Performance Evaluation and Attribution of Security Portfolios Introduction to the Series ii The aim of the Handbooks in Economics series is to produce Handbooks for various branches of economics, each of which is a definitive source, reference, and teaching supplement for use by professional researchers and advanced graduate students Each Handbook provides self-contained surveys of the current state of a branch of economics in the form of chapters prepared by leading specialists on various aspects of this branch of economics These surveys summarize not only received results but also newer developments, from recent journal articles and discussion papers Some original material is also included, but the main goal is to provide comprehensive and accessible surveys The Handbooks are intended to provide not only useful reference volumes for professional collections but also possible supplementary readings for advanced courses for graduate students in economics KENNETH J ARROW and MICHAEL D INTRILIGATOR Performance Evaluation and Attribution of Security Portfolios by Bernd Fischer and Russell Wermers Academic Press is an imprint of Elsevier The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK 225 Wyman Street, Waltham, MA 02451, USA First edition 2013 Copyright © 2013 Elsevier Inc All rights reserved SOLNIK, BRUNO, McLEAVEY, DENNIS, GLOBAL INVESTMENTS, 6th Edition, © 2009, Reprinted by permission of Pearson Education, Inc., Upper Saddle River, NJ 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 the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: permissions@elsevier.com Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is availabe from the Library of Congress ISBN–13: 978-0-12-744483-3 For information on all Academic Press publications visit our web site at store.elsevier.com Printed and bound in the US 12 13 14 15 16 10 9 8 7 6 5 4 3 2 1 Preface v This book is intended to be the scientific state-of-the-art in performance evaluation—the measurement of manager skills—and performance attribution—the measurement of all of the sources of manager returns, including skill-based We have attempted to include the best and most promising scientific approaches to these topics, drawn from a voluminous and quickly expanding literature Our objective in this book is to distill hundreds of both classic and the best cuttingedge academic and practitioner research papers into a unified framework Our goal is to present the most important concepts in the literature in order to provide a directed study and/or authoritative reference that saves time for the practitioner or academic researcher Sufficient detail is provided, in most cases, such that the investment practitioner can implement the approaches with data immediately, without consulting the underlying literature For the academic, we have provided enough detail to allow an easy further study of the literature, as desired We have contributed in two dimensions in this volume—both of which, we believe, are missing in currently available textbooks Firstly, we provide a timely overview of the most important performance evaluation techniques, which allow an accurate assessment of the skills of a portfolio manager Secondly, we provide an equally timely overview of the most important and widely used performance attribution techniques, which allow an accurate measure of all of the sources of investment returns, and which are necessary for precise performance reporting by fund managers We believe that our text is timely An estimated $71.3 trillion was invested in managed portfolios worldwide, as of 2009 (source: www.thecityuk.com) Managing this money, thus, is a business that draws perhaps $700 billion per year in management fees and other expenses for asset managers, in addition to a perhaps similar magnitude in annual trading costs accruing to brokers, market makers, and other liquidity providers (i.e., Wall Street and other financial centers) Our book is the first comprehensive text covering the latest science of measuring the main output of portfolio managers: their benchmarkrelative performance (alpha) Our hope is that investors use these techniques to improve the allocation of their money, and that portfolio management firms use them to better understand the quality of their funds’ output for investors We intend this book to be used in at least two ways: First, as a useful reference source for investment practitioners—who may wish to read only one or a few chapters We have attempted to make chapters self-contained to meet this demand We have also included chapter-end questions that both test the reader’s vi Preface understanding and provide examples of applications of each chapter’s concepts The audience for this use includes (at least) those studying for the CFA exams; performance analysts; mutual fund and pension fund trustees; portfolio managers of mutual funds, pension funds, hedge funds, and fund-of-funds; asset management ratings companies (e.g., Lipper and Morningstar); quantitative portfolio strategists, regulators, financial planners, and sophisticated individual investors Second, the book serves as an efficient way for mathematically advanced undergraduate, masters, or Ph.D students to undertake a thorough foundation in the science of performance evaluation and attribution After reading this book, students will be prepared to handle new developments in these fields We have attempted to design each chapter of this book to contain enough detail to bring the reader to a point of being able to apply the concepts therein, including the chapter-end problems In cases where further detail may be needed, we have cited the most relevant source papers to allow further reading We have divided our book into two sections: Part of the book