International Mergers and Acquisitions Activity Since 1990
Quantitative Finance Series
Aims and Objectives • • • • • •
Books based on the work of financial market practitioners and academics Presenting cutting-edge research to the professional/practitioner market Combining intellectual rigour and practical application Covering the interaction between mathematical theory and financial practice To improve portfolio performance, risk management and trading book performance Covering quantitative techniques
Market Brokers/Traders; Actuaries; Consultants; Asset Managers; Fund Managers; Regulators; Central Bankers; Treasury Officials; Technical Analysis; and Academics for Masters in Finance and MBA market.
Series Titles Return Distributions in Finance Derivative Instruments: theory, valuation, analysis Managing Downside Risk in Financial Markets Economics for Financial Markets Performance Measurement in Finance Real R&D Options Advanced Trading Rules, Second edition Advances in Portfolio Construction and Implementation Computational Finance Linear Factor Models in Finance Initial Public Offerings: an international perspective Funds of Hedge Funds Venture Capital in Europe Forecasting Volatility in the Financial Markets, Third Edition International Mergers and Acquisitions Activity Since 1990 Corporate Governance and Regulatory Impact on Mergers and Acquisitions
Series Editor Dr Stephen Satchell Dr Satchell is the Reader in Financial Econometrics at Trinity College, Cambridge; Visiting Professor at Birkbeck College, City University Business School and University of Technology, Sydney. He also works is a consultative capacity to many firms, and edits the journal Derivatives: use, trading and regulations and the Journal of Asset Management.
International Mergers and Acquisitions Activity Since 1990 Recent Research and Quantitative Analysis
Edited by
Greg N. Gregoriou and Luc Renneboog
AMSTERDAM • BOSTON • HEIDELBERG • LONDON • NEW YORK OXFORD • PARIS • SAN DIEGO • SAN FRANCISCO • SINGAPORE SYDNEY • TOKYO Academic Press is an imprint of Elsevier
Academic Press is an imprint of Elsevier 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, California 92101-4495, USA 84 Theobald’s Road, London WC1X 8RR, UK This book is printed on acid-free paper. ⬁ Copyright © 2007, Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone: (⫹44) 1865 843830, fax: (⫹44) 1865 853333, E-mail:
[email protected]. You may also complete your request on-line via the Elsevier homepage (http://elsevier.com), by selecting “Support & Contact” then “Copyright and Permission” and then “Obtaining Permissions.” Library of Congress Cataloging-in-Publication Data Application submitted British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN 978-0-7506-8289-3 For information on all Academic Press publications visit our Web site at www.books.elsevier.com Typeset by Charon Tec Ltd (A Macmillan Company), Chennai, India www.charontec.com Printed in the United States of America 07 08 09 10 11
9 8 7 6 5 4 3 2 1
Contents
Acknowledgments About the editors List of contributors 1
Understanding mergers and acquisitions: activity since 1990
vii ix xi 1
Greg N. Gregoriou and Luc Renneboog
Part One International M&A Activity and Takeover Performance 2
Cross-border mergers and acquisitions: the facts as a guide for international economics
21 23
Steven Brakman, Harry Garretsen and Charles van Marrewijk 3
Searching for value-enhancing acquirers
51
Manolis Liodakis and Che Pang 4
The Long-term operating performance in European mergers and acquisitions
79
Marina Martynova, Sjoerd Oosting and Luc Renneboog 5
How do bondholders fare in mergers and acquisitions?
117
Luc Renneboog and Peter G. Szilagyi 6
Mix-and-match facilities and loan notes in acquisitions
135
Marc Goergen and Jane Frecknall-Hughes
Part Two 7
Special Types of Mergers and Acquisitions
Mergers and acquisitions in IPO markets: evidence from Germany
167 169
David B. Audretsch and Erik E. Lehmann 8
Reverse mergers in the United Kingdom: listed targets and private acquirers
Peter Roosenboom and Willem Schramade
181
vi
9
Contents
The profile of venture capital exits in Canada
195
Douglas Cumming and Sofia Johan
Part Three Valuation and Irrationality in Takeover Decision Making 10 Executive compensation and managerial overconfidence: impact on risk taking and shareholder value in corporate acquisitions
221 223
Sudi Sudarsanam and Jian Huang 11 Opportunistic accounting practices around stock-financed mergers in Spain María J. Pastor-Llorca and Francisco Poveda-Fuente 12 Size does matter—firm size and the gains from acquisitions on the Dutch market
261
279
Roman Kräussl and Michel Topper Index
295
Acknowledgments
We thank Karen Maloney, Dennis McGonagle, and Anne McGee at Elsevier for guidance throughout the publishing process, as well as Emily Thompson, copyeditor, and Charon Tec (A Macmillan Company) and its team. We are also grateful to all the anonymous referees for carefully reviewing and selecting the final papers during this process.
This page intentionally left blank
About the editors
Greg N. Gregoriou is Professor of Finance in the School of Business and Economics at State University of New York at Plattsburgh. He obtained his Ph.D. (finance) from the University of Quebec at Montreal and is hedge-fund editor for the peer-reviewed scientific journal Derivatives Use, Trading and Regulation and editorial board member for the Journal of Wealth Management, and the Journal of Risk and Financial Institutions. He has written more than 50 articles on hedge funds and managed futures in various U.S. and U.K. peerreviewed publications, including (among others) the Journal of Portfolio Management, Journal of Derivatives Accounting, Journal of Futures Markets, European Journal of Operational Research, Annals of Operations Research, European Journal of Finance, and Journal of Asset Management. He has edited 18 books for Elsevier, Wiley, Palgrave-MacMillan, and Risk and has coauthored one book for Wiley. Luc Renneboog is Professor of Corporate Finance at Tilburg University and a research fellow at the CentER for Economic Research and the European Corporate Governance Institute (ECGI, Brussels). He graduated from the Catholic University of Leuven with degrees in management engineering (MSc) and in philosophy (BA), from the University of Chicago with an MBA, and from the London Business School with a Ph.D. in financial economics. He held appointments at the University of Leuven and Oxford University and visiting appointments at London Business School, European University Institute (Florence), HEC (Paris), Venice University, and CUNEF (Madrid). He has published in the Journal of Finance, Journal of Financial Intermediation, Journal of Law and Economics, Journal of Corporate Finance, Journal of Banking and Finance, Journal of Law, Economics & Organization, Cambridge Journal of Economics, European Financial Management, and others. He has co-authored and edited several books on corporate governance, dividend policy, and venture capital with Oxford University Press. His research interests are corporate finance, corporate governance, dividend policy, insider trading, law and economics, and the economics of art.
This page intentionally left blank
List of contributors
David B. Audretsch is the Ameritech Chair of Economic Development; Director of the Institute for Development Strategies at Indiana University; Director of the Entrepreneurship, Growth and Public Policy Group at the Max Planck Institute in Jena, Germany; and a Research Fellow of the Centre for Economic Policy Research (London). Audretsch’s research has focused on the links between entrepreneurship, government policy, innovation, economic development, and global competitiveness. He has consulted with the World Bank, National Academy of Sciences, U.S. State Department, the United Nations, European Commission, the European Parliament, and the OECD, as well as numerous private corporations and governments. He is a member of the Advisory Board to the Zentrum fuer Europaeisch Wirtschaftsforschung (ZEW) in Mannheim, the Hamburgisches Welt-Wirtschafts-Archiv (HWWA), and the Swedish Foundation for Research on Entrepreneurship and Small Business. His research has been published in more than one hundred scholarly articles in the leading academic journals. He has published thirty books, including, Innovation and Industry Evolution, with MIT Press. He is co-founder of Small Business Economics: An International Journal. He was awarded the 2001 International Award for Entrepreneurship and Small Business Research by the Swedish Foundation for Small Business Research. Steven Brakman is Professor of International Economics at the University of Groningen, The Netherlands, and also holds an appointment at the Radboud University of Nijmegen, The Netherlands. He studied economics at the University of Groningen where he graduated with an MA in 1981. He subsequently worked for the Research Department of the Central Bank of The Netherlands, focusing mainly on monetary issues. He returned to the University of Groningen in 1984 to work on his Ph.D., which was completed in 1991. He has a wide range of research interests, encompassing international economics, economic growth, geographic economics, development economics, and macroeconomics. Douglas Cumming—B.Com. (Hons.) (McGill), M.A. (Queen’s), J.D. (University of Toronto Faculty of Law), Ph.D. (Toronto), CFA, is an Associate Professor of Finance at the Schulich School of Business at York University in Toronto, Canada. He previously taught at the University of Alberta School of Business, the University of New South Wales School of Banking and Finance, and Rensselaer Polytechnic Institute. He has also held the following Visiting Professorships: ABN AMRO Bank Professor of Finance at the University of Amsterdam Graduate School of Business, Center for Financial Studies of the University of Frankfurt Scholar,
xii
List of contributors
University of Cambridge ESRC Centre for Business Research Judge Institute of Management Scholar, and Fellow of Clare Hall at Cambridge. His research is interests span areas in law and finance, venture capital, private equity, IPOs, hedge funds, and market surveillance. His recent publications have appeared in the International Review of Law and Economics, Journal of Banking and Finance, Journal of Corporate Finance, Financial Management, Oxford Economic Papers, and Journal of Business and others. He was the recipient of the 2004 Ido Sarnat Award for the best paper published in the Journal of Banking and Finance. He is a research associate and consultant for a variety of governmental and private organizations around the world. Jane Frecknall-Hughes holds two first degrees in Literae Humaniores and English Language and Literature, respectively from the University of Oxford, and a Ph.D. in revenue law and taxation practice from the University of Leeds. She is a chartered accountant and member of the Institute of Chartered Accountants in England and Wales (ICAEW) and is also a member of the ICAEW’s Tax Faculty. She is a chartered tax adviser and member of the Chartered Institute of Taxation (CIOT). Jane moved from professional practice with KPMG to an appointment at the University of Leeds, and from there, in 2005, to her current post as Senior Lecturer in Accounting and Taxation at the University of Sheffield Management School. Her research interests center on taxation, with particular reference to the development of the taxation profession in the United Kingdom the ethics and morality of taxation, tax planning, transfer pricing, taxation and the multinational enterprise, decision-making and taxation, and taxation/financial history. She has published in Applied Economics, Applied Economics Letters, and the European Management Journal, and has contributed chapters to a number of edited books. Harry Garretsen is Professor of International Economics at Utrecht University in The Netherlands. His primary research area is the new economic geography or geographic economics. Other research areas include monetary issues, macroeconomic theory, and policy making. He is the co-author of a leading textbook on geographic economics (Cambridge UP, 2001) and has published widely in this field in journals such as Regional Studies, Kyklos, Regional Science and Urban Economics, Journal of Economic Geography, Journal of Regional Science, and Journal of Urban Economics. At present, he is co-editor of Spatial Economic Analysis, editorial board member of Papers in Regional Science, and has acted (2006) as guest-editor for Regional Science and Urban Economics. He has edited books as well as contributed to books for Cambridge UP, MIT Press, and Routledge. He is also research fellow of CESifo Munich. Marc Goergen has a degree in economics from the Free University of Brussels, an MBA from Solvay Business School, and a DPhil from the University of Oxford. He has held appointments at the University of Manchester Institute of Science & Technology (UMIST), Manchester Business School, and the ISMA
List of contributors
xiii
Centre (University of Reading). He currently holds a chair in finance at the University of Sheffield Management School. Marc Goergen’s research interests are in corporate ownership and control, corporate governance, mergers and acquisitions, dividend policy, corporate investment models, insider trading, and initial public offerings. Marc has widely published in academic journals such as the European Financial Management, the Journal of Business Finance & Accounting, the Journal of Corporate Finance, the Journal of Finance, and the Journal of Law & Economics. He has also contributed chapters to several edited books and written two books on corporate governance (published by Edward Elgar and Oxford University Press). Marc is a Research Associate of the European Corporate Governance Institute and a fellow of the International Institute for Corporate Governance & Accountability. Jian Huang is a doctoral researcher at Cranfield School of Management, U.K. He has obtained an MSc in Finance with Distinction from Lancaster University Management School, U.K. His research interests are executive compensation, risk taking, and mergers and acquisitions. Sofia Johan is a Ph.D. candidate within the Center for Business Law and a member of the Tilburg Law and Economics Center at the University of Tilburg, The Netherlands. She has been a visiting fellow at the Judge Institute, University of Cambridge. Sofia has an LLB degree from the University of Liverpool and an LLM in International Economic Law from the University of Warwick. Prior to joining Tilburg University, Sofia was Head Legal Counsel of the Malaysian Venture Capital Management Bhd, the largest government-owned private equity management fund in Malaysia. Sofia was also Legal Counsel for the derivatives clearing house in Malaysia. Sofia Johan has published in European Financial Management and is also a research associate with Capital Markets CRC (Sydney). Roman Kräussl obtained a Masters in Economics with a specialization in Financial Econometrics at the University of Bielefeld, Germany. He completed his Ph.D. in Financial Economics on the Role of Credit Rating Agencies in International Financial Markets at Johann Wolfgang Goethe-University, Frankfurt/Main, Germany, in 2002. As the Head of Quantitative Research at Cognitrend GmbH, he was closely involved with the financial industry. Currently, he is an Assistant Professor of Finance at Free University of Amsterdam and a research fellow with the Centre for Financial Studies, Frankfurt/Main. He specializes in Venture Capital and Private Equity and has written numerous papers on these topics. Erik E. Lehmann is Professor of Management, Organization, and Business Administration at the University of Augsburg. Previously, he was an assistant professor at the University of Konstanz and at the Max Planck Institute of Economics. His research focuses on corporate governance, small business economics, and technology management. He has published articles in Research
xiv
List of contributors
Policy, Small Business Economics, Journal of Economic Behavior and Organization, European Finance Review, Journal of Technology Transfer, and other academic journals. Together with David Audretsch and Max Keilbach, he has recently published the book Entrepreneurship and Economic Growth with Oxford University Press. Manolis Liodakis is a managing director and head of the European quantitative research group at Citigroup Investment Research. The group is responsible for producing quantitative research and strategy, as well as advising institutional clients on all aspects of the investment process, from stock selection/screening to portfolio construction and risk management. Since 1999, the team has been ranked among the top three in all major external surveys. It was ranked first in the 2005 Institutional Investor survey for quantitative analysis. Manolis joined the firm in 1999 and has also worked as an emerging-markets strategist for Morgan Stanley Dean Witter. He holds an MBA from the University of Birmingham and a Ph.D. in Finance from Cass Business School. His academic research has been published in various academic and professional journals, including the Journal of Portfolio Management and the Financial Analysts Journal. Marina Martynova is Assistant Professor of Finance at the University of Sheffield. She graduated with a Ph.D. in Financial Economics from Tilburg University and was a research fellow of the European research program, “New Forms of Governance,” coordinated by the European University Institute in Florence. She also graduated from the Center for Economic Research and Graduate Education of Charles University (CERGE-EI) with an MA degree in economics and from St. Petersburg State Engineering-Economic Academy with an MSc in economics and management. Marina is a member of Tilburg Law and Economics Center (TILEC). Her research interests are corporate governance regulation, mergers and acquisitions, corporate finance, dividend policy, and managerial remuneration. Marina’s current research is dedicated to the empirical analysis of regulatory environments and other determinants of mergers and acquisitions patterns in Europe. Sjoerd Oosting graduated with as a Master in Financial Management from Tilburg University and the Helsinki School of Economics. He has worked in the Corporate M&A division of Phillips and is currently employed by a corporate finance consultancy in Amsterdam. Che Pang is an Equity Portfolio Manager at Barclays Global Investors. Prior to joining BGI in 2006, Che spent two years in Citigroup’s European Quantitative Research team as a Quantitative Research Analyst. Che has also spent time as a project manager in the investment strategy group of Ziff Brothers Investments in New York and as a programmer/analyst on the Goldman Sachs Equity Arbitrage desk in London. He holds an MBA from Oxford University and an MEng degree in Computer Science from Imperial College London.
