105
custodian 33
Global Macro 80
execution brokers 33
Long/Short Equity 85
fund of funds 31–4
Risk Arbitrage 68
investment advisors/managers 32
Short Selling 94
Systematic Technical 98 managed accounts 33
Value-at-Risk (VaR) 172 management company/fund of funds
Markowitz, H. 173, 234 investment manager 32
mass psychology 98 manager due diligence 32
Mathlab 156 offshore centres 33
Measurisk 198 open-ended funds 33
Merger Arbitrage see Risk Arbitrage prime broker 32
Mexico crisis 126 structured notes 33–4
Misys 198 Zertifikate 34
MKP Capital Management LLC 139 see also active risk management in
MLM index 96 multi-manager portfolios
model risk 137 multi-manager portfolios 208–211
Convertible Arbitrage 47 funds of funds 208–211

.... 299 ....

8409 Index p289-306 11/4/02 1:16 PM Page 300

Index

NASDAQ crash 2000–1 21, 126–8, Systematic Technical 99

268 Value-at-Risk (VaR) 172
Natal, John 140 operational risk management 193
National Futures Association (NFA) authorizations 193
18, 95, 241 dual entries and reconciliation 193
new investment paradigm 7–8 incentives 193
high investment liquidity 7 internal and external audits 194
manager strategies and risks 7 monitoring 193
monitoring 8 price verification 193
risk management 8 risk-adjusted performance
transparency characteristics 7–8 compensation 193
transparency and manager activity 7 senior management involvement 193
Niederhoffer 11, 217 separation of functions 193
non-directional equity risk 146 over-fitting risk, Systematic Technical 99
non-symmetric return distributions 26
non-transparent strategies 11–12 Panalytix 198
parametric approach 178
offshore centres 33, 143 Park ratios 242
administrator 228–9 Parker Currency Trading Index 133
oil price boom 126 Parker Discretionary Index 133
open-ended funds 12, 33, 224 Parker FX Index 113, 133
operational risk 137, 148, 191–4 Parker Systematic Index 133
actuarial models 192 Pediatrix Medical Group Inc. 92
audit oversight 192 performance characteristics 229
causal networks 192 performance drivers 136
critical self-assessment 192 performance measurement challenges
earnings volatility 192 112–13
Equity Market Neutral 59 Credit Suisse First Boston
Equity Market Timing 90 (CSFB)/Tremont indices 113
Extreme Value Methodology 192–3 HFR (Hedge Fund Research) indices
Fixed Income Arbitrage 53 113
key risk indicators 192 Parker FX Index 113
Long/Short Equity 86 selection bias 113
Risk Arbitrage 69 survivorship bias 112

.... 300 ....

8409 Index p289-306 11/4/02 1:16 PM Page 301

Index

weighting within indices 112 recession 126

Zurich Managed Futures indices 113 redemption periods 1
performance measurement risk 140–41 Reech Capital 198
Pimco 88 registration risk, Convertible Debenture
Ponzi schemes 140 Arbitrage (Reg D) 77
portfolio construction, optimizers 55 regulation, Alternative Investment
portfolio diversification 164–8 Strategies (AIS) 266
capital formation 167 Regulation D see Convertible
liquidity 168 Debenture Arbitrage
Relative Value Arbitrage 168 regulatory risk 143
risk transfer 168 Convertible Arbitrage 47
portfolio risk 142 Relative Value Arbitrage 168
Short Selling 94 Convertible Arbitrage 168
post-investment risk management 9
Equity Market Neutral 168
post-registration risk, Convertible
Fixed Income Arbitrage 168
Debenture Arbitrage (Reg D) 77
Risk Arbitrage 168
pre-investment risk management 9
Relative Value and Arbitrage strategies
pre-registration risk, Convertible
28
Debenture Arbitrage (Reg D) 77
residual risk, Equity Market Neutral 59
prepayment risk, Fixed Income
return 111, 136
Arbitrage 53
Reuters 161, 198
President’s Working Group on
Financial Markets 206 risk
price aggregation 207
transparency and efficiency 27 definition 171
verification 193 monitoring 207
pricing risk 141–2 premium strategies 27
prime broker 32, 228 sources 230–32
PrimeRisk (CSFB) 197–8 technology 15
properties summary 130–31 see also traditional measures
risk in AIS 136–70
quantitative tools 172–3, see also portfolio diversification 164–8
Value-at-Risk (VaR) risk factors 136–43
quantitative valuation models 27 risk factors of different AIS sectors
Quantum Fund 11, 21, 81, 217 145–51

.... 301 ....