covers the area of performance evaluation Chapter provides a short overview of the basics of empirical asset-pricing as applied to performance assessment, including basic factor models, the CAPM, the Fama-French three-factor model and the research on momentum, and the characteristic-based stock benchmarking model of Daniel, Grinblatt, Titman, and Wermers Chapter provides an overview of returns-based factor models, and the issues involved in implementing them Chapter discusses the issue of luck vs skill in generating investment returns, and presents the fundamental performance evaluation measures, including those based on the Chapter factor models In addition, extensions of these factor models are introduced that contain factors that capture the ability of portfolio managers to time the stock market or to time securities over the business cycle Chapter presents the latest approaches to using portfolio holdings to more precisely measure the skill of a portfolio manager Chapter provides a complete system for evaluating the skills of a portfolio manager using her portfolio holdings and net returns Many managed portfolios generate non-normal returns Chapter shows how to apply bootstrap techniques to generate more precise estimates of the statistical significance of manager skills in the presence of non-normal returns and alphas Chapter covers a very new topic: how to capture the time-varying abilities of a portfolio manager (as briefly introduced in Chapter 3) Specifically, this chapter shows how to predict which managers are most likely to generate superior alphas in the current economic climate Chapter also covers a very recent topic in performance evaluation: the assessment of the proportion of a group of funds that are truly skilled using only their net returns This approach is very useful in assessing whether the highest alpha managers are truly skilled, or are simply the luckiest in a large group of managers Preface Finally, Chapter is a “capstone chapter,” in that it provides an overview of the research findings that use the principles outlined in the first chapters As such, it is a very useful summary of what works (and what does not) when looking for a superior asset manager (a “SAM”) and trying to avoid an inferior asset manager (an “IAM”) Part of the book primarily concerns performance attribution and related topics Since attribution analysis has become a crucial component within the internal control system of investment managers and institutional clients, ample space is dedicated to a thorough treatment of this field The focus in this part lies on the practical applications rather than on the discussions of the various approaches from an academic point of view This (practitioner’s) approach is accompanied by a multitude of examples derived from practical experience in the investment industry Great emphasis was also put on the underlying mathematical detail, which is required for an implementation in practice Chapter 10 provides an overview of the basic approaches for the measurement of returns In particular, the concepts of time-weighted return and internal rate of return, as well as approximation methods for these measures are discussed in detail Attribution analysis, in practice, requires a deep understanding of the benchmarks against which the portfolios are measured Chapter 11 provides an introduction to the benchmarks commonly used in practice, and their underlying concepts Chapter 12 covers fundamental models for the attribution analysis of equity portfolios developed by Gary Brinson and others Furthermore, basic approaches for the treatment of currency effects and the linkage of performance contributions over multiple periods are considered Chapter 13 contains an introduction to attribution analysis for fixed income portfolios from a practitioner’s point of view The focus lies on a methodology that is based on a full valuation of the bonds and the option-adjusted spread In addition, various other approaches are described Based on the methodologies for equity and fixed income portfolios, Chapter 14 presents different methodologies for the attribution analysis of balanced portfolios This chapter also illustrates the basic approaches for a risk-adjusted attribution analysis and covers specific aspects in the analysis of hedge funds Chapter 15 describes the various approaches for the consideration of derivatives within the common methodologies for attribution analysis The final chapter (Chapter 16) deals with Global Investment Performance Standards, a globally applied set of ethical standards for the presentation of the performance results of investment firms The authors are indebted to many dedicated academic researchers and tireless practitioners for many of the insights in this book Professor Wermers wishes to thank the many investment practitioners that have provided data or insights into the topics vii viii Preface of this book, including through their professional investment management activities: Robert Jones of Goldman Sachs Asset Management (now at System Two and Arwen), Rudy Schadt of Invesco, Scott Schoelzel and Sandy Rufenacht of Janus (now retired, and at Three Peaks Capital Management, respectively), Bill Miller and Ken Fuller of Legg-Mason, Andrew Clark, Otto Kober, Matt Lemieux, Tom Roseen, and Robin Thurston of Lipper, Don Phillips, John Rekenthaler, Annette Larson, and Paul Kaplan of Morningstar, Sean Collins and Brian Reid of the Investment Company Institute Professor Wermers also wishes to thank all of the classes taught on performance evaluation and attribution since 2001—at Chulalongkorn University (Bangkok); the European Central Bank (Frankfurt); the Swiss Finance Institute/FAME Executive Education Program (Geneva); Queensland University of