List of contributors
xv
María J. Pastor-Llorca is Assistant Professor of Finance in the Financial Economics Department at University of Alicante, Spain. She obtained her Ph.D. from the University of Alicante on the topic long-run performance of equity issues. Her research is on different topics in corporate finance, mainly on equity offerings. Her current research interest is the study of the impact of accounting practices and analysts’ behavior on the corporate market value. Francisco Poveda-Fuente is Assistant Professor of Accounting in the Financial Economics Department at University of Alicante, Spain. He graduated with a Ph.D. from the University of Alicante, and his research is concentrated on accounting information in capital markets. He specializes in earnings-management practices in the Spanish market and estimates the abnormal component of accounting results. Peter Roosenboom is Associate Professor of Corporate Finance at RSM Erasmus University and a member of ERIM. He holds a Ph.D. in finance from Tilburg University. His research interests include corporate governance, venture capital, and initial public offerings. His work has been published in the Journal of Corporate Finance, European Financial Management Journal, Applied Economics, International Review of Financial Analysis, Pacific-Basin Finance Journal, International Journal of Accounting, and the Journal of Management & Governance. He is the co-editor of the book The Rise and Fall of Europe’s New Stock Markets (in the book series Advances in Financial Economics). He has also contributed chapters to books on initial public offerings, mergers and acquisitions, venture capital, and corporate governance. He is one of the members of a team that conducts a quarterly survey of European CFOs. This joint project with Duke University and CFO Europe Magazine has received media attention in the Wall Street Journal Europe and the survey findings have been discussed on CNN and CNBC Europe. The initiative won the ERIM Impact Award in 2005. Willem Schramade is Assistant Professor of Corporate Finance at the Erasmus School of Economics and advisor in the PriceWatershouseCoopers Valuation & Strategy practice in Amsterdam. He holds a Ph.D. in Finance from RSM Erasmus University and obtained his MSc from Tilburg University. His work is mainly in the area of security issuance and has been published in the Journal of Corporate Finance and the Pacific-Basin Finance Journal. Sudi Sudarsanam is Professor of Finance and Corporate Control at Cranfield School of Management. Previously, he was Professor of Finance and Accounting at Cass Business School in London. His primary interests are corporate restructuring, mergers and acquisitions, valuation of intellectual assets, and corporate strategy. He is one of the leading authorities on mergers and acquisitions in Europe and author of The Essence of Mergers and Acquisitions (Prentice Hall), translated into five European and Asian languages. His recent
xvi
List of contributors
book, Creating Value from Mergers and Acquisitions: The Challenges, an International and Integrated Perspective (FT Prentice Hall, 2003), has received both practitioner and academic acclaim. He is a co-author of Mergers, Acquisitions and Divestitures: Control and Audit Best Practices (The Institute of Internal Auditors Research Foundation, Florida, 2002). He has been a visiting professor at U.S. and European business schools. Sudi has published articles in top U.S. and European journals on mergers and acquisitions, corporate turnaround, corporate governance, and valuation of intellectual assets. Sudi Sudarsanam is also a member of the U.K. Competition Commission and of its Expert Committee on cost of capital. Peter G. Szilagyi joined Tilburg University in The Netherlands in 2003, where he is undertaking doctoral studies in finance. He is a Marie Curie fellow and member of the European Corporate Governance Training Network. Previously, Peter worked as a freelance correspondent with the BBC World Service and as a financial markets consultant for the Asian Development Bank. He has published in the International Journal of the Economics of Business and the Journal of Corporate Citizenship. Peter has also co-authored and edited volumes on European and Japanese fixed-income markets and their derivatives, published by John Wiley & Sons and Elsevier Science, respectively. His current research interests range from shareholder activism, mergers and acquisitions, and payout policy to bond market and financial market development. Michel Topper studied International Financial Economics and Business Administration at the Vrije Universiteit in Amsterdam and the Wirtschafts Universita¨ t in Vienna. After graduation, he joined a U.S. investment bank in London. Charles van Marrewijk is Professor of Economics at Erasmus University Rotterdam, The Netherlands, specializing in International Economics and Economic Growth and Agglomeration. He studied horticulture and worked as a grower before studying economics in Holland at Erasmus University Rotterdam (BA and MA) and in the United States at Purdue University (MSc and Ph.D.). He regularly visits foreign universities (Cornell, Yale, Cambridge, Princeton, Adelaide) and has a wide range of research interests, including international economics, economic growth, geographic economics, development economics, multinationals, and macroeconomics. He has published, for example, in the Review of World Economics, Oxford Economic Papers, Journal of Regional Science, International Journal of Industrial Organization, Journal of International Economics, Journal of Development Economics, Regional Science and Urban Economics, and the International Economic Review. He has also (co-)authored several books, published by Cambridge University Press and Oxford University Press.
1 Understanding mergers and acquisitions: activity since 1990 Greg N. Gregoriou and Luc Renneboog
Abstract This chapter discusses the trends in international market for corporate control. Each mergers and acquisitions (M&A) wave has been characterized by a different set of underlying triggers. However, we consistently find that takeovers early in the wave are triggered by industry shocks. Takeovers are more likely to occur during periods of economic recovery, and the takeover market may be driven by regulatory changes as well as by industrial and technological shocks. Managers’ personal goals may have further impact on takeover activity: We find that managerial hubris and herding behavior tend to increase during takeover waves, often leading to inefficient acquisitions. Finally, takeover activity usually collapses alongside a market decline and an economic recession. The chapter also positions the papers of this book in the international literature.
1.1
Introduction
Understanding the drivers of mergers and acquisitions means understanding their cyclical nature (see Golbe and White, 1993, for one of the earliest documentations of this phenomenon). It is commonly accepted that there have been five waves of major merger activity: the 1890s, the 1920s, the 1960s, the 1980s, and the 1990s. The scale of the final wave is remarkable for its breadth and geographic distribution. This wave saw tremendous U.S. M&A growth, but it was also witness to soaring levels of European M&A activity, as firms started to partner actively with U.S. and U.K. firms. M&A activity has been on the rise again since June 2003, perhaps suggesting a new wave. This recent increase in takeover activity could have wide-ranging ramifications and raises many interesting questions. We briefly review the historical and recent literature on M&A activity by wave for the U.S., U.K. and Continental Europe. We find that takeover activity is often triggered by excessive heterogeneity, generally ending with some type of economic shock such as a recession. Economic recovery seems to drive takeover waves, which often coincide with periods of rapid credit expansion. Regulatory changes are also important drivers of takeover waves. The earlier waves of the 1890s and 1920s are believed to have been driven by antitrust legislation, while that of the 1980s appears to have been brought on by widespread market deregulation (Martynova and Renneboog, 2005).
2
1.2 1.2.1
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Historical background The 1890s and the 1910s to the 1920s: the first and second waves
The first wave of mergers in the 1890s was generated by an economic depression, legislation governing incorporation, and the rise of industrial stocks (see, e.g., O’Brien, 1988). The main goal of this first wave was to consolidate industrial production and reduce competition (Lamoreaux, 1985). This wave led to the creation of companies that became virtual monopolies in their respective industries. The equity market crash caused this first wave to come to an end around 1905. M&A activity stayed at moderate levels from then until the late 1910s, largely owing to World War I. Around 1910, antitrust legislation began to take hold both in the U.S. and Europe, probably as a result of the previous monopolization attempts. The only option for firms desiring to expand was vertical expansion; thus this second wave can be seen as creating oligopolistic structures (see Stigler, 1950). The resulting conglomerates of the 1920s focused on economies of scale (for detailed studies of the first and second merger waves, see, e.g., Eis, 1969, Markham, 1955, and Thorp, 1941).
1.2.2
The 1950s to the 1970s: the third wave
Several decades passed before the advent of a new takeover wave, largely owing to the economic depression of the 1930s and World War II. The third M&A wave is widely accepted to have taken off during the 1950s and to have come to an end in 1973 as a result of the oil crisis and subsequent recession. As Sudarsanam (2003) notes, here we see a difference between U.S. and U.K. takeover activity: Whereas U.S. takeovers focused on creating large conglomerates, the hallmark of U.K. takeovers at this time was horizontal integration (see Fairburn, 1989, for a more detailed discussion). It is notable, however, that the beginning of this third M&A wave in the U.S. coincided with tighter antitrust regulations—regulations that not only made horizontal expansion more difficult, but caused more firms to combine with those outside their industries. As Matsusaka (1996) notes, though, some countries that did not have such tough antitrust policies, such as Canada, Germany, and France, also saw a wave of diversification during the 1960s. It is likely that, during this time, companies were beginning to search more actively for opportunities to boost value and reduce earnings volatility. There is more than one plausible explanation for the rise of the third M&A wave. Diversifications during the 1960s can be attributed to such assorted causes as stricter antitrust regulations, less well developed external capital markets, and labor inefficiencies, as well as a host of economic, social, and technological changes (for additional explanations of the motives behind this third takeover wave, see, e.g., Lintner, 1971, Markham, 1973, and Reid, 1968).
Understanding mergers and acquisitions: activity since 1990
1.2.3
3
The 1980s: the fourth wave
The fourth takeover wave is widely accepted to have ranged from 1980, at which time the stock market had regained its footing after the economic recession, through 1989. It was a time of antitrust policy changes, financial services deregulation, new financial instruments and markets, and increased technological progress. There were also a record number of divestitures, hostile takeovers, and transactions such as leveraged buyouts (LBOs), suggesting increased investor focus on corporate control (Renneboog and Simons, 2006; Renneboog, Simons and Wright, 2007). This fourth takeover wave appears to have emerged as a result of the inefficiencies created by the previous wave’s diversifications (Bhagat, Shleifer, and Vishny, 1990; and Shleifer and Vishny, 1991). The hallmarks of this wave included loosened antitrust regulations, more competitive capital markets, and improved shareholder control. Companies began to see the benefits of “de-diversifying” and refocusing on core business ideals (Blair, 1993). This decade also saw the rise of hostile raiders, who were always ready to swoop in and pick off slower, less efficient companies. Some authors believe that the outside capital markets had also become more efficient, owing to the host of economic, technological, and regulatory changes seen during the 1980s (Martynova and Renneboog, 2006a). This may have begun to render internal capital markets less necessary (Bhide, 1990). But the structure of the conglomerate was also starting to be seen as inefficient. Its size meant it was slow to react to shocks caused by deregulation, political events, or economic factors (Mitchell and Mulherin, 1996; see also Jensen, 1986 and 1993; Morck, Shleifer, and Vishny, 1988; and Andrade and Stafford, 2004). For example, in the medical and pharmaceuticals sectors, the introduction of a new reimbursement policy in 1983 triggered a wave of takeover activity aiming to take advantage of potential cost reductions. In the oil sector, political events such as the 1973 OPEC embargo set off a wave of corporate restructuring. And in food processing, low population growth during the 1980s drove a wave of restructuring. To conclude, the drivers of the takeover wave of the 1980s include industrial shocks, the reining in of managerial power, and the trend toward smaller, more nimble companies. Activity at this time was driven further by more and stricter disclosure of corporate information and the subsequent focus on maximizing shareholder value.
1.3 1.3.1
Recent M&A activity The 1990s: the fifth wave
It is commonly accepted that the fifth takeover wave, unprecedented in both deal value and deal volume, began in 1993. It also took off alongside an economic bull
4
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
market, then collapsed in 2000, a victim of the equity market downturn that year. The United States had approximately 119,000 M&A deals during this wave, and Europe had 117,000 (these data come from the Thomson Financial Securities Database). In comparison, the fourth wave had only 34,000 and 13,000, respectively, in the United States and Europe. But the fifth wave dwarfs the fourth wave in other ways: Total (global) value reached USD 20 trillion, more than five times the total of the fourth wave (Martynova and Renneboog, 2006 a,b). The fifth takeover wave saw dramatically more activity abroad as well. In fact, during this period, the European wave was almost as large as the U.S. wave, and a substantial takeover market emerged in Asia. Many M&As conducted during this fifth wave were cross-border transactions, reflecting the increase in capital market globalization. Stronger competition from abroad meant that U.S. companies needed to consider takeovers in other countries just to survive. The increase in deregulation and privatization during this period tended to trigger cross-border acquisitions in sectors such as finance and telecoms. According to the Thomson Database, M&A activity during the fifth wave, whether crossborder or domestic, occurred primarily intra-industry. The proportion of M&A divestitures, although still relatively high, was decreasing. This indicates that the main takeover motive during the 1990s wave was growth, which was necessary to participate in global markets. But to expand, companies need financing, and they may choose to issue equity or debt to get it. Thus we see a relationship between the bull market of the 1990s and the widespread use of equity in M&A deals (see Shleifer and Vishny, 2003). Bidders used equity to buy assets of undervalued companies. We suggest that the mispricing premium was an important source of M&A value during this period. The corporate bond market also grew tremendously during this period. The higher amount of activity during this wave may also have been driven by lower interest rates and easier credit terms (Renneboog and Szilagyi, 2007). Note that the number of hostile bids in the United Kingdom and the United States fell dramatically during the 1990s compared to the 1980s, according to the Thomson Database. This decline may be attributable to the bull market: Target shareholders have been shown to be more receptive to takeover bids when their shares are overvalued (Martynova and Renneboog, 2006b). Regulatory changes during the 1980s are also responsible for the decrease in hostile takeovers. Strict anti-takeover laws were enacted at this time in some states. And Holmström and Kaplan (2001) put forth another reason: the rise of alternative governance mechanisms, such as stock options and shareholder activism, which may mean that hostile takeovers are no longer the preferred means of policing management behavior. Note that, interestingly, hostile takeover activity in continental Europe increased during the 1990s. In fact, it began to be seen even in countries with no history of hostile takeovers. In sum, the fifth wave of M&A activity was driven by a wide range of factors, with globalization playing perhaps the largest part, followed by technological innovation, the financial bull market, deregulation, and privatization. Many articles posit that takeovers at that time were mainly concerned with
Understanding mergers and acquisitions: activity since 1990
5
cost-cutting, expansion overseas, and exploiting over- or undervaluations. Goergen, Martynova, and Renneboog (2005) discuss the impact of regulation on M&A activity. But several important empirical studies have shown that M&A deals undertaken in the late 1990s may have actually destroyed value (e.g., Moeller, Schlingemann, and Stulz, 2005).
1.3.2
From 2003: the sixth wave?
Because takeover activity has been increasing since 2003 in the United States, Europe, and Asia, we may be seeing what will become a sixth wave. As with the other waves, this wave seems to have been triggered by the market recovery after the 2000 downturn. According to the Thomson Database, M&A volume saw a 71% increase in 2004, for a total of about USD 1 trillion, compared to 2002 when it totaled about USD 500 billion. A similar trend has been seen in Europe. In 2004, total takeover value was approximately U.S. USD 760 billion, up from USD 517 billion in 2002. In fact, cross-border acquisitions from 2002 through mid-2005 account for more than 43% of the total value of all European M&A’ and 13% of the total value of all U.S. M&A’. In China, the numbers have also increased dramatically, from about U.S. USD 3 billion in 2002 to almost USD 19 billion in the first half of 2005. We cannot draw conclusions yet about the drivers of any new wave, but some things are apparent. First, the events of September 11, 2001, are believed to have played a large part, causing a delay in certain transactions that are now coming to fruition. Second, there has been an increase in governments’ selling shares in major national companies, thus increasing the supply of target firms (this is especially true in China). Third, firms afloat with cash from the recent bull market seem to be seeking to expand into new markets. And, fourth, private equity investments in sectors like real estate and retail, have escalated dramatically recently (Wright, Renneboog, Scholes, and Simons, 2006).
1.4
M&A clustering: theory
We now briefly discuss the theoretical models behind the motives for takeovers. There are three main groups: (1) neoclassical models, which suggest that takeover waves emerge from industrial, economic, political, or regulatory shocks; (2) models proposing that takeover clustering is driven by the self-interest of managers, e.g., herding, hubris, or agency problems; and (3) models attributing takeovers to general capital market development, thereby positing that waves occur as a result of managerial market timing.
1.4.1
Neoclassical models
Neoclassical models revolve around the rational economic factors that motivate firms to restructure simultaneously. Coase (1937) was an early proponent of the
6
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
model suggesting that takeover activity is driven by technological change. A later model by Gort (1969) claims that economic disturbances, such as market disequilibrium, may cause wholesale industry restructuring. Jovanovic and Rousseau (2001, 2002) built on Gort’s theory, and developed the Q-theory of takeovers, which posits that economic and technological changes cause a higher degree of corporate growth opportunities. Such changes may cause capital to be reallocated to more productive and efficient firms. Some authors explain takeover activity by citing the relationship between industry-specific shocks and the availability of low-cost capital. Harford’s (1999) model predicts that M&As are more likely to occur when companies have large cash reserves or less access to external financing (see Martynova and Renneboog, 2006b, for empirical evidence). Thus takeover clustering occurs in periods of capital market growth. Neoclassical models explain takeover clustering by industry and by country. But waves can also result from firms’ responding to the actions of their competitors. Thus, if one firm conducts a series of successful M&As, this may increase the resolve of other firms, especially in the same industry, to follow suit (Persons and Warther, 1997).