8409 Index p289-306 11/4/02 1:16 PM Page 302

Index

(risk in AIS cont’d) performance measurement risk

risk from investor’s view 143–5 140–41
risk management for different AIS portfolio risk 142
sectors 152–64 pricing risk 141–2
risk analysis systems 196–8 regulatory risk 143
outsourcing 197–8 systemic risk 143
Risk Arbitrage 60–69 risk factors of different AIS sectors
Relative Value Arbitrage 168 145–51
risk management 156–7 beta risk 151
risk bucketing see risk budgeting capacity constraints 150
risk budgeting 163, 195, 232 credit risk 147
risk factor mapping 175–6 directional market risk 146
commodities 176 idiosyncratic equity risk 149
equity indices 176 lack of transparency 149
foreign exchange rates 176 leverage risk 150
yield and credit spread curves 176 liquidity risk 148
risk factors 136–43 non-directional equity risk 146
administrative risk 142 operational risk 148
asset-liability mismatch risk 140 yield curve risk 147
capacity risk 140 risk from investor’s view 143–5
common factor risk 137 fraud 144–5
corporate event risk 137 inappropriate strategy sector
counter-party risk 139 allocation 143
credit risk 137 investment underperformance 143
data risk 140 severe decline in account value 143–4
event risk 137 Risk Magazine 198
fraud risk 140 risk management 1, 4, 6–8, 160–161,
lack of transparency risk 138 262–6
legal risk 142 challenges 14–15
leverage risk 139 complexity 6
liquidity risk 137 rapid change 6
manager risk 138–9 risk dimension 6
market risk 137 see also tools and principles of risk
model risk 137 management
operational risk 137 risk management best practices 205–8

.... 302 ....

8409 Index p289-306 11/4/02 1:16 PM Page 303

Index

Basel Committee on Banking risk and return properties of AIS see

Supervision 208 empirical properties
Group of 30 (G–30) 206 risk transfer 27, 168
Highly Leveraged Institutions Managed Futures 168
Working Group (HLIWG) 208 premium 28
International Organization of risk-adjusted performance compensation
Securities Commissions 208 193
Investor Risk Committee (IRC) 207 Risk-Adjusted Return On Capital
LTCM (Long-Term Capital (RAROC) 242
Management) 205 risk-reward characteristics 1
President’s Working Group on RiskBox 198
Financial Markets 206 RiskMetrics 172, 177, 198
risk aggregation 207 RiskWatch (Algorithmics) 197
risk monitoring 207 Robertson, Julian 11, 20–21, 81
strategy drift monitoring 207 Russian default crisis 1998 52, 126–7
risk management for different AIS
sectors 152–64 S&P500 Index 119, 129
Convertible Arbitrage 152–3 Salomon Smith Barney (SSB) 139
Convertible Debenture Arbitrage Salomon Smith Barney World
(Reg D) 158–9 Government Bond Ten Market
Distressed Securities 157–8 Index 119, 129
Equity Market Neutral 155–6 scenario analysis see stress testing and
Equity Market Timing 161 scenario analysis
Fixed Income Arbitrage 154 SEC registration 158–9, 241
Global Macro 159 sector allocation 8
Long/Short Equity 160–61 sector risk
Managed Futures 162–4 Equity Market Neutral 58
Risk Arbitrage 156–7 Equity Market Timing 90
Short Selling 162 Long/Short Equity 85
risk premiums 26, see also economic Securities and Futures Authority (SFA)
functions and risk premiums 95, 241
risk and return 113–19 security valuation 177
AIS strategies 114–15 selection bias 113
Hedge fund strategies 116 senior management involvement 193
Managed Futures strategies 117–18 separation of functions 193

.... 303 ....