Technology (Brisbane); Stockholm University; the University of Technology, Sydney; and the University of Vienna Special thanks are due to students in that first class of the SFI/FAME program during those dark days in September 2001, 10 days after the 9-11 attacks Professor Wermers is also indebted to his loving family, Johanna, Natalie, and Samantha, for the endless hours spent away from them while preparing and teaching this subject He gratefully acknowledges Thomas Copeland and Richard Roll of UCLA and Josef Lakonishok of University of Illinois (and LSV Asset Management) for early inspiration, as well as Wayne Ferson, Robert Stambaugh, Lubos Pastor, and Mark Carhart for their recent contributions to the field In addition, he owes his career to the brilliant mentoring of Mark Grinblatt and Sheridan Titman at UCLA, pioneers in the subject of performance evaluation This text would not have been possible from such humble beginnings without their selfless support and guidance Dr Fischer is indebted to his colleagues at IDS GmbH—Analysis and Reporting Services, an international provider of operational investment controlling services Over the past years he has greatly benefited from numerous discussions surrounding practical applications Thanks are also due to Dr Fischer’s former team members at Cominvest Asset Management GmbH The design and the implementation of a globally applicable attribution software from scratch, and the implementation of the Global Investment Performance Standards were exciting experiences which left their mark on the current treatise He also wishes to thank various colleagues (Markus Buchholz, Detlev Kleis, Ulrich Raber, Carsten Wittrock, and others), with whom he co-authored papers in the past Several sections in this book are greatly indebted to the views expressed there Dr Fischer is also indebted to the CFA institute and the Global Investment Performance Committee for formative discussions surrounding the draft of the GIPS in 1998/1999 and during his official membership term from 2000 to 2004 Both authors wish to thank J Scott Bentley of Elsevier, whose vision it was to create such a book, and whose patience it took to see it through To those whose contributions we have overlooked, our sincere apologies; such an ambitious undertaking as condensing a huge literature necessitates that the authors Preface choose topics that are either most familiar to us or viewed by us as most widely useful Surely, we have missed some important papers, and we hope to have a chance to create a second edition that expands on this one Finally, to the asset management practitioner: we dedicate this volume to you, and hope that it is useful in furthering your goal of providing high-quality investment management services! ix Chapter An Introduction to Asset Pricing Models ABSTRACT This chapter provides a brief overview of asset pricing models, with an emphasis on those models that are widely used to describe the returns of traded financial securities Here, we focus on various models of stock returns and fixed-income returns, and discuss the reasoning and assumptions that underlie the structure of each of these models 1.1 HISTORICAL ASSET PRICING MODELS Individuals are born with a sense of the perils of risk, and they develop mental adjustments to penalize opportunities that involve more risk.1 For example, farmers not plant corn, which requires a great deal of rainfall (which may or may not happen), unless the expected price of corn at harvest time is sufficiently high Currency traders will not take a long position in the Thai baht and short the U.S dollar unless they expect the baht to appreciate sufficiently In essence, the farmer and the currency trader are each applying a “personal discount rate” to the expected return of planting corn or investing in baht The farmer’s discount rate depends on his assessment of the risk of rainfall (which greatly affects his total corn crop output) and the risk of a price change in the crop The currency trader’s discount rate depends on the relative economic health of Thailand and the U.S., and any potential government intervention against currency Gibson and Walk (1960) performed a famous experiment that was designed to test for depth perception possessed by infants as young as six months old Infants were unwilling to crawl on a transparent glass plate that was placed over a several-foot drop, proving that they possessed depth perception at a very early age Another inference which can be drawn from this experiment is that infants already perceive physical risks and exhibit risk-averse behavior at a very early age (probably before they are environmentally taught to avoid risk) Performance Evaluation and Attribution of Security Portfolios http://dx.doi.org/10.1016/B978-0-08-092652-0.00001-7 © 2013 Elsevier Inc All rights reserved For End-of-chapter Questions: © 2012 CFA Institute, Reproduced and republished with permission from CFA Institute All rights reserved Keywords Asset Pricing Models, CAPM, Factor Models, Fama French three-factor model, Carhart four-factor model, DGTW stock characteristics model, Estimating beta, Expected return and risk ... economics KENNETH J ARROW and MICHAEL D INTRILIGATOR Performance Evaluation and Attribution of Security Portfolios by Bernd Fischer and Russell Wermers Academic Press is an imprint of Elsevier The Boulevard,... most important and widely used performance attribution techniques, which allow an accurate measure of all of the sources of investment returns, and which are necessary for precise performance reporting... Copeland and Richard Roll of UCLA and Josef Lakonishok of University of Illinois (and LSV Asset Management) for early inspiration, as well as Wayne Ferson, Robert Stambaugh, Lubos Pastor, and