1.4.2
Models of managerial hubris, herding, and agency problem
As we see in the empirical literature, a significant percentage of M&As may be considered to have destroyed rather than added value. Thus some theoretical models focus on facets of managerial decision-making to explain this phenomenon. Jensen’s (1986, 2004) agency explanation cites overcapacity generated by industrial shocks or financial bull markets. Roll (1986) focuses on managerial hubris, positing that overconfident managers overestimate the creation of synergistic value. The hubris and herding hypotheses may provide additional explanations for the cyclical nature of M&A activity (examples of financial herding models are cited in Scharfstein and Stein, 1990; Graham, 1999; Boot, Milbourn, and Thakor, 1999; and Devenow and Welch, 1996). Herding refers to firms’ imitating the actions of a leader or a first-mover firm. Thus successful takeovers may encourage other companies to try for similar transactions. However, the other companies may not be acting from clear economic foundations; hence their takeovers may not result in the same efficiency. Thus herding combined with hubris may mean that inefficient takeovers are more likely to follow efficient ones. Note that there is also a behavioral explanation for takeover waves, as suggested by Auster and Sirower (2002). They put forth three distinct stages of a takeover wave: development, diffusion, and dissipation. The way a wave develops is determined by macro factors and the competitiveness of the environment. They hypothesize that if M&A activity does not result in positive economic outcomes, market forces will cause it to decline rapidly. An alternative view put forth by Gorton, Kahl, and Rosen (2000) shows that value-destroying takeovers can also precede profitable ones. These authors posit that managers will always prefer to keep their firms independent, and may
Understanding mergers and acquisitions: activity since 1990
7
actively take over other firms as a defense against being taken over themselves. If managers believe they are in danger of being taken over, this fear may result in a wave of inefficient takeover activity.
1.4.3
Models of managerial market timing
Models seen in more recent articles posit that takeover waves arise from managerial market timing. Following Myers and Majluf (1984), these models hypothesize that managers may use equity overvaluations to acquire real assets. But takeover waves may also occur because financial bull markets tend to overvalue stocks in the short run (Shleifer and Vishny, 2003). Overvaluation can vary significantly from company to company, so a bidding firm can actually purchase the assets of a less overvalued firm by using their own (overvalued) equity (in other words, this is an example of the mispricing premium). The assumption here is that target managers will maximize their own short-term benefits and will accept an all-equity bid, even at the expense of target shareholders. Following the predictions of this model, takeover waves are seen as positive for stock market value because managers can take advantage of temporary market inefficiencies.
1.5
Empirical evidence on M&A profitability
The empirical literature on M&A profitability is extensive. Several surveys provide useful overviews (see Jensen and Ruback, 1983; Jarrell, Brickley, and Netter, 1988; Sudarsanam, 2003; Martynova and Renneboog, 2005). There are many ways to assess the success of a takeover. We can evaluate M&As either from the perspective of the target’s shareholders or the bidder’s shareholders, or we can calculate the combined shareholder effect. Event studies are the predominant approach for analyzing short-term shareholder wealth effects. The pivotal point in the event study approach is that the M&A announcement constitutes new market information, and that investor expectations will be updated immediately and reflected in the share price. The difference between the realized returns and a benchmark return would equal an abnormal return if the takeover bid did not take place. The benchmarks that are commonly calculated use asset pricing models like the market model. Some studies also calculate the operating performance of the merging firms by comparing measures like net income, sales, and return on assets or equity before and after the takeover. However, because operating performance is also affected by a variety of other factors, this approach has limitations. To compensate, the literature suggests adjusting for the performance of merging firms for industry trends as well as for size and book-to-market ratios of nonmerging companies, such that the question of whether or not merging companies outperform nonmerging ones before and after the bid can be answered effectively.
8
1.6
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Short-term wealth effects
According to the empirical literature, takeovers create value for the target and bidder shareholders combined (with the target shareholders reaping the majority of the gains). But the results for the bidder shareholders are mixed: Some gain small positive abnormal returns; others suffer small losses. See Table 1.1 for an overview.
1.7
Total takeover gains
As Table 1.1 shows, takeovers on average are expected to increase the value of the combined firms’ assets. The target shareholders tend to earn large positive abnormal returns, and the bidder shareholders are not likely to lose value. An interesting study by Bradley, Desai, and Kim (1988) found an abnormal return of 7–8% over the period 1963–1984 accruing to an investor who, having owned an equal share in both the bidder and the target one week before the event date, then sold the shares one week later. For the period 1985–2000, Bhagat, Dong, Hirshleifer, and Noah (2004) found a decrease in total takeover gains over this period compared to previous decades. Bhagat Dong, Hirshleifer, and Noah (2004) and Harford (2003) found that total announcement wealth effects of M&As from periods outside the takeover waves are always significantly lower than gains earned during waves. Interestingly, both studies also found that the highest combined M&A gains come at the beginning of takeover waves.
1.8
Operating performance
To gauge the combined gains of takeovers, 25 major accounting studies have been conducted. Post-merger, 14 studies found a decline in the profitability of merging firms, 6 studies found firm profitability to be changed insignificantly, and 5 found a significantly positive increase in operating returns (see, e.g., Ravenscraft and Scherer, 1987; Linn and Switzer, 2001; and Carline, Linn, and Yadav, 2002). But if we consider post-merger corporate growth, the results are less clear. Cosh, Hughes, and Singh (1980) found that post-merger asset growth of U.K. companies that conducted M&As from 1967 to 1969 improved systematically. Mueller (1980) found a significant decline in the growth rate of U.S. companies during the third wave. Ghosh (2001), however, finds no statistically significant changes in the growth rate of U.S. companies for the period of the fourth wave. Note that measurement errors and statistical problems may arise in post-merger operating performance studies (similar to those found for long-term wealtheffect studies). It may not be meaningful, therefore, to compare results across
Table 1.1 M&A announcements—short-term effects Benchmark return model
Event Window (days)
Sample size: T/B/C
Type of M&A
CARs target, %
CARs bidder, %
Panel A: Third takeover wave, 1950s–1973 Asquith, 1983 (U.S.) 1962–1976
BMCP
(⫺2, 0) (⫺20, 0)
211/196 211/196
M
⫹6.20a ⫹13.30a
⫹0.20 ⫹0.20
Eckbö, 1983 (U.S.)
1963–1978
MM
(⫺1, ⫹1) (⫺20, ⫹10)
57/102 57/102
HM
⫹6.24a ⫹14.08a
⫹0.07 ⫹1.58
Franks, Broyles, and Hecht, 1977 (U.K.)
1955–1972
MM, TTA
(0, ⫹20)
70
M
⫹16.0*
⫹4.60*
Eckbö and Langohr, 1989 (France)
1966–1982
MM
(0, ⫹5)
90/52
TO–Public
⫹16.48a
⫺0.29
EV/PA
(⫺2, ⫹1)
326 34 57 120 115
All MA RMA RMA UMA UMA
⫺0.70 ⫹1.54 ⫹2.88 ⫹0.23 ⫺4.09b ⫺1.23
Study (sample country)
Sample period
Panel B: Fourth takeover wave, 1981–1989 Morck, Shleifer, and 1975–1987 Vishny, 1990 (U.S.) 1975–1979 1980–1987 1975–1979 1980–1987 Byrd and Hickman, 1992 (U.S.)
1980–1987
MM
(⫺1, 0)
128
TO
Franks and Mayer, 1996 (U.K.)
1985–1986
MAM
(0, ⫹20)
34 32
FA HA
Doukas, Holmen, and Travlos, 2001 (Sweden)
1980–1995
MM
(⫺5, ⫹5)
46 46
RMA UMA
CARs combined, %
⫹8.60*
⫹18.44a ⫹29.76a ⫹2.74a ⫺2.37c (continued)
Table 1.1 (continued) Study (sample country)
Sample period
Benchmark return model
Event Window (days)
Kang, Shivdasani, and Yamada, 2000 (Japan)
1977–1993
MM
(⫺5, (⫺1, (⫺1, (⫺1, (⫺1,
MAM
Panel C: Fifth takeover wave, 1993–2001 Mulherin and Boone, 1990–1999 2000 (U.S.)
Sample size: T/B/C
Type of M&A
154 104 50 95 59
All MA RMA UMA Stock Mixed
(⫺1, ⫹1)
376/281/281
MA–Public
⫹5) 0) 0) 0) 0)
Sudarsanam and Mahate, 2003 (U.K.)
1983–1995
4 methods, results are for MAM
(⫺1, ⫹1) (⫹2, ⫹40)
519
All deals
Goergen and Renneboog, 2004 (Europe)
1993–2001
6 methods, results are for MM (TTA)
(⫺2, ⫹2)
40/41 53/55 28/32 88/86 30/33 18/23
M FA HA Cash Stock Mixed
CARs target, %
CARs bidder, %
CARs combined, %
⫹2.22a ⫹1.4b ⫹0.8 ⫹1.0b ⫹1.4c ⫹21.2a
⫺0.37 ⫺1.39a ⫹0.14
⫹12.62a ⫹11.33a ⫹17.95a ⫹13.56a ⫹11.38a ⫹13.24a
⫹4.35a ⫹1.94a ⫺3.43a ⫹0.90c ⫹2.57a ⫹0.22
⫹3.56a
van Schaik and Steenbeek, 2004 (Japan)
1993–2003
MM
(⫺1, ⫹1)
136
All deals
⫹0.57
Bae, Kang, and Kim, 2002 (Korea)
1981–1997
MM
(⫺5, ⫹5)
107 66 41
M all RM UM
⫹2.666b ⫹3.904a ⫹0.672
Martynova and Renneboog, 2006b (Europe)
1993–2001
6 methods, Results are for MM (TTA)
(⫺5, ⫹5)
259/1659 380/329 123/120 405/754 185/285 92/412 525/1334 234/774
M FA HA Cash Stock Mixed RMA UMA
⫹6.25a ⫹20.19a ⫹22.36a ⫹20.17a ⫹11.10a ⫹17.48a ⫹15.16a ⫹17.36a
⫹1.07a ⫺0.29 ⫺0.18 ⫹1.03a ⫹0.66 ⫹1.03c ⫹0.98a ⫹0.45
Source: Martynova and Renneboog (2005). Types of M&A: TO ⫽ tender offer; M ⫽ merger; MA ⫽ M&A ⫽ horizontal M&A; VMA ⫽ vertical M&A; RMA ⫽ related M&A (nonconglomerate); UMA ⫽ unrelated M&A (conglomerate or diversification); A ⫽ acquisition; FA ⫽ friendly acquisition; HA ⫽ hostile acquisition; stock ⫽ all-stock offer; cash ⫽ all-cash offer; mixed ⫽ combination of stock and cash offer; public (pub) ⫽ target company is public; private (priv) ⫽ target company is private. Benchmark return models: MM ⫽ market model; MAM ⫽ market-adjusted model; CAPM ⫽ capital asset pricing model; BMCP ⫽ beta-matched control portfolio (CRSP); FFM ⫽ Fama-French model; VPE ⫽ valuation prediction error; PSM ⫽ probability scaling method; TTA ⫽ thin trade-adjusted; EV/PA ⫽ ratio of change in bidder equity value to acquisition price; SBM ⫽ size and book-to-market ratio-matched portfolio. Close is the date the target was delisted from public trading. Sample size: T/B/C ⫽ number of observations for target firms/bidding firms/combined firms, respectively. If the three samples have the same number of observations, only one number is reported. *Significance is not reported, a/b/c ⫽ statistical significance at 1%/5%/10% levels, respectively.
12
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
countries and merger waves. Results may also be affected by accounting standard changes and country variations, and even by noise in the accounting data.
1.9
Conclusion and overview of the research presented in this volume
Each M&A wave has been characterized by a different set of underlying triggers. However, we consistently find that takeovers early in the wave are triggered by industry shocks. Takeovers are more likely to occur during periods of economic recovery, and the takeover market may be driven by regulatory changes as well as by industrial and technological shocks. Managers’ personal goals may further affect takeover activity. We find that managerial hubris and herding behavior tend to increase during takeover waves, often leading to inefficient acquisitions. Finally, takeover activity usually collapses alongside a market decline and an economic recession. Most M&As improve efficiency and generate substantial share price increases at the announcement, most of which will accrue to the target shareholders. It is possible that heterogeneity in takeover wave triggers may account for differences in M&A patterns and profitability across decades. Different types of shocks, whether economic or technological, have different impacts on corporate profitability and hence on any gains. Thus understanding whether takeovers will create or destroy value requires an understanding of why and when merger waves occur. The literature on M&As leaves a number of questions on takeovers unanswered. Issues such as cross-border mergers and acquisitions merit more comprehensive theoretical and empirical analysis. It is also important to determine how differences in corporate law, governance, and accounting quality influence cross-border acquisitions (see Volume 2). Noneconomic factors, such as manager compensation, education, and even social networks, may also play large roles in takeover decisions and thus need a great deal of further explication. This volume intends to close this gap partially. Section 1 of this volume describes the expected and long-term performance of forms involved in mergers and acquisitions (M&As) as well as how the market for corporate control has evolved over the past 15 years. Brakman, Garretsen and van Marrewijk collect a very large dataset on cross-border merger and acquisitions in order to analyze empirically the properties of cross-border M&As regarding country characteristics, regional composition, gross and net flows, size, and inequality. In Chapter 2, Brakman, Garretsen, and van Marrewijk discuss established and new theories on foreign direct investments and the extent to which these may be helpful for understanding the facts. The authors provide an overview of M&A deals for virtually all countries throughout the world over the period 1986–2005. About 50% of cross-border M&As appear to be horizontal activities
Understanding mergers and acquisitions: activity since 1990
13
deals. Most acquirers are public firms, and the largest proportion of targets are subsidiaries. In most cases, the M&A is completely paid for in cash and leads to complete ownership of the target firm. The largest acquirers and targets are still those based in the OECD countries, particularly in Western Europe and North America. When Brakman, Garretsen, and van Marrewijk divide the world into nine “global regions,” they find that about half of all inter-regional M&As occur between Western Europe and North America. The third-largest flows are those from Western to Eastern Europe. Using Gini coefficients, they document that the tendency of inequality to change over time and the change in inequality are strongly correlated with merger waves. In a nutshell, they assert that (1) most FDI is in the form of cross-border M&As; (2) firms engaged in cross-border M&As seem to be “market-seeking”; (3) cross-border M&As come in waves (the most recent wave is still unfolding); (4) economic integration (international deregulation) has stimulated M&As; and (5) both the size of and the inequality between M&As grows over time which is strongly correlated with the wave phenomenon. In Chapter 3, Liodakis and Pang search for the alpha. They investigate which acquisitions create value and show that, although the typical acquirer underperforms the market in the long term, not all acquisitions destroy value. The authors search for factors that could help investors screen for valueenhancing acquirers. For instance, cash-financed deals within the same industry that involve relatively large targets enjoy better fortunes, and bidders that trade at a valuation discount to their sector typically do better. The bid premium and the short-term market reaction to the deal announcement are also important leading indicators. Liodakis and Pang’s basket of “potential valuecreating” acquirers outperformed the market by 24% and the typical bidder by 37%. Although numerous research papers have been written on stock-price performance following mergers and acquisitions, the empirical evidence on changes in post-acquisition operating performance is relatively sparse and the conclusions vary widely. Martynova, Oosting, and Renneboog argue in Chapter 4 that the main reason for such widely differing views lies in the different, sometimes flawed, methodologies used to compare pre- and post-acquisition operating performance. Martynova, Oosting, and Renneboog adjust for industry, size, and preevent performance; and they utilize pure cash-flow performance, including changes in working capital. They investigate a sample of 155 European corporate mergers and acquisitions completed between 1997 and 2001 and study a period of 3 years pre-acquisition until 3 years post-acquisition. Their first main result is that, after adjusting for industry, size, and pre-acquisition, the combined operating performance does not change significantly following mergers, whereas the unadjusted “raw” operating performance declines significantly. In addition, the acquirer and the target significantly outperform their industry median peers prior to the merger. This shows that to estimate changes in performance reliably, it is important to adjust raw performance not only for industry-median performance,
14
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
but also for (size and) pre-event performance, thereby taking into account possible mean reversion of a company’s performance. Renneboog and Szilagyi state in Chapter 5 that, even though bondholders are among the most important corporate stakeholders, academic research on how M&As affect them has been fairly limited. Until recently, only three main hypotheses have been tested in the empirical literature. First, bondholders may benefit more from diversifying deals because the cash-flow streams of the merging firms are less well correlated. Second, firms may reverse any bondholder gains by paying for acquisitions with borrowed cash, thereby inducing leverage. And third, the actual changes in bondholder wealth should be influenced by how the pre-merger risk profiles of bidder and target compare. The authors show that cross-border deals also provide a platform for interactions between governance and legal systems. Thus, if they expose firms to jurisdictions with better creditor protection, they may even allow creditors to strengthen their legal positions. In Chapter 6, Goergen and Frechknall-Hughes investigate the acquisitions that offer target shareholders a choice of different types of means of payment (considerations), including the potential to mix and match these considerations. They also discuss the reasons that bidders may want to issue loan notes and why target shareholders may want to take up loan notes rather than cash or shares. In contrast to existing studies, the authors do not take the eventual payment for an acquisition as a given, but rather take into account the choice of different types of considerations offered to the target shareholders. Furthermore, in contrast with most other studies, which consider loan notes to be equivalent to cash, these authors clearly distinguish between the two. Goergen and FrechknallHughes draw three important conclusions. First, the popularity of loan notes increased from the late 1990s through the early 2000s. Second, contrary to what theories on asymmetric information argue and findings from empirical studies on wealth gains suggest, target shareholders do not always choose (exclusively) cash instead of shares. Third, clear benefits derive from loan notes issued both to the target shareholders and to the bidders. More precisely, using the tax models developed, the tax position of target shareholders may, in certain circumstances, make loan notes a more attractive choice than other types of consideration. Section 2 of this volume deals with special types of takeovers: acquisitions of recently floated companies, reverse mergers, mergers in the insurance industry, and investments by venture-capital investors. In Chapter 7, the first chapter of this section, Audretsch and Lehmann observe that many new public firms are acquired within a short period after their IPO, a phenomenon often described as a double-exit strategy (old shareholders have two opportunities to exit: at the IPO and at acquisition). They ask a particular question: Why do firms forgo a private takeover but sell them soon after the IPO, even though the cost of going public often accounts for more then 10% of the funds raised? To answer this question, the authors analyze the determinants of being acquired within 3 years after IPO using a unique and hand-collected dataset of 285 IPOs in
Understanding mergers and acquisitions: activity since 1990
15
Germany, mostly in the high-tech industry. What makes a recently floated firm an interesting acquisition target? About 12% of the firms are acquired within 3 years of their IPO. They find that the identity of controlling initial owners, and hence the corporate governance structure of firms, affects the likelihood of acquisitions after IPO. One factor that conditions the likelihood of newly public firms is the percentage of equity held by the different initial owners at time of IPO. Although venture capitalists may desire liquidity and returns more than control of the firm, Audretsch and Lehmann do not find a significant impact of the percentage of equity held by venture capitalists on the likelihood of being an acquisition target within three years after IPO. However, they demonstrate that banks as initial owners of IPO firms significantly increase the likelihood of being taken over after IPO. Roosenboom and Schramade show in Chapter 8 that initial public offerings (IPOs) are not the only way of going public: Reverse mergers may offer a costefficient way of floating a company. In reverse mergers a private company de facto acquires a public target company whereas de iure it is the public company that acquires the private company. This way the private company, the acquirer in economic terms, can obtain a stock market listing for its shares via the “backdoor” without the costs and time associated with conducting an IPO. Target firms in reverse mergers are typically unprofitable shell companies. The authors study reverse takeovers in the U.K. They first calculate the abnormal returns to reverse-merger announcements and subsequently focus on the determinants. Finally, they measure the long-run stock price and operating performance of reverse-merger firms and compare them with IPO firms. They find that target returns in reverse mergers are significantly positive and are higher when target firms are in a bad financial condition. Roosenboom and Schramade conclude that this is consistent with the existence of large, previously unused tax shields. They also document that when bidders are relatively large, their chances of successful takeover completion are higher. They show that reverse-merger firms and matched IPO firms display similar stock and operating performance, which is inconsistent with the conventional wisdom that reverse mergers are mainly conducted by poorquality private firms that want to escape regulatory scrutiny. The final chapter of this section deals with the entry and exit of venture capitalists in Canadian firms over the period 1991–2004. Cummings and Johan study the complete series of exit possibilities: initial public offerings (IPOs, or new listings on a stock exchange for sale to the general public), acquisitions (in which the investor and entrepreneur sell to a larger company), secondary sales (in which the investor sells to another company or another investor, but the entrepreneur does not sell), buybacks (in which the entrepreneur repurchases the interest of the investor), and write-offs (liquidations). Their findings show that the patterns of exit vary, depending on the exit year, the characteristics of the venture-capital investor (private limited partnership, corporate, and government), the characteristics of the investee firm (industry and stage of development at first investment), and the characteristics of the transaction (capital requirements, syndication, and security design).
16
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Section 3 of this volume focuses on irrationalities in the decision process of takeovers. In Chapter 10 Sudarsanam and Huang argue that one way to reduce risk-related agency conflicts is to structure the managers’ compensation in such a way that it provides the proper risk incentives to managers (e.g., through executive stock options). The authors maintain that both the wrong monetary risk incentives and overconfidence can lead to excessive risk-taking. Avoiding such excessive risk-taking or, conversely, inadequate risk-taking by risk-averse managers requires a new monitoring role for corporate governance. Sudarsanam and Huang present an integrated model in which risk-taking and consequent firm performance are subject to the interacting influences of executive compensation structure, a behavioral bias, and corporate governance. In Chapter 11, Pastor and Poveda believe that in stock-financed mergers the managers of acquiring companies have incentives to overstate the earnings because higher earnings may push up the stock prices, thereby engendering a more favorable exchange ratio in the stock-for-stock takeovers. The authors disclose that, as in the United States, acquiring Spanish firms show an increase in unexpected accruals prior to the merger announcement. They also find a reversion after the merger. So it seems that, despite the risk of detection in due diligence, the results point out that managers of acquiring firms make use of the discretion allowed in the accounting rules to overstate the accounting results in the year of the merger announcement. In Chapter 12, Kräussl and Topper argue that size is the main factor explaining value in the Dutch M&A market. Their empirical results indicate that small companies earn significantly higher returns (of 2.65%) than do large companies upon the announcement of a transaction. In the final chapter of this book, Ali states that corporations routinely use buybacks to return excess capital to their shareholders, manage their capital structures and convey signals to the market about the corporation’s financial performance. This chapter examines how buybacks can be used by Australian corporations to achieve those aims as well as undisclosed objectives such as consolidating management’s control of the corporation and creating deterrence to takeover bids.
References Andrade, G., and Stafford, E. (2004). Investigating the Economic Role of Mergers. Journal of Corporate Finance, 10:1–36. Asquith, P. (1983). Merger Bids, Uncertainty, and Stockholder Returns. Journal of Financial Economics, 11:51–83. Auster, E., and Sirower, M. (2002). The Dynamics of Merger and Acquisition Waves: A Three-M Stage Conceptual Framework with Implications for Practice. Journal of Applied Behavioral Science, 38:216–244. Bae, K., Kang, J., and Kim, J. (2002). Tunneling or Value Added? Evidence from Mergers by Korean Business Groups. Journal of Finance, 57(6): 2695–2740.
Understanding mergers and acquisitions: activity since 1990
17
Bhagat, S., Dong, M., Hirshleifer, D., and Noah, R. (2004). Do Tender Offers Create Value? New Methods and Evidence. Working Paper No. 2004-4, Dice Center. Bhagat, S., Shleifer, A., and Vishny, R. (1990). Hostile Takeovers in the 1980s: The Return to Corporate Specialization. Brookings Papers on Economic Activity, Special Issue, pp. 1–72. Bhide, A. (1990). Reversing Corporate Diversification. Journal of Applied Corporate Finance, 3:70–81. Blair, M. (1993). The Deal Decade: What Takeovers and Leveraged Buyouts Mean for Corporate Governance. Brookings Institution, Washington, D.C. Boot, A., Milbourn, T., and Thakor, A. (1999). Megamergers and Expanded Scope: Theories of Bank Size and Activity Diversity. Journal of Banking and Finance, 23:195–214. Bradley, M., Desai, A., and Kim, E. H. (1988). Synergistic Gains from Corporate Acquisitions and Their Division between the Stockholders of Target and Acquiring Firms. Journal of Financial Economics, 21(1): 3–40. Byrd, J., and Hickman, K. (1992). Do Outside Directors Monitor Managers? Evidence from Tender Offer Bids. Journal of Financial Economics, 32:195–214. Carline, N., Linn, S., and Yadav, P. (2002). The Impact of Firm-Specific and Deal-Specific Factors on the Real Gains in Corporate Mergers and Acquisitions: An Empirical Analysis. Working Paper, University of Oklahoma. Coase, R. (1937). The Nature of the Firm. Economica, 4:386–405. Cosh, A. D., Hughes, A., and Singh, A. (1980). The Causes and Effects of Takeovers in the UK: An Empirical Investigation for the Late 1960’s. In: The Determinants and Effects of Mergers (Miller. D. C., ed.). Cambridge: Oelgeschlager, Gunn & Hain. Devenow, A., and Welch, I. (1996). Rational Herding in Financial Economics. European Economic Review, 40:603–615. Doukas, J., Holmen, M., and Travlos, N. (2001). Corporate Diversification and Firm Performance: Evidence from Swedish Acquisitions. SSRN Working Paper. Eckbo, B. E. (1983). Horizontal Mergers, Collusion, and Stockholder Wealth. Journal of Financial Economics, 11(1–4):241–274. Eckbo, B. E., and Langohr, H. (1989). Information Disclosure, Method of Payment, and Takeover Premiums: Public and Private Tender Offers in France. Journal of Financial Economics, 24:363–403. Eis, C. (1969). The 1919–1930 Merger Movement in American Industry. Journal of Law and Economics, 12:267–296. Fairburn, J. A. (1989). The Evolution of Merger Policy in Britain. In: Mergers and Merger Policy (Fairburn, J. A., and Kay, J., eds.). Oxford: Oxford University Press, pp. 193–230. Fama, E., Fisher, L., Jensen, M., and Roll, R. (1969). The Adjustment of Stock Prices to New Information. International Economic Review, pp. 1–21.
18
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Franks, J. R., Broyles, J. E., and Hecht, M. J. (1977). An Industry Study of the Profitability of Mergers in the UK. Journal of Finance, 32:1513–1525. Franks, J., and Mayer, C. (1996). Hostile Takeovers and the Correction of Managerial Failure. Journal of Financial Economics, 40:163–181. Ghosh, A. (2001). Does Operating Performance Really Improve Following Corporate Acquisitions? Journal of Corporate Finance, 7(2):151–178. Goergen, M., and Renneboog, L. (2004). Shareholder Wealth Effects of European Domestic and Cross-Border Takeover Bids. European Financial Management, 10(1):9–45. Goergen, M., Martynova, M., and Renneboog, L. (2005). Corporate Governance Convergence: Evidence from Takeover Regulation Reforms in Europe, Oxford Review of Economic Policy, 21(2):1–27. Golbe, D. L., and White, L. J. (1993). Catch a Wave: The Time Series Behaviour of Mergers. Review of Economics and Statistics, 75:493–497. Gort, M. (1969). An Economic Disturbance Theory of Mergers, Quarterly Journal of Economics, 83:624–642. Gorton, G., Kahl, M., and Rosen, R. (2000). Eat or Be Eaten: A Theory of Mergers and Merger Waves. Unpublished Working Paper, University of Pennsylvania, Philadelphia. Graham, J. R. (1999). Herding among Investment Newsletters: Theory and Evidence. Journal of Finance, 54:237–268. Harford, J. (1999). Corporate Cash Reserves and Acquisitions. Journal of Finance, 54(6):1969–1997. Harford, J. (2003). Efficient and Distortional Components to Industry Merger Waves. Unpublished Working Paper, AFA 2004, San Diego Meetings. Holmström, B., and Kaplan, S. N. (2001). Corporate Governance and Merger Activity in the United States: Making Sense of the 1980s and 1990s. Journal of Economic Perspectives, 15:121–144. Jarrell, G. A., Brickley, J. A., and Netter, J. M. (1988). The Market for Corporate Control: The Empirical Evidence since 1980. Journal of Economic Perspectives, 2:49–68. Jensen, M. C. (1986). Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review, 76(2):323–329. Jensen, M. C. (1993). The Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems. Journal of Finance, 48:831–880. Jensen, M. C. (2004). Agency Costs of Overvalued Equity. Harvard NOM Working Paper No. 04-26, ECGI Finance Working Paper No. 39/2004. Jensen, M. C., and Ruback, R. S. (1983). The Market for Corporate Control: The Scientific Evidence. Journal of Financial Economics, 11:5–50. Jovanovic, B., and Rousseau, P. (2001). Mergers and Technological Change: 1885–2001. Unpublished Working Paper, New York University. Jovanovic, B., and Rousseau, P. (2002). The Q-Theory of Mergers. American Economic Review, 92(2):198–204. Kang, J., Shivdasani, A., and Yamada, T. (2000). The Effect of Bank Relations on Investment Decisions: An Investigation of Japanese Takeover Bids. Journal of Finance, 55:2197–2218.
Understanding mergers and acquisitions: activity since 1990
19
Lamoreaux, N. R. (1985). The Great Merger Movement in American Business, 1895–1904. Cambridge: Cambridge University Press. Linn, S., and Switzer, J. (2001). Are Cash Acquisitions Associated with Better Post Combination Operating Performance Than Stock Acquisitions? Journal of Banking and Finance, 25:1113–1138. Lintner, J. (1971). Expectations, Mergers and Equilibrium in Purely Competitive Securities Markets. American Economic Review, 61(2): 101–111. Markham, J. W. (1955). Survey of the Evidence and Findings on Mergers. In: Business Concentration and Price Policy. Princeton: Princeton University Press. Markham, J. W. (1973). Conglomerate Enterprise and Public Policy. Boston: Harvard University Press. Martynova, M., and Renneboog, L. (2006a). Mergers and Acquisitions in Europe. In: Advances in Corporate Finance and Asset Pricing (Renneboog, L., ed.). Amsterdam: Elsevier, pp. 13–75. Martynova, M., and Renneboog, L. (2005), Takeover Waves: Triggers, Performance and Motives (with M. Martynova). Discussion Paper CentER, Tilburg University and European Corporate Governance Institute. Martynova, M., and Renneboog, L. (2006b). The Performance of the European Market for Corporate Control: Evidence from the 5th Takeover Wave, Discussion Paper CentER, Tilburg University and European Corporate Governance Institute. Matsusaka, J. (1996). Did Tough Antitrust Enforcement Cause the Diversification of American Corporations? Journal of Financial and Quantitative Analysis, 31:283–294. Mitchell, M., and Mulherin, J. H. (1996). The Impact of Industry Shocks on Takeover and Restructuring Activity. Journal of Financial Economics, 41:193–229. Moeller, S. B., Schlingemann, F. P., and Stulz, R. M. (2005). Wealth Destruction on a Massive Scale? A Study of Acquiring-Firm Returns in the Recent Merger Wave. Journal of Finance, 60(2):757–782. Morck, R. M., Shleifer, A., and Vishny, R. W. (1988). Characteristics of Targets of Hostile and Friendly Takeovers. In: Corporate Takeovers: Causes and Consequences (Auerbach, A. J., ed.). Chicago: National Bureau of Economic Research. Morck, R., Shleifer, A., and Vishny, R. (1990). Do Managerial Objectives Drive Bad Acquisitions? Journal of Finance, 45(1):31–48. Mueller, D. C. (1980). The United States, 1962–1972. In: The Determinants and Effects of Mergers: An International Comparison (Mueller, D. C., ed.). Cambridge: Oelgeschlager, Gunn & Hain, pp. 271–298. Mulherin, J. H., and Boone, A. L. (2000). Comparing Acquisitions and Divestitures. Journal of Corporate Finance, 6:117–139. Myers, S., and Majluf, N. (1984). Corporate Financing and Investment Decisions When Firms Have Information that Investors Do Not Have. Journal of Financial Economics, 13:187–221.
20
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
O’Brien, A. P. (1988). Factory Size, Economies of Scale, and the Great Merger Wave of 1898–1902. Journal of Economic History, 48:639–649. Persons, J. C., and Warther, V. A. (1997). Boom and Bust Patterns in the Adoption of Financial Innovations. Review of Financial Studies, 10(4):939–967. Ravenscraft, D. J., and Scherer, F. M. (1987). Mergers, Sell-offs and Economic Efficiency. The Brookings Institution, Washington, D.C. Renneboog, L., and Simons, T. (2006). Public to Private Transactions: Motives, Trends, Theories and Empirical Literature on LBOs, MBOs, MBIs and IBOs, Discussion Paper CentER, Tilburg University and European Corporate Governance Institute. Renneboog, L., Simons, T., and Wright, M. (2007). Why Do Public Firms Go Private in the UK? Journal of Corporate Finance, (forthcoming). Renneboog, L., and Szilagyi, P. (2007). Corporate Restructuring and Bondholder Wealth. European Financial Management, 3 (forthcoming.) Renneboog, L., and Szilagyi, P. (2006). How Do Mergers and Acquisitions Affect Bondholders in Europe? Evidence on the Impact and Spillover of Governance and Legal Standards, Discussion Paper CentER, Tilburg University and European Corporate Governance Institute. Reid, S. R. (1968). Mergers, Managers, and the Economy. New York: McGraw-Hill. Roll, R. (1986). The Hubris Hypothesis of Corporate Takeovers. Journal of Business, 59:197–216. Scharfstein, D., and Stein, J. (1990). Herd Behavior and Investment. American Economic Review, 80(3):465–479. Shleifer, A., and Vishny, R. W. (1991). Takeovers in the ’60s and the ’80s: Evidence and Implications. Strategic Management Journal, 12:51–59. Shleifer, A., and Vishny, R. W. (2003). Stock Market Driven Acquisitions. Journal of Financial Economics, 70:295–311. Stigler, G. (1950). Monopoly and Oligopoly Power by Merger. American Economic Review, 40:23–34. Sudarsanam, S. (2003). Creating Value from Mergers and Acquisitions: The Challenges. Harlow. UK: Prentice Hall/Financial Times. Sudarsanam, S., and Mahate, A. A. (2003). Glamour Acquirers, Method of Payment and Post Acquisition Performance: The UK Evidence. Journal of Business Finance and Accounting, 30:299–341. Thorp, L. W. (1941). The Increasing Responsibility of Management. Journal of Accountancy, 72(50):403. van Schaik, D., and Steenbeek, O. W. (2004). Price and Volume Effects of Merger Bids in Japan. SSRN Working Paper. Wright, M., Renneboog, L., Scholes, L., and Simons, T. (2006). Leveraged Buyouts in the UK and Europe: Retrospect and Prospect. Journal of Applied Corporate Finance, 18(3):38–55.
Part One International M&A Activity and Takeover Performance
This page intentionally left blank
2 Cross-border mergers and acquisitions: the facts as a guide for international economics1 Steven Brakman, Harry Garretsen and Charles van Marrewijk
Abstract Using a detailed and large -border dataset on cross merger and acquisitions, we discuss the relationships between theory and the following observed empirical characteristics: (1) Most FDI is in the form of M&As, (2) firms engaged in M&As seem to be “market-seeking,” (3) M&As come in waves (the most recent wave is still unfolding), (4) economic integration (international deregulation) stimulates M&As, (5) the size of and inequality between M&As grows over time. Our contention is that these stylized facts drive and should drive recent theoretical contributions in the field of international economics that try to understand cross-border mergers and acquisitions. Although some models, notably Neary’s (2003) explain a number of characteristics, a full-fledged model of cross-border M&As that can, at least in principle, deal with all the characteristics is still lacking.
2.1
Introduction
Theoretical developments in international economics are sometimes motivated by empirical findings. The “new trade theory,” for example, was inspired to a large extent by empirical work on intra-industry trade (Neary, 2004b). This also holds for the recent outburst of research on foreign direct investment (FDI) as one of the driving forces behind the current wave of globalization. Many observers have noted that FDI grows much faster than world merchandise trade (BarbaNavaretti and Venables, 2004). This is clearly a stylized fact in search of an explanation. For years, students of FDI used Dunning’s (1993) OLI-categorization scheme to understand why firms engage in FDI. Notwithstanding its usefulness in the case of FDI, a categorization scheme is not a model. New theories are being developed in which the firm’s decision on FDI engagement is determined in a full-fledged microeconomic model.
1
We thank Utz Weitzel for his help with the Thomson dataset.
24
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Interestingly, looking at FDI as a broad category obscures the fact that most FDI is in the form of so-called cross-border mergers and acquisitions (henceforth M&As). Figure 2.1 shows a decomposition of FDI from which it is clear that M&As constitute the bulk of FDI, whereas Greenfield FDI is less important than M&As. The main difference between these two investments is that in an M&A “control of assets and operations is transferred from a local to a foreign company, the former becoming an affiliate of the latter” (UNCTAD, 2000, p. 99). Only recently have models in international economics been developed that enable us to understand M&As (Neary, 2004b). Neary’s model takes the standard explanations for M&As a step further. Usually two motives are mentioned to explain M&As: a strategic motive (competition reduction) and an efficiency motive (cost reductions). An explanation of cross-border M&As, however, also must explain the cross-border part of the deals. Trade theory suggests that comparative advantage could be included in full explanations of M&As; see Neary (2004a). A different but equally novel line of research in international economics (see Barba-Navaretti and Venables, 2004, or Helpman, 2006, for excellent surveys) seeks to understand the conditions under which firms decide to locate (part of) their production abroad (the off-shoring decision). When they decide to off-shore, some firms do so under the flag of FDI, while other firms go for outsourcing. In this literature, and in contrast to the empirical relevance illustrated in Figure 2.1, the role of cross-border M&As is, however, largely ignored. The aim of this chapter is to present stylized facts on cross-border M&As. This is interesting in its own right (see also Evenett, 2004), but it may also act as
Foreign Direct Investment 22%
78%
Greenfield investments
Mergers and Acquisitions (M&As) 97%
3%
Mergers
Acquisitions 16%
65% 15% Full acquisition
More than 50% acquisition
10–49% acquisition
Figure 2.1 Distribution of different types of FDI. Source: Brakman, Garretsen, and van Marrewijk (2006); data UNCTAD (2000); 78–22% in value terms, other percentages in number of deals.
Cross-border mergers and acquisitions: the facts as a guide for international economics
25
a guide for the recent upsurge of interest in FDI and its alternatives in international economics regarding the facts that the modern theory of FDI should be able to explain. When highlighting the stylized facts in this chapter, we therefore briefly point out those FDI models in international economics that are capable of coping with the facts under consideration. We proceed as follows. Section 2.2 presents basic characteristics of M&As using the database of Thomson Financial Securities Data (hereinafter, Thomson). The advantage of this source over UNCTAD data is that it consists of individual data on each M&A, enabling us to look at M&As at a very detailed level. Section 2.3 provides information at the country level. Section 2.4 looks at the regional composition of target and acquirer, both of which are typically found in the OECD countries. Section 2.5 confronts gross M&As with net M&As and discusses some developments over time, confirming that emerging markets, such as China and Eastern Europe, are increasingly becoming net targets. Section 2.6 argues that the inequality within the set of M&As tends to increase over time. Section 2.7 discusses the characteristics of firms involved in FDI. Section 8 concludes and summarizes our findings.
2.2
Cross-border M&As: basic characteristics
Our overview of the structure and developments of cross-border M&As is based on Thomson’s Global Mergers and Acquisitions database, which provides the best and most extensive data source for M&As to date. Thomson gathers information on M&As exceeding 1 million U.S. dollars. Its main sources of information are financial newspapers and specialized agencies like Bloomberg and Reuters. Our Thomson dataset begins in 1979 and ends in August 2006. Initially, Thomson focused on American M&As. Systematic M&A data for almost all countries are available for about the last 20 years. In presenting the data we therefore focus on the period 1986–2005 usually grouped in four 5-year subperiods, thereby mitigating the large annual fluctuations characteristic of M&As and allowing us to discern longer term trends. We collected information on all completed/unconditional cross-border M&As with a deal value of at least $10 million. In the period 1986–2005 this provided us with 27,541 cross-border M&As; see the overview in Table 2.1. Usually, the time difference between the date of announcement of an M&A deal and the date the deal is effective is nonexistent or very short (such that 99.7% of the deals are immediately effective within the same year). The announced date is exactly the same as the effective date for about 38% of the M&A deals. On average, the difference between these two dates is 0.18 year. We therefore used the date of announcement for classifying the M&A deals over time; see also Brakman, Garretsen, and van Marrewijk (2005, 2006). In general, a large share of a company (on average, 75.5%) is acquired by the deal, leading to a majority ownership after the deal is completed (on average, owning 80.1% of the acquired company). This indicates that most firms already have intimate knowledge of the firm being
26
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 2.1 Overview of cross-border M&As Number of deals Cross-border M&As, 1986–2005 Effective M&As Average percentage of shares acquired Average percentage of shares owned after deal Number of tender offers Number of horizontal M&As (2-digit level)
27,541 27,461
Percentage
2,476 13,605
99.7 75.5 80.1 9.0 49.4
Public status of target government joint venture subsidiary public private unknown /other
658 977 11,053 7,343 7,489 21
2.4 3.5 40.1 26.7 27.2 0.1
Public status of acquirer government joint venture subsidiary public private unknown/other
298 499 6,814 15,796 4,067 67
1.1 1.8 24.7 57.4 14.8 0.2
Number of deals involving cash if so, average share of payment
25,665
93.2 94.4
Number of deals involving stock if so, average share of payment
2,635
9.6 73.1
acquired. Payment for the acquisition usually involves cash (93.2% of the deals); and, if so, it is usually completely paid for in cash (on average, 94.4% of the deals involving cash are paid for in cash). Payment of the deal using shares occurs regularly (9.6% of the deals); and, if so, it is usually completely paid for in shares (on average, 73.1% of the deals involving stocks are paid for in stocks). The fact that many takeovers are financed with cash does not imply that shares are not important in those deals: Raising cash is greatly facilitated if stock prices of the firms involved are high. This might be the motive behind announcing takeovers before the actual takeover takes place; announcements tend to affect share prices in an upward direction (see also Box 2.1).2 There are substantial differences between the public status of acquiring and target firms. The majority of acquiring firms are public companies (57.4%), followed by subsidiaries (24.7%), and private firms (14.8%), respectively. The target company, on the other hand, is usually a subsidiary (40.1%), followed by 2
As to the (negative) relationship between profits and share prices with respect to M&As, see Fridolfsson and Stennek (2005).
Cross-border mergers and acquisitions: the facts as a guide for international economics
27
Box 2.1 Cross-border M&A Profitability For this chapter it is instructive to present a simple way of looking at a crossborder M&A. It is more a way of organizing thoughts than a complete model, but it illustrates the key issues involved. Let 1 and 0 indicate the post- and pre-merger situation, respectively. Then the gain of taking over a home firm, GH, by a foreign firm is given by the following expression: GH ⫽ ⎡⎢ π1*(n ⫺ 1, n* ⋅ ) ⫺ π*0 (n, n* ⋅ )⎤⎥ ⫺ π0 (n, n* ⋅ ) > 0 ⎣ ⎦
(2.1)
The first term (in square brackets) relates to the gain in profitability from reduced competition by taking over the domestic firm; the number of domestic firms is reduced by 1, from n to (n ⫺ 1). The number of foreign firms, n*, does not change. The second term indicates the cost of acquiring the domestic firm. This is a function of profits of the target—the more profitable a target is, the higher the takeover costs—and the cost of financing the takeover. If the acquirer has a windfall gain, for example, higher share prices arising from the takeover, the finance costs are smaller. The . indicates that other variables are taken as given. The balance between the change in profits and the costs involved in the M&A determines whether or not a takeover will take place. Whether the increase in profits really materializes after the M&A has taken place is another issue, but the equation illustrates how in international economics (the equation is taken from Neary, 2004a) the firm decision on whether or not to engage in a cross-border M&A is very simple. The firm (and its organizational setup) itself is something of a black box, and the focus is on how changes in the external environment (fall in transportation costs, lowering of tariffs) might have an impact on equation (2.1), and thus on the M&A decision.
a private company (27.2%) and a public company (26.7%), respectively. The share of subsidiaries and private companies among the target companies is therefore substantially larger and the share of public companies is substantially lower. To classify M&As between horizontal and other types of deals (be they vertical or conglomerate), we used the SIC classification of target and acquirer as provided by Thomson at the 2-digit level; a deal is therefore a horizontal M&A in our classification if the acquirer and target are active in the same 2-digit sector. On average, about half of the M&As are horizontal deals (49.4%, see the following for further details). Thus, to a large extent, investments take place in the same sector. One can speculate why this might be the case. Strategic motives may, of course, be at work here; but as we will argue, the most likely explanation is probably that most cross-border M&As belong to the category of market-seeking FDI. Taking a competitor out of the market reduces competition and increases
28
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
profits. Buying a firm outside one’s own sector might be motivated by an efficiency motive: It can be profitable to control a larger part of the value chain. Both motives increase profits after the takeover. We also argue that, since most crossborder M&As belong to the category of horizontal FDI, market-seeking motives play a dominant role in M&As. Figure 2.2 illustrates that the share of horizontal M&As is very stable over time when measured using the number of deals, fluctuating relatively little around the average of 49%, which ranges from a low of 45.1% in 1986 to a high of 51.5% in 1996. Horizontal M&As are substantially more volatile when measured using the value of the deals, fluctuating around the average of 56%, ranging from a low of 46.7% in 1988 to a high of 73.0% in 1999. Using either measure, we find little support for the argument that the share of horizontal M&As is declining. Those who would argue that the value of horizontal M&As has declined since 1999 are obviously obscuring the fact that this peak in 1999 is not representative over a long-time horizon. The current (2005) value of horizontal M&As of 55.2% is very close to the long-run average of 56%. From an international economics perspective, see our introduction; the question is if existing theories of FDI can explain the dominance of horizontal FDI. At first sight, this is not the case. Assuming that during our sample period 1985–2005 trade costs, broadly defined, have (if anything) decreased, the standard FDI model then predicts that horizontal FDI should become less important. With falling trade costs, foreign markets might ceteris paribus be better served by exporting
0.8
Horizontal M&As (share)
Value 0.6
Number of deals 0.4
0.2
0.0 1985
1990
1995 Year
2000
2005
Figure 2.2 Horizontal (2-digit) cross-border M&As; share of total, number of deals and value. (Horizontal lines indicate averages for the period 1986–2005.)
Cross-border mergers and acquisitions: the facts as a guide for international economics
29
3,000
1,500
2,500
1,250
2,000
1,000
1,500
750 Number of deals 500
1,000 Value 500 0 1985
250
1990
1995
2000
Year
Figure 2.3 Cross-border M&As, 1985–2005; number of deals and value.
0 2005
Value in 2005 ($ billion)
No. of deals
instead of FDI; and in the well-known proximity-concentration tradeoff, a drop in trade costs shifts the tradeoff in favor of exporting. However, Neary (2005) shows that falling trade costs might still explain the rise of horizontal FDI and thus of the bulk of cross-border M&As, once we allow for an FDI model that explicitly incorporates the possibility of cross-border M&As instead of merely looking at FDI as a black box (see Neary, 2004a). A historical perspective reveals a remarkable characteristic of M&As. Figure 2.2 depicts the evolution of all cross-border M&As over time, both measured as the number of deals and the value of deals (in constant 2005 $ bn, using the U.S. GDP deflator). Clearly, there is substantial variation over time, with periods of rapid increase followed by periods of rapid decline. Five merger waves have been identified during the 20th century, three of which are recent (Andrade, Mitchell, and Stafford, 2001). The 3rd wave took place in the late 1960s to the early 1970s. The 4th wave ran from about the mid-1980s until 1990. The 5th wave started around 1995 and ended in 2000 with the collapse of the “new economy.” Figure 2.3 shows that a subsequent 6th (still ongoing) merger wave started in the 21st century around 2003. Note that the data used in this chapter cover the last two waves. Merger waves are positively correlated with increases in share prices and p/e ratios, as well as with the general overall business cycle. However, the causality of the relation is not always clear. On the one hand, an upswing of the business cycle increases share prices, and high share prices reduce the cost of financing an M&A. On the other hand, the same upswing of the business cycle increases the profits of the target and increases takeover costs (see also Box 2.1). When one sticks to standard M&A motives, such as the efficiency argument, it is rather
30
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
difficult to explain the synchronicity of M&As. Gugler, Mueller, Yurtoglu, and Zulehner (2004) argue that merger waves can be understood if one acknowledges that M&As do not boost efficiency and hence do not increase shareholders’ wealth. Instead, they find that M&A waves are best looked upon as the result of overvalued shares and managerial discretion. For the case of the United States and restricting their sample to publicly traded firms, Andrade, Mitchell, and Stafford (2001) show that with each merger wave, the value of the M&A deals (measured by firms’ market capitalization) increases strongly. Merger waves in Europe seem to follow those in the United States with a short lag. During the 5th merger wave, European firms engaged in a number of mega-M&As, with the cross-border takeover of Mannesmann (Germany) by Vodafone (U.K.) for $203 bn in 1999/2000 the largest M&A to date. It turns out that this part of M&A waves is especially difficult to model. First, an M&A wave must start at some point. Equation (2.1) points to a difficulty in this respect. A reduction of competition makes an M&A profitable; this implies that it is rational to wait for other M&As to go first because waiting reduces competition, making the next M&A more profitable than the first. Second, an M&A wave must stop at some point. Both elements should be incorporated in a full M&A model. Neary (2004a) does just that: Waves have to start at some point or else M&A profits are forgone. Moreover, since it is a general equilibrium model, the excess supply on the labor market following an M&A (lower wages resulting in higher profits) finally stops the wave.
2.3
Countries and M&As in 2005
In this section we provide an overview of the currently (2005) most active countries in M&As. There were 2,154 cross-border M&As in 2005, with a total value of about $774 bn. Table 2.2 provides an overview of the top 20 countries ranked in order of acquirer value. Unsurprisingly, the United States tops the list, both in value and number of deals, acquiring 514 foreign firms with a total value of about $158 bn (20.4% of the total). The United States was also the largest target country in 2005 when measured in number of deals (356) and the second largest target in value terms ($125 bn). The United Kingdom was the second largest acquiring country (286 deals and $94 bn) and the largest target country in value ($144 bn; second largest in number of deals). Among the other countries listed in Table 2.2 are the “usual suspects” of high-income (European) countries: Spain, France, Germany, Italy, Switzerland, The Netherlands, Sweden, Denmark, Norway, Israel, Australia, Canada, Japan, Russia, and Hong Kong. More remarkable, presumably, are the high ranks for Egypt, United Arab Emirates (UAE), and even tiny Luxembourg. As suggested by the fact that the United States and the United Kingdom take the two top spots in Table 2.2, both as acquirer and target, there is substantial coincidence between acquirers and targets (large acquiring countries are usually also large target countries, and vice versa). Indeed, of the 20 countries listed as
Cross-border mergers and acquisitions: the facts as a guide for international economics
31
Table 2.2 Cross-border M&As; top 20 countries in 2005 (ranked according to acquirer value) Country
1 2 3 4 5
United States United Kingdom Spain France Germany
In value terms ($ million)
In number of deals
Acquirer
% of total
Target
Acquirer
Target
157,924 94,104 59,953 58,606 48,081
20.4 12.2 7.7 7.6 6.2
124,764 143,754 22,531 36,733 65,053
514 286 49 86 79
356 262 58 110 136
6 7 8 9 10
Italy Australia Switzerland Netherlands Sweden
37,897 31,722 30,973 28,664 19,555
4.9 4.1 4.0 3.7 2.5
48,593 10,048 6,710 32,416 17,799
48 137 35 64 63
58 106 22 38 44
11 12 13 14 15
Egypt Canada Luxembourg United Arab Emirates Japan
16,992 15,679 14,584 14,565 12,034
2.2 2.0 1.9 1.9 1.6
2,227 26,943 7,808 86 3,538
6 121 21 11 70
9 87 7 1 26
16 17 18 19 20
Russia Denmark Hong Kong Israel Norway
11,088 9,341 9,213 8,847 8,799
1.4 1.2 1.2 1.1 1.1
7,818 20,933 10,107 2,001 7,329
22 27 60 17 20
28 33 63 18 33
the largest acquirers in value terms in Table 2.2, 15 also appear among the top 20 as largest targets in value terms. Only Switzerland, Egypt, UAE, Japan, and Israel would have to be replaced by Belgium, China, Turkey, Czech Republic, and South Korea. This coincidence is illustrated in Figure 2.4 using logarithmic scales. The figure also indicates that the Czech Republic is indeed a relatively large target and the UAE is a relatively large acquirer. What can we conclude from the fact that M&As mostly take place between high-income countries? As stated before, an important classification in the literature is the difference between so-called horizontal and vertical FDI. The difference is important because in the case of horizontal FDI, firms are marketseeking (looking for large and profitable markets), whereas in the case of vertical FDI, firms have a factor-market motive. In the former case, firms are interested in the high wages of consumers instead of the low cost in factor markets (for example, low wages) in the latter case. Thus, both forms need very different models. As horizontal FDI seems to dominate the data, models that stress marketseeking reasons to engage in M&As are potentially the most appropriate for empirical research. Having acknowledged this (see also the previous section), we can see that these models have trouble explaining FDI in the face of
32
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
1,000,000 UK 100,000
USA
Target value
Czech Rep 10,000
1,000
100
UAE
10 10
100
1,000 10,000 Acquirer value
100,000
1,000,000
Figure 2.4 Cross-border; 2005, values ($ million, log scales). (The thin line is the 45° line.)
increased economic integration (falling trade costs); see also Evenett (2004, p. 427). It is here that the models in international economics might gain (Neary, 2005) from distinguishing more clearly between various forms of FDI, notably by including cross-border M&As as a separate category of FDI.
2.4
Regional distribution of cross-border M&As
In Section 2.3 we showed that the majority of cross-border M&As are between relatively rich countries. However, in the public debate on off-shoring—which includes all forms of FDI and thus also cross-border M&As as well as outsourcing—there is a strong undercurrent that looks at off-shoring, and hence at FDI and its main component cross-border M&As, as threatening. Workers in the industrialized countries would lose out because of the relocation of their jobs to other, notably low-wage countries. This fear is far from new, as illustrated by the former U.S. presidential hopeful Ross Perot’s “giant sucking sound” comments (in)famously made in 1992 on the alleged migration of jobs from the United States to Mexico. To assess these developments over time it is useful to define more or less coherent groups of countries, which we label “global regions.” We identify nine global regions: six developing regions and three high-income regions. The six developing regions are based on the World Bank’s grouping in global regions (see the appendix for details): EAP: East Asia and Pacific (includes China and Indonesia) ECA: EastEurope and Central Asia (includes Turkey and Russia)
Cross-border mergers and acquisitions: the facts as a guide for international economics
33
LAC: Latin America and Caribbean (includes Brazil and Mexico) MNA: Middle East and North Africa (includes Egypt) SAS: South Asia (includes India) SSA: Sub-Saharan Africa (includes Nigeria and South Africa)
The World Bank’s group of high-income countries is subdivided into three global regions, following van Marrewijk (2002, Ch. 1; see also Table 2.A1 in the appendix): AAS: AustralAsia (includes Australia, Japan, and South Korea) EUR: Western Europe (includes France, United Kingdom, and Germany) NAM: North America (includes Canada and the United States)
Figures 2.5 and 2.6 depict the evolution over time of the global regions in terms of acquirer and target in cross-border M&As as a percentage of the total value of M&As in the respective year. Western Europe (EUR) is by far the largest acquirer (on average about 55% of the total), followed by North America (30%) and AustralAsia (10%). Over time, the share of Western Europe as an acquirer has increased and of North America has decreased. At the world scale, the importance of East Asia and Pacific (EAP) and Latin America (LAC) as acquirers is limited (between 1 and 2%), and of the other global regions is minimal (less than 1%). Western Europe SAS
SSA
MNA
100
LAC EAP
AAS
ECA 80
60
NAM
40
20
EUR
0 1985
1990
1995
2000
2005
Figure 2.5 Regional distribution of M&A acquirers; value, percentage of total. (EAP ⫽ East Asia and Pacific; ECA ⫽ East Europe and Central Asia; LAC ⫽ Latin America and Caribbean; MNA ⫽ Middle East and North Africa; SAS ⫽ South Asia; SSA ⫽ Sub-Saharan Africa; AAS ⫽ AustralAsia; EUR ⫽ Western Europe; NAM ⫽ North America.)
34
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
SSA
MNA
100
SAS EAP
LAC AAS
ECA
80
60
NAM
40
EUR
20
0 1985
1990
1995
2000
2005
Figure 2.6 Regional distribution of M&A targets; value, percentage of total. (For abbreviations, see Figure 2.5.)
and North America are about equally important as the world’s largest target regions for M&As, on average about 44 and 38% of the world total, respectively. Western Europe has clearly become a more important target region over time, whereas North America’s position has clearly declined. AustralAsia is again third (about 7%), closely followed by Latin America (5%). The importance of Eastern Europe as a target region has clearly increased, as has, to a lesser extent, the importance of East Asia and Pacific and Latin America. The importance of South Asia (SAS), the Middle East and North Africa (MNA), and Sub-Saharan Africa (SSA) as a target region is minimal (less than 1%). In light of the fear of globalization debate alluded to at the beginning of this section, the increased importance of Eastern Europe and also of East Asia and the Pacific and Latin America as target regions provides some evidence that crossborder M&As are increasingly used as vehicles to invest from high-income countries to low-income countries. The changes are, however, (still) modest; it remains true that, even in our regional classification, the vast majority of FDI takes place between and within the three high-income regions. Table 2.3 provides more detail in this respect by giving the regional distribution of cross-border M&As in percentages of the total for acquirer and target region for each of the four 5-year subperiods. It shows, for example, that EUR acquired 48.8% of the cross-border M&As in the period 1986–1990, of which 26.4 percentage points were destined for NAM and 19.8 percentage points for EUR itself. Since then, EUR’s share as an acquirer has exceeded 50%, while its share as a target has been close to 50%. Also note the relative importance of the intra-regional M&As.
Table 2.3 Regional distribution of cross-border M&As; 5-year averages (% of total) Acquirer
Target AAS
EAP
ECA
EUR
LAC
MNA NAM
SAS
SSA
Average value 2001–2005 AAS 3.7 1.0 0.1 EAP 0.4 0.5 0.0 ECA 0.0 0.0 1.4 EUR 2.1 0.5 2.9 LAC 0.0 0.0 0.0 MNA 0.0 0.0 0.0 NAM 2.3 0.7 0.5 SAS 0.0 0.0 0.0 SSA 0.1 0.0 0.0 8.8 2.7 4.9
2.5 0.2 0.2 34.6 0.0 0.4 13.3 0.0 0.1 51.4
0.1 0.1 0.0 2.0 1.4 0.0 1.4 0.0 0.0 5.1
0.0 0.1 0.0 0.4 0.0 0.1 0.1 0.0 0.0 0.7
2.0 0.0 0.1 10.2 0.8 0.0 11.2 0.0 0.1 24.3
0.0 0.0 0.0 0.3 0.0 0.0 0.2 0.1 0.0 0.6
0.0 0.0 0.0 1.1 0.0 0.0 0.1 0.1 0.2 1.5
9.6 1.4 1.7 54.2 2.2 0.5 29.8 0.3 0.5 100
Average value 1996–2000 AAS 2.6 1.6 0.1 EAP 0.4 0.2 0.1 ECA 0.0 0.0 0.1 EUR 1.8 0.4 1.2 LAC 0.0 0.0 0.0 MNA 0.0 0.0 0.0 NAM 2.2 0.3 0.2 SAS 0.0 0.0 0.0 SSA 0.1 0.0 0.0 7.1 2.5 1.7
1.1 0.1 0.0 33.0 0.0 0.0 9.8 0.0 0.5 44.5
0.1 0.0 0.0 3.5 1.6 0.0 2.3 0.0 0.0 7.6
0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.2
1.8 0.1 0.0 21.9 0.2 0.0 11.0 0.0 0.1 35.1
0.1 0.0 0.0 0.1 0.0 0.0 0.2 0.1 0.0 0.5
0.0 0.0 0.0 0.2 0.0 0.0 0.2 0.0 0.4 0.8
7.5 0.9 0.2 62.2 1.9 0.0 26.1 0.2 1.0 100
Average value 1991–1995 AAS 3.0 0.7 0.2 EAP 0.5 0.2 0.0 ECA 0.1 0.0 0.1 EUR 2.6 0.3 1.4 LAC 0.0 0.0 0.0 MNA 0.0 0.0 0.0 NAM 2.3 0.3 0.4 SAS 0.0 0.0 0.0 SSA 0.2 0.0 0.0 8.7 1.5 2.1
1.3 1.2 0.0 34.0 0.1 0.1 12.3 0.0 0.5 49.3
0.2 0.0 0.0 1.4 0.9 0.0 2.4 0.0 0.2 5.1
0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.1
3.1 0.1 0.0 15.4 0.7 0.0 12.7 0.0 0.3 32.3
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1
0.0 0.0 0.0 0.4 0.0 0.0 0.1 0.0 0.2 0.7
8.5 2.0 0.2 55.6 1.8 0.1 30.5 0.1 1.4 100
Average value 1986–1990 AAS 2.5 0.4 0.0 EAP 0.0 0.1 0.0 ECA 0.0 0.0 0.0 EUR 1.0 0.0 0.1 LAC 0.0 0.0 0.0 MNA 0.0 0.0 0.0 NAM 1.3 0.0 0.0 SAS 0.0 0.0 0.0 SSA 0.0 0.0 0.0 4.9 0.5 0.1
4.0 0.0 0.0 19.8 0.0 0.0 6.6 0.0 0.1 30.6
0.0 0.0 0.0 1.2 0.1 0.0 0.4 0.0 0.0 1.8
0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1
9.2 0.3 0.0 26.4 0.3 0.0 24.9 0.0 0.0 61.1
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.1 0.0 0.0 0.3 0.0 0.0 0.4 0.0 0.1 0.9
16.2 0.5 0.0 48.8 0.4 0.1 33.7 0.0 0.2 100
36
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Most noteworthy in Table 2.3 is, of course, the large share of European firms’ buying other European firms, which has been close to one-third of the world total since 1990. It seems easy to argue that the intra-European M&A activity has been stimulated by the process of EU integration, the completion of the single market. But if this is the case, the modern FDI models that serve as benchmarks for our chapter lose some explicatory power because they predict that (horizontal) FDI would become less important. One explanation (BarbaNavaretti and Venables, 2004, chapter 3) might be that (independently from the level of trade costs) the fixed cost of taking over another European firm has fallen because of the streamlining of national legislation. Table 2.3 also shows that the share of intra-regional M&As has been high for AustralAsia and North America (see the following) and that South Asia and the Middle East and North Africa are virtually absent as acquiring and target regions throughout the period. Table 2.4 highlights the changes in the distribution of cross-border M&As by subtracting the percentages in the period 1986–1990 from the percentages of the period 2001–2005 and rounding to the nearest integer. AustralAsia and North America have decreased most substantially as acquirers (minus 7 and minus 4 percentage points, respectively), whereas West and East Europe and Latin America have increased their positions (plus 5, 2, and 2 percentage points, respectively). At the expense of North America (minus 37 percentage points) all the other regions have become more important targets, particularly West and East Europe, AustralAsia, and Latin America (plus 21, 5, 4, and 3 percentage points). The inside of the table shows that the most important distributional change has been European firms’ buying European instead of American firms, and similarly for American firms. Table 2.4 Change in regional distribution of cross-border M&As; 2001–2005 5-year average minus 1986–1990 5-year average, rounded to nearest integer Acquirer
Target AAS
EAP
ECA
Average value 2001–2005 AAS 1 1 EAP ECA 1 EUR 1 1 3 LAC MNA NAM 1 1 SAS SSA 4
2
5
For abbreviations, see Figure 2.5.
EUR
LAC
MNA NAM
⫺1
SAS
SSA
⫺7
15
1 1
⫺16
7
1
⫺14
21
3
1
⫺37
1
⫺7 1 2 5 2 ⫺4
1
1
Cross-border mergers and acquisitions: the facts as a guide for international economics
37
Finally, we focus attention on inter-regional M&As, which gives us an indication of the extent to which different global regions interact with one another. These flows can obviously be (roughly) deduced from Table 2.3 or the various subperiods by disregarding the diagonal entries (which sum to ⬃50% of the total) and re-adjusting the remaining entries to sum to 100% inter-regional M&As. Figure 2.7 graphically depicts the inter-regional cross-border connections for the most recent 5-year period (2001–2005), rounded to the nearest integer. Since there are 9 global regions, there are 72 different inter-regional connections. Only 19 of these are actually shown in Figure 2.7 because the remaining 53 are rounded to 0%. First, we note that by far the largest inter-regional M&As are from North America to Western Europe (28% of the total), and vice versa (22% of the total). Together these two flows account for 50% of all interregional M&As and clearly dwarf the importance of all other inter-regional connections. Second, Western Europe is substantially buying up firms in Eastern Europe (6%). Third, the other connections between the high-income regions (between EUR and AAS and between NAM and AAS) are substantial (about 5% each). Fourth, M&As toward East Asia and the Pacific are still rather
Developing region
The thickness of an arrow is proportional to the size of the M&A flow (percent of total interregional M&As, round to nearest integer)
High income region
LAC Latin America & Caribbean
SSA Sub-Sahara Africa
ECA East Europe & Central Asia
2
1
2
4
3
6
22 NAM North America 1
EUR West Europe
28
5
5
1 4
EAP East Asia & Pacific
1 5
2 AAS Austral Asia 1
SAS South Asia
1
1 MNA Middle East & North Africa
Figure 2.7 Inter-regional cross-border; percentage of total (value), 2001–2005. [NB: All intra-regional M&As are excluded from the figure. The total value of interregional M&As is 100%; only flows above 0.5% are shown (this excludes 53 of 72 possible arrows).]
38
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
small, certainly compared to the attention this activity receives in the popular media. Fifth, and finally, Western Europe is the only global region with connections to all other regions. This is reminiscent of the dominance of Western Europe in inter-regional trade flows; see van Marrewijk (2007, Ch. 1). So, it seems safe to conclude this section with the observation that, indeed, most FDI and M&As take place between the relatively wealthy parts of the world. This observation is in accord with our previous findings that cross-border M&As are mainly of the horizontal type.
2.5
Countries and M&As over time
In view of the high coincidence between acquiring and target countries discussed in Sections 2.3 and 2.4, it is interesting to make a distinction between the largest gross acquirers and targets and the largest net acquirers and targets of M&As. Looking at net figures corrects for (country) size differences and reveals possible changes in the direction of FDI flows. Since the value and number of cross-border M&As varies substantially, even for the world as a whole (see Figure 2.3), it should come as no surprise that this variation is even more substantial at the country level, certainly when we look at net M&A flows. This variations is illustrated in Figure 2.8 for the two largest net acquiring countries (United Kingdom and France) and net target countries (United States and 200
UK France
150
Net acquiring M&As
100
USA
50 UK 0
France
UK
UK
Brazil 1985
2005 USA
−50
USA
Brazil
USA
UK
−100 USA −150
Figure 2.8 Cross-border M&As; four largest net acquirers and net targets.
year
Cross-border mergers and acquisitions: the facts as a guide for international economics
39
Brazil) for the period 1985–2005. For the United Kingdom, for example, the fluctuations around the average of $19.1 bn per year range from a low of ⫺$78 bn in 2004 to a high of $295 bn in 1999. Similarly, for the United States, the fluctuations around the average of ⫺$31.3 bn per year range from a low of ⫺$205 bn in 2000 to a high of $46 bn in 2003. To mitigate the impact of fluctuations over time and to identify important trends over longer time periods, Table 2.5 lists the most important countries for each of the four categories identified above for the period 1986–2005 as a whole, subdivided into four 5-year subperiods. Table 2.5a lists the top 10 acquiring countries, consisting of the United States, Canada, Australia, Japan, and six European countries (United Kingdom, France, Germany, The Netherlands, Switzerland, and Spain). The United States and the United Kingdom are about equally important in this respect, although the United States tops the list in three of the four subperiods. The role of The Netherlands and Spain as acquiring nations has become more important in the last 10 years and that of Australia in the last 5 years. In contrast, the role of Japan as an acquiring nation has clearly reduced over time. Table 2.5b lists the top 10 target countries. Except for Italy and Sweden (which replace Switzerland and Japan), it consists of the same countries as those listed in Table 2.5a. The United States is undisputedly the largest target country, followed by the United Kingdom and Germany. The role of the United Kingdom as a target country has clearly increased over time. Similarly, to a lesser extent, has the role of other European countries, particularly in the last 5 years. Table 2.5c lists the top five net acquiring countries, consisting of five European countries: France, United Kingdom, Switzerland, The Netherlands, and Spain. Of these five, Switzerland and The Netherlands have been stable net acquiring countries throughout the time period, whereas the net position of France has been more volatile. The United Kingdom’s net position recently switched from acquirer to target, and vice versa for Spain. Finally—and most interestingly from the globalization-debate perspective— Table 2.5d lists the top five net target countries, consisting of the United States, Brazil, Germany, China, and Argentina. Of these five, Brazil and Argentina have been stable net target countries throughout the period, whereas China, like Germany, became an important net target in the last 10 years only. The United States has been a primary net target most of the time, switching roles with the United Kingdom only in the last 5 years. The analysis reveals that, despite the dominant position of the United States, recently high-income countries are turning toward emerging markets, of which China stands out as the most recent net target. Folk wisdom about the increasing importance of China—and other promising markets—thus seems correct in this respect. This trend also implies a challenge for FDI modeling. Typically (see Barba-Navaretti and Venables, 2004), when the possibility of M&A as an FDI option is taken into account, it occurs in models of horizontal FDI, which (given the facts we have presented so far) should not come as a surprise. But the information provided by Table 2.5d suggests that (increasingly?) cross-border M&As are also aimed at low(er)-income
40
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 2.5 Largest M&A countries; acquiring and targets, gross and net flows Country
Annual average acquiring flows 1986–1990
1991–1995
1996–2000
2001–2005
1986–2005
a. Ten largest acquiring M&A countries, 1986–2005 (constant 2005 $ billion) 1 United States 41.1 42.3 142.3 118.0 2 United Kingdom 37.2 27.0 200.3 76.7 3 France 17.0 13.2 85.6 34.9 4 Germany 6.4 10.7 68.7 31.5 5 Netherlands 4.3 8.1 39.8 32.8 6 Canada 13.3 7.5 29.9 24.1 7 Switzerland 6.1 8.0 28.8 15.8 8 Spain 2.0 2.9 27.1 24.5 9 Australia 8.2 3.7 14.1 21.4 10 Japan 16.0 3.7 13.7 8.9
85.9 85.3 37.7 29.3 21.2 18.7 14.7 14.1 11.9 10.6
b. Ten largest target M&A countries, 1986–2005 (constant 2005 $ billion) 1 United States 86.5 44.6 238.4 99.6 2 United Kingdom 29.6 22.7 119.7 92.7 3 Germany 4.1 7.9 83.3 40.3 4 Canada 11.3 6.9 37.6 22.2 5 France 5.8 12.9 28.9 26.1 6 Netherlands 3.0 5.7 29.4 20.6 7 Australia 4.1 8.4 18.0 13.5 8 Italy 3.8 5.8 10.4 21.8 9 Sweden 1.7 4.9 23.7 10.3 10 Spain 3.1 5.0 11.1 11.8
117.3 66.2 33.9 19.5 18.4 14.7 11.0 10.5 10.2 7.8
c. Five largest net acquiring M&A countries, 1986–2005 (constant 2005 $ billion) 1 France 11.3 0.3 56.8 8.8 19.3 2 United Kingdom 7.6 4.3 80.6 ⫺16.0 19.1 3 Switzerland 3.5 5.5 20.3 8.0 9.3 4 Netherlands 1.3 2.4 10.3 12.3 6.6 5 Spain ⫺1.1 ⫺2.1 16.0 12.7 6.4 d. 1 2 3 4 5
Five largest net target M&A countries, 1986–2005 (constant 2005 $ billion) United States 45.4 2.3 96.1 ⫺18.4 31.3 Brazil 0.2 0.6 18.6 2.8 5.6 Germany ⫺2.4 ⫺2.7 14.5 8.8 4.6 China 0.0 0.2 11.4 5.3 4.2 Argentina 1.8 1.3 11.0 1.8 4.0
countries where the market-seeking aspect is probably far less relevant than the (labor) cost-saving argument. This means that cross-border M&A should be part of models of vertical FDI as well. It might be that cross-border M&As become an increasingly viable alternative for Greenfield FDI or outright outsourcing in view of the well-known asymmetric information problems (the holdup problem) associated with the FDI-versus-outsourcing decision.
Cross-border mergers and acquisitions: the facts as a guide for international economics
2.6
41
Inequality between cross-border M&As
One of the reasons the M&A phenomenon attracts attention inside—and certainly outside—academia is undoubtedly the involvement of national pride in M&A deals (either positively or negatively). Another, perhaps even more important, reason for this attention is the size of some of the cross-border M&A. Indeed, some of the deals are so large that they can have a substantial influence on a country’s position as a (net) acquirer or target. Table 2.6 lists the largest deals by year of announcement, valued in current and constant dollars, as well as the two countries involved in each deal. Several conclusions can be drawn from this table. First, there is substantial variation in the maximum value over time (a 50-fold difference between the highest and lowest value). Second, a single deal can indeed have a substantial influence. The Vodafone takeover of Mannesmann (see Section 2) is by far the largest M&A. It has also clearly influenced the net acquiring position of the United Kingdom and the net target position of Germany. Third, the United States is by far the most popular target country for these mega-deals (12 out of 20 observations), while Europe is the Table 2.6 Value of largest cross-border M&As (announced year) Year
Value of deal (bn) Current $
Constant 2005 $
Firm and country information Acquiring
Target
Firm
Country
Firm
Country
Campeau BP America Campeau Beecham Matsuhita E
Canada U.S.A. Canada U.K. Japan
Allied Stores Standard Oil Fed Dep St. Smith Kline MCA
U.S.A. U.S.A. U.S.A. U.S.A. U.S.A.
Altus Fin. Reed Metro etc. Roche Hoechst
France U.K. Malaysia Switzerl. Germany
Ex. Life Elsevier ASKO etc. Syntex Marion etc.
U.S.A. Netherl. Germany U.S.A. U.S.A.
1986 1987 1988 1989 1990
3.6 7.9 6.5 7.9 7.4
5.6 12.0 9.7 11.4 10.3
1991 1992 1993 1994 1995
3.3 4.6 6.3 5.3 7.3
4.3 6.0 8.0 6.5 8.8
1996 1997 1998 1999 2000
4.2 17.1 48.2 202.8 46.0
5.0 19.8 54.9 228.7 51.1
Fresenius Zürich Vers BP Vodafone France Tel
Germany Switzerl. UK U.K. France
Nat Med ca BAT Ind Amoco Mannesmann Orange
U.S.A. U.K. U.S.A. Germany U.K.
2001 2002 2003 2004 2005
12.8 15.3 11.1 74.6 31.7
14.0 16.2 11.6 76.5 31.7
Citigroup HSBC Manulife R D Petrol Telefonica
U.S.A. UK Canada Netherl. Spain
Banacci Household I J Hancock Shell Transp O2
Mexico U.S.A. U.S.A. U.K. U.K.
42
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
most popular acquiring region (13 out of 20 observations). Fourth, and finally, even when measured in constant dollars, there seems to be a tendency for the maximum value to increase over time.3 This has led to the suggestion in the literature that the size distribution of M&As has become more unequal over time (Evenett, 2004). This section analyzes that suggestion in more detail. A proper understanding of the degree of inequality of a distribution must, of course, take all observations into consideration, rather than focusing just on the maximum value. An excellent, and popular, method is to construct Lorenz curves, where the observations are ordered in increasing value, with the share of the cumulative number of deals on the horizontal axis and the share of the cumulative value of these deals on the vertical axis. Figure 2.9 provides examples of these curves in the years 1991, 1999, and 2005. If all the observations in a particular year had an equal value, the Lorenz curve would coincide with the diagonal. The area below the diagonal and above the curve (times 2) therefore provides a measure of the inequality of the observations, a number between 0 (complete equality) and 1 (complete inequality) known as the Gini coefficient. We calculated the Gini coefficient for each year of our dataset.
Cumulative value of deals (share)
1
1991 2005 1999
0 0
Cumulative number of deals (share)
1
Figure 2.9 Lorenz curves of cross-border M&As; selected years. 3
A trendline of the logarithm of the maximum value in the period 1985–2005 explains about half of the variance and suggests a rate of increase at 0.13% per year.
Cross-border mergers and acquisitions: the facts as a guide for international economics
43
Figure 2.10 provides an overview of the evolution of the Gini coefficient over time for the period 1986–2005. There is indeed a tendency of the Gini coefficient to increase over time, supporting the suggestion that the degree of inequality in cross-border M&As increases over time.4 The variation from year to year is substantial, however, ranging from a low of 0.649 in 1991 to a high of 0.853 in 1999; see also the associated Lorenz curves in Figure 2.9. More importantly, by including the evolution over time of the total value of cross-border M&As in the same diagram, Figure 2.10 draws attention to the relationship between inequality as measured by the Gini coefficient and the wave phenomenon. Clearly, the Gini coefficient increases during the 4th wave of the late 1980s, then declines after this peak has been reached, to increase again during the 5th wave of the late 1990s, to decline again after the absolute peak in 1999, and starts to increase again during the 6th wave starting in 2003. Figure 2.11 illustrates the coincidence of changes in inequality, measured by the relative change in the Gini coefficient, and merger waves, measured by the relative change in the value of cross-border M&As. There is a clear positive relationship between these two phenomena. If we let GIt be the Gini coefficient in year t, Vt be the value of cross-border M&As (in constant 2005 $ bn), ~ ⬅ (x ⫺ x )/x and ~ denote a relative change, that is x t t t⫺1 t⫺1 for xt ⫽ GIt,Vt, then we get (t-values in parentheses) ≈ ⫺1.1207 ⫹ 0.1085 ⋅ V GI t t (⫺1.76)
R2 ⫽ 0.75
(2.2)
(7.38)
1,500
Value in 2005 $ billion
Gini coefficient
0.9
Gini coefficient
Value
0.6 1985
1990
1995
2000
0 2005
Year
Figure 2.10 Cross-border M&As, 1985–2005; Gini coefficients and value. 4
A trendline of the Gini coefficient explains almost half of the variance and suggests an increase in the Gini value at a rate of about 0.0061 per year.
44
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Relative change in Gini coefficient (%)
12 1998 8
4 2005 0 −60
−30
0
30
60
90
−4 −8 1991 −12 Relative change in value of M&A (%)
Figure 2.11 Relative changes in value of M&As; Gini coefficient, 1986–2005.
A 1% increase in the value of cross-border M&As therefore causes about a 0.1% increase in the Gini coefficient. The gradual increase in the real value of crossborder M&As over time is therefore probably largely responsible for the observed increase in inequality.5 This begs the question of what causes the increase in the value of the M&As during the 1990s. The most important reason is that regulations with respect to M&A have changed over time. Especially the financial service sector, banking sector, (tele)communication sector, and media firms have been allowed to merge with or acquire overseas firms (Evenett, 2004; Muelfeld, Sahib, and van Witteloostuijn, 2007). Once the regulations became more relaxed, the local giants were looking for profitable M&As. Does this make sense from the perspective of the modern FDI theories that play such a dominant role in the current research in international economics? To start with, the idea of merger waves can be explained, as equation (2.1) already suggests, that once the initial mergers have taken place and competition is reduced, it becomes profitable for other firms also to engage in the M&A activity; however, the problem of explaining the initial mergers remains. The association of the merger wave (with a strong European flavor) with changes in regulation can be aligned with theoretical models of (horizontal) FDI as long as we look upon it as a decrease in the organizational costs of setting up and arranging an M&A. If these costs are seen as a fall in trade costs—and hence as a manifestation of increased economic integration—these models are, as we have argued before, ill equipped to explain the 5
It should be noted, moreover, that changes in the maximum M&A value are only weakly (positively) correlated with changes in the Gini coefficient.
Cross-border mergers and acquisitions: the facts as a guide for international economics
45
data in Figure 2.10. Maybe the limits of existing FDI models which more or less all build on equation (2.1) come to the fore here; thus, to understand what is driving merger waves, we might look at alternative theories like the managerial hubris theory (Roll, 1986). Managers tend to err positively when it comes to the valuation of targets and hence tend to overpay. Especially during the heyday of the dotcom bubble in the late 1990s, this phenomenon could explain the increase in value of the M&As. Although managerial hubris is not part of our categorization scheme from Box 2.1, it correlates with the fact that M&A are facilitated in the upswings of business cycles.
2.7
Looking more closely at individual firms that engage in M&As
Until now we have not discussed individual firms.6 In this section we briefly discuss the main insight that results from the research on FDI and firm heterogeneity: namely, that within a sector there is considerable firm heterogeneity to the effect that only the most productive firms are expected to be engaged in FDI and thus in cross-border M&A (as an acquiring firm). The idea that firms from the same sector differ (considerably) is probably not a path-breaking observation, but for the fact that there is a systematic relationship between plant productivity and the mode of entry in international trade. Bernard, Eaton, Jensen, and Kortum (2003) showed that a systematic relationship exists between productivity and whether or not firms are engaged in exports. They show that of 200.000 (U.S.) firms in their sample, only 21% report any export. Fewer than 5% of these firms export more than 50%, which shows that even if firms are engaged in international trade, most are still most active in domestic markets—two-thirds of the exporters export less than 10% of their output. Most interestingly, those that export have higher productivity levels, and thus are able to charge a higher markup. Given the fact that international trade is more costly than domestic sales, only productive firms are able to cover trade costs. Despite these trade costs they can still be competitive in foreign markets, just because they are efficient. So, export reveals highly productive plants. Helpman, Melitz, and Yeaple (2004) take this line of reasoning one step further not by only looking at the export decision, but also by taking the FDI decision into account. Because FDI is even more expensive than exports, only the most efficient firms are able to engage in FDI. They find strong evidence for a sample of U.S. and European firms that only the most productive firms are engaged in FDI. Studies like these confirm the notion that transportation costs are important to describe not only international trade patterns, but also FDI flows. 6
The Thomson data do not allow us to calculate the productivity measures as used by Bernard., Eaton, Jensen, and Kortum (2003); neither can we differentiate between domestic sales exports or FDI at the plant level. In this section we review some of the relevant literature that has original plant level data on productivity.
46
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
We report these results because they also explain why most FDI is between rich countries (see Sections 2.3 and 2.4). Instead of emphasizing market-seeking arguments, the firm heterogeneity argument points out that most FDI, and hence cross-border M&As, is between rich countries because that is where the most productive firms are located. Our dataset, the Thomson dataset on M&A, does not allow for an easy differentiation of firms in terms of productivity, but additional stylized facts on the productivity of firms engaged in cross-border M&As could help to establish if this new firm heterogeneity literature makes sense when applied to M&As. We leave this for future research.
2.8
Conclusion
The well-known advice of Leamer and Levinsohn (1995) to “estimate, don’t test” implies that, given the current state of the theories and the quality of the data, the Popperian test of falsifying a theory is hardly possible: “We may statistically ‘reject’ the theory, but leave it completely unharmed nonetheless. After all, we already knew it wasn’t literally true” (ibid. p. 1314). What empirical work should be doing, according to them, is “not to test the validity of the theory but to determine if the theory is working adequately in its limited domain” (ibid. p. 1342). So, in practice the distinction between verifying or falsifying theories is less clearcut than one would ideally want. This boils down to asking theorists to think about the link between theory and observable phenomena. The aim of this chapter is to present the correlations in the data on crossborder M&As, and ask the theorists to develop useful models that give us some understanding about the underlying causation. Our chapter provides guidelines for theory on an very important phenomenon, cross-border M&As, as to what the most important correlations might be. Using the well-known Thomson dataset, we show that cross-border M&As) have a number of features: Most FDI is in the form of cross-border M&As. Firms engaged in cross-border M&As seem to be “market-seeking.” Cross-border M&As come in waves (the most recent wave is still unfolding). Economic integration (international deregulation) stimulates M&As. The size of and inequality between M&As grows (over time).
Our contention in this chapter is that these stylized facts drive and should drive theoretical contributions from international economics that try to understand cross-border M&As. A number of recent models that are firmly rooted in the first principles of trade theory (see Neary, 2003) go a long way in explaining some of these facts. What is still missing, given our stylized facts, is a full-fledged model of M&As. It might be, of course, that tools of modern international economics do not allow for such an all encompassing theory, but ongoing research by economists like Neary, Helpman, or Melitz suggests that our understanding of cross-border M&As will improve in the near future. This is real progress because, from the perspective of mainstream international economics, cross-border M&As have too long been a case of interesting facts in search of a theory.
Cross-border mergers and acquisitions: the facts as a guide for international economics
47
Appendix Table 2.A1 Global regions: country composition EAP: East Asia and Pacific; 27 developing countries American Samoa Marshall Islands Cambodia Micronesia China Mongolia Fiji Myanmar Indonesia Nauru Kiribati N. Mariana Islands Korea, North Palau Laos Papua New Guinea Malaysia Philippines ECA: Europe and Central Asia; 28 developing countries Albania Hungary Armenia Kazakhstan Azerbaijan Kyrgyz Republic Belarus Latvia Bosnia-Herzegovina Lithuania Bulgaria Macedonia Croatia Moldova Czech Republic Poland Estonia Romania Georgia
Samoa Solomon Islands Taiwan Thailand Timor, East Tonga Tuvalu Vanuatu Vietnam Russia Serbia and Montenegro Slovak Republic Tajikistan Turkey Turkmenistan Ukraine Uzbekistan Yugoslavia
LAC: Latin America and the Caribbean; 33 developing countries Argentina Ecuador Nicaragua Barbados El Salvador Panama Belize French Guiana Paraguay Bolivia Grenada Peru Brazil Guatemala St Kitts and Nevis Chile Guyana St Lucia Colombia Haiti St Vincent & Grenadines Costa Rica Honduras Suriname Cuba Jamaica Trinidad And Tobago Dominica Martinique Uruguay Dominican Republic Mexico Venezuela MNA: Middle East and North Africa; 14 developing countries Algeria Jordan Syria Djibouti Lebanon Tunisia Egypt Libya West Bank Iran Morocco Yemen Iraq Oman SAS: South Asia; 8 developing countries Afghanistan India Bangladesh Maldives Bhutan Nepal
Pakistan Sri Lanka
48
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
References Andrade, G., Mitchell, M., and Stafford, E. (2001). New Evidence and Perspectives on Mergers, Journal of Economic Perspectives, 15(2):103–120. Barba-Navaretti, G., and Venables, A. J. (2004). Multinational Firms in the World Economy. Princeton, N.J., Princeton University Press. Bernard, A., Eaton, J., Jensen, B., and Kortum, S. (2003). Plants and Productivity in International Trade. American Economic Review, 93:1268–1290. Brakman, S., Garretsen, H., and van Marrewijk, C. (2005). Cross-Border Mergers and Acquisitions: On Revealed Comparative Advantage and Merger Waves. CESifo Working Paper No. 1602. Brakman, S., Garretsen, H., and van Marrewijk, C. (2006). Comparative Advantage, Cross-Border Mergers, and Merger Waves: International Economics Meets Industrial Organization. CESifo Forum 1, pp. 22–26. Brakman, S., Garretsen, H., van Marrewijk, C., and van Witteloostuijn, A. (2006). Nations and Firms in the Global Economy. Cambridge, U.K., Cambridge University Press. Dunning, J. H. (1993). Multinational Enterprises and the Global Economy. Wokingham, Addison–Wesley. Evenett, S. J. (2004). The Cross Border Mergers and Acquisitions Wave of The Late 1990s. In (Baldwin, R. E., and Winters, L. A., eds.), Challenges to Globalization. Chicago, University of Chicago Press. Fridolfsson, S.-O., and Stennek, J. (2005). Why Mergers Reduce Profits and Raise Share Prices—A Theory of Preemptive Mergers. Journal of the European Economic Association, 3(5):1083–1104. Gugler, K., Mueller, D. C., Yurtoglu, B. B., and Zulehner, C. (2003). The Effects of Mergers: An International Comparison. International Journal of Industrial Organization, 21:625–653. Helpman, E., Melitz, M. J., and Yeaple, S. R. (2004). Exports versus FDI with Heterogeneous Firms, American Economic Review, 94:300–316. Helpman, E. (2006). Trade, FDI, and the Organization of Firms. NBER Working Paper, No. 12091, Cambridge. Leamer, E. E., and Levinsohn, J. (1995). International Trade Theory: The Evidence. In Handbook of International Economics (Grossman, G. M., and Rogoff, K., eds.), Vol. 3. Amsterdam, North-Holland. Muelfeld, K., Sahib, P. R., and van Witteloostuijn, A. (2007). Completion or Abandonment of Mergers and Acquisitions: Evidence from the Newspaper Industry, 1981–2000. Journal of Media Economics, forthcoming. Neary, J. P. (2003). Globalization and Market Structure. Journal of the European Economic Association 1:245–271. Neary, J. P. (2004a). Cross-Border Mergers as Instruments of Comparative Advantage (mimeo). Dublin. Neary, J. P. (2004b). Monopolistic Competition and International Trade Theory. In The Monopolistic Competition Revolution in Retrospect
Cross-border mergers and acquisitions: the facts as a guide for international economics
49
(Brakman, S., and Heijdra, B. J., eds.). Cambridge, U.K., Cambridge University Press. Neary, J. P. (2005). Trade Costs and Foreign Direct Investment. CESifo Summer Institute Workshop Recent Developments on International Trade: Globalization and the Multinational Enterprise, Venice, July 2005. Roll, R. (1986). The Hubris Hypothesis of Corporate Takeovers, Journal of Business, 59:197–216. UNCTAD, 2000. World Investment Report. Geneva, Switzerland. van Marrewijk, C. (2002). International Trade and the World Economy. Oxford, U.K., Oxford University Press. van Marrewijk, C. (2007). International Economics: Theory, Application, and Policy. Oxford, U.K., Oxford University Press.
This page intentionally left blank
3 Searching for value-enhancing acquirers Manolis Liodakis and Che Pang
Abstract Although the typical acquirer underperforms the market in the long term, not all acquisitions destroy value. We search for factors that could help investors screen for value-enhancing acquirers. For instance, cash-financed deals within the same industry involve relatively large targets, enjoying better fortunes, and bidders who trade at a valuation discount to their sector typically do better. The bid premium and the short-term market reaction to the deal announcement are also important leading indicators. Our basket of “potential value–creating” acquirers outperformed the market by 24% and the typical bidder by 37%.
3.1
Introduction
Mergers and acquisitions (M&A) are the key to driving European equities higher. Every day, it seems, a new deal is announced or a share price has spiked in anticipation of the next bid. The mood is febrile. With the M&A activity accelerating in Europe, investors cannot afford to ignore these transactions. But is this situation likely to continue? Most market commentators think this is just the beginning. The absolute numbers might be big ($767 bn completed U.S./European M&As in 2005), but as a percentage of overall market, they are still just half the two previous peaks reached in the late 1980s and 1990s (see Figure 3.1). In our view, the current M&A cycle is mainly driven by low interest rates. The profit recovery since 2003 has allowed companies to repair balance sheets. With profitability remaining strong, increasingly confident company CEOs are using cheap-debt financing to drive a new cycle of industry consolidation. In addition, cashflow that might be spent on capex is being diverted toward M&A. At this stage of the cycle CEOs prefer to buy rather than to build. At the same time, private equity investors are using that cheap debt to remove companies from the stock market. All debt-financed deals are currently accretive. Cheap debt is driving the current M&A feeding frenzy. While interest rates remain so low, it seems unlikely to stop. So how can investors profit from this trend? The obvious way is to attempt to predict takeover targets (Liodakis and Brar, 2005). Takeover targets are the
52
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Forecast
25
20
(%)
15
10
5
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006e 2007e 2008e 2009e
0
EuropeóCompleted M&A, % of market capitalisation USóCompleted M&A, % of market capitalisation
Figure 3.1 M&A cycle (% of market capitalization). Source: SDC and Citigroup Investment Research.
companies that receive most of the benefits from acquisitions. Forecasting targets, however, involves taking a great deal of risk. Another option is to focus on acquirers. But do acquisitions create any value for the shareholders of the bidding firms? Should investors continue to hold these acquiring companies in their portfolios, and under what conditions should they consider selling them? Our report focuses on two main research issues. What is the short- and longterm performance of acquirers following the announcement of a deal? And what are the characteristics of value-enhancing, outperforming acquirers?
3.1.1
A closer look at our M&A database
To address these issues we have put together a comprehensive database of publicly announced M&A deals from January 1990 to the end of February 2006 involving European acquirers. Our data source is Thomson SDC Database and Bloomberg. Having excluded deals with a transaction value of less than USD 100 m, we found data on 4866 announced transactions. Of those deals, 4070 or 84% of the total were completed. Figures 3.2 and 3.3 show the breakdown of deals by country and by year. We report both announced and failed deals. The U.K. dominates the deal activity in the last 15 years with 1276 completed deals, followed by France with 654. Most of the deals took place at the peak of the bull market in 1999 and 2000. More than 25% of our sample was concentrated in these 2 years.
Searching for value-enhancing acquirers
53
1600 1400 Number of deals
1200 1000 800 600 400 200
Announced deals
France
United Kingdom
Italy
Germany
Spain
Sweden
Netherlands
Belgium
Switzerland
Finland
Norway
Portugal
Denmark
Ireland-Rep
Austria
Greece
Luxembourg
Poland
Hungary
Iceland
Czech Republic
0
Failed deals
Figure 3.2 Deal breakdown by country. Source: Thomson SDC and Citigroup Investment Research.
800 700
Number of deals
600 500 400 300 200 100 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 YTD Deals completed
Failed deals
Figure 3.3 Deal breakdown by years. Source: Thomson SDC and Citigroup Investment Research.
54
3.2
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Do takeovers create any value?
Our interest is in calculating both the short-term market reaction and the longterm share price performance of bidders after the deal announcement date. Short-term returns are measured from 10 days prior to and 10 days after the bid announcement date. The rather wide window of 20 days ensures that we capture any bid speculation as well as important details that often become public just after the initial announcement. The long-term performance of acquirers is evaluated using a 3-year window. Synergies and cost benefits often require a significant amount of time to materialize, so evaluating the success of a deal using a 3-year horizon seems sensible. As the performance figures can be influenced by broad market movements, we calculate returns in excess of the market index. We use the Pan-European S&P/Citigroup BMI as our market index. We do this daily for each bidder by taking the difference between the bidder’s return and the market index return. Therefore, each bidder’s excess return is calculated over each day of the event period as follows: Excess return1 ⫽ Bidder stock return ⫺ S&P/Citigroup BMI Return
(3.1)
The average abnormal return (AAR) is then calculated by taking the equal weighted average of the excess returns of all bidders. To reduce the effects of outliers, we use trimmed means to remove any observations outside of 3 standard deviations. Also, to avoid look-ahead bias, excess returns of all the bidders, and not just those of bidders involved in completed deals, are included in the sample. If the deal is withdrawn at a later date, the bidder will then be excluded from that date onward. The final step is to calculate the cumulative average abnormal return (CAAR) for each day over the entire event window. For deals that were done less than 3 years ago, we include the performance of the bidder in calculations for each day available, thereby ensuring that our results are as complete as possible and that we do not neglect the performance of deals done within the past 3 years. Figure 3.4 shows the CAARs of acquirers from 10 days prior to 10 days post the deal announcement. The results show that these companies outperform modestly (⫹60 bps) around the announcement of the transaction. A small fraction of these returns come from trading before the announcement, but the bulk of the outperformance is concentrated on the day the bid intention becomes public. 1
If we were to measure excess returns just over the short-term horizon, we could have used an alternative “beta-adjusted” approach to calculate excess returns by regressing the returns of the bidder in question against the returns of the market index over a period prior to the announcement date. Given the estimated alpha and beta from this regression, the one-day expected return for a bidder would be equal to the alpha plus the beta times the return on the market index. The one-day excess return for the bidder would then be equal to the bidder’s return less its expected return. However, because we were also measuring excess returns over the longer term of 3 years, we believed that using this method, which assumes a constant beta over this time period, was not sensible.
Searching for value-enhancing acquirers
55
Although the short-term market reaction seems negligible, there is a considerable variation in bidders’ returns. We ranked all the acquirers by their 20-day CAARs and allocated them into quintiles (5 groups with equal numbers of stocks) and calculated the average cumulative outperformance as shown in Figure 3.5.
Average market relative returns (%)
0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 −10 −9 −8 −7 −6 −5 −4 −3 −2 −1 0
1
2
3
4
5
6
7
8
9 10
No. of trading days around merger announcement
Figure 3.4 Short-term market relative performance of acquirers around merger announcement. Source: Exshare, Thomson SDC and Citigroup Investment Research.
Cumulative average excess return (−10, 10 days) (%)
15 10 5 0 −5 −10 −15 Most value Quintile 2 destroying acquirers
Quintile 3
Quintile 4
Most value creating acquirers
Figure 3.5 Acquirer’s dispersion in short-term returns. Source: Exshare, Thomson SDC and Citigroup Investment Research.
56
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
We find that the worst performing group underperforms by 10% on average, whereas the best performing group outperforms by 11%. This extraordinary cross-sectional volatility indicates that the market does not treat all acquisitions in the same way. This presents an interesting opportunity for active investors. We have also examined the long-term track record of acquirers. The results here were somewhat disappointing. Figure 3.6 shows that the typical acquirer underperforms the broad market index by 13% 3 years after the deal announcement. Interestingly the share price weakness is fairly consistent throughout the postmerger period. Although the average number suggests that acquisitions destroy value there is again a considerable dispersion in acquirers’ long term returns. Focusing just on the average return tends to hide the enormous variability in performance and hence the opportunity that investors have to add value by backing the “right” acquirers. As Figure 3.7 shows, the bottom 20% of the worst performing acquirers (by the 3-year CAAR) have underperformed by 42% while the top 20% have outperformed by 57%. Table 3.1 shows the short- and long-term market reaction to M&A announcement for each year separately. As expected, the worst returns for bidders occurred at the peak of the equity market cycle (1999 and 2000). Also interestingly, despite M&A’s poor long-term track record, acquirers have been rewarded by the market recently. The average 10-day post-deal announcement excess return of bidders was 3.7% in 2006. We have also broken down the average short- and long-term returns of acquirers by sector. Our results suggest that takeovers have destroyed value in
Average market relative returns (%)
2 0 −2 −4 −6 −8 −10 −12 −14 −16 −10 Days
1Yr
2Yr
Number of years post merger announcement
Figure 3.6 Long-term market relative performance of acquirers post merger announcement. Source: Exshare, Thomson SDC and Citigroup Investment Research.
3Yr
Cumulative excess return (−10, 3 years) (%)
Searching for value-enhancing acquirers
57
80 60 40 20 0 −20 −40 −60
Most value Quintile 2 destroying acquirers
Quintile 3
Quintile 4
Most value creating acquirers
Figure 3.7 Dispersion in acquirer’s long-term returns. Source: Exshare, Thomson SDC and Citigroup Investment Research. Table 3.1 Performance breakdown by year Year
10 Days (%)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
⫺1.2 0.6 ⫺0.7 1.1 ⫺0.6 ⫺0.3 2.0 1.0 1.7 1.3 ⫺1.6 0.4 0.5 1.8 1.5 1.7 3.7
1 Year (%) ⫺5.1 2.7 ⫺1.4 1.1 ⫺2.0 ⫺4.7 ⫺9.