8409 Index p289-306 11/4/02 1:16 PM Page 304

Index

September 11, 2001 52, 126, 128, 257 quantitative valuation models 27
settlement risk, Distressed Securities risk premiums 26
72–3 transaction costs 26
severe decline in account value 143–4 specific stock risk, Equity Market
accepted risk 144 Neutral 58
style drift or bet 144 stock risk, Short Selling 94
unanticipated factor 144 strategy
sGFII 96, 129 drift monitoring 207
Sharpe ratio 111, 141, 238, 240, 242 timing 211
Sharpe, W. 173 strategy sector selection 136, 143, 232–7
short rebate risk, Short Selling 95 attractiveness analysis 236
short risk correlations analysis 237
Equity Market Neutral 59 market environment assessment 236
Global Macro 81 risk budgeting 232
Long/Short Equity 86 risk factors assessment 236
Short Selling 94 strategy sector selection/allocation 230
Short Selling 91–5 stress tests and scenario analysis 184-5,
risk management 162 252
short squeezes, Short Selling 94 structural changes risk, Systematic
shorting risk Technical 98
Convertible Arbitrage 47 structure of AIS multi-manager funds
Risk Arbitrage 68 31-4
situation-specific risk, Distressed structured notes 13, 33-4, 224-5
Securities 72 style
skill-based strategies 17 drift 139, 144, 213
Soros, George 11, 20–21, 81 factors 31
Sortino ratios 242 Sumitomo 191
sources of AIS returns 24–7 summary of properties 130-31
alpha return 24 Summit Systems 198
Arbitrage Pricing Theory (APT) 26 SunGard’s Infinity 198
beta return 24 SunGard’s Opus 198
Capital Asset Pricing Model (CAPM) SunGard’s Panorama 198
24–7 survivorship bias 112
efficiency market hypothesis (EMH) Systematic Technical 97-9
25 systemic risk 143, 151

.... 304 ....

8409 Index p289-306 11/4/02 1:16 PM Page 305

Index

tail risk 111-12 transparency 4, 9–13, 211–20, 262–6

Equity Market Neutral 58 best practices 219
Fixed Income Arbitrage 52 black box 215–16
TASS+ 132 characteristics 7–8
tax or legal risk, Fixed Income confidentiality 10
Arbitrage 53 disclosure 218
technology bubble 22, 127 example 218
Tiger Funds 11, 81, 217 fund of fund managers 11
timing risk, Distressed Securities 72 funds of funds 220
TMT crash 126 investor information management
tools and principles of risk management skill 10
171-204 lack of 1, 138, 149
active risk management 195–6 LTCM (Long-Term Capital
credit risk 189–91 Management) 212–13
enterprise-wide risk management manager activity 7
194–5 manager disclosure 10
Extreme Value Theory (EVT) 185–8 manager risk 213–14
Liquidity Risk 188 market risk 214–15
operational risk 191–4 non-transparent strategies 11–12
quantitative tools 172–3 tools and software 10–12
risk analysis systems 196–8 Treynor’s measure 242
stress testing and scenario analysis TRS Associates 133
184–5 Tuna Indices 132
traditional measures and MPT 173–4
Value-at-Risk (VaR) 174–9 Ubitrade 198
VaR back testing and model uncertainty 171
verification 180 unconditional correlation properties
VaR problems 180–82 120–23
VaR variations 183–4 AIS and traditional investment
tools and software 10–12 classes 121
trading advisors 222 universe of AIS 17–37, 223–5
traditional measures and MPT 173–4 definition 17–19
traditional portfolio and AIS see benefits development of AIS 20–24
in traditional portfolio economic functions and risk premiums
transaction costs 26 27–31

.... 305 ....

8409 Index p289-306 11/4/02 1:16 PM Page 306

Index

(universe of AIS cont’d) lower partial moments (LPM) 183–4

Hedge funds 18 marginal and incremental VaR
Managed Futures 18 183, 251
sources of AIS returns 24–7 value/growth risk, Long/Short Equity 85
structure of AIS multi-manager funds Van Hedge Fund Advisors International
31–4 133
US Justice Department, Risk Arbitrage 63 vega risk, Convertible Arbitrage 45
volatility 111, 171
volatility risk, Equity Market Neutral 59
Value-at-Risk (VaR) 172–9, 249–51
back testing and model verification
Weibull distribution 186
180
weighting within indices 112
calculation of VaR 177–9
whip-saw markets 98–9, 105
covariance matrix calculation 177 Wilshire 198
problems 180–82
risk factor mapping 175–6 yield curve risk 147
techniques 154, 157, 159, 163 Fixed Income Arbitrage 53
variations 183–4
see also calculation of VaR Zertifikate 34
Value-at-Risk (VaR) variations 183–4 Zurich Capital Markets 133
expected shortfall (conditional) VaR Zurich Financial Services 133
183, 250 Zurich Managed Futures Index 113, 129

.... 306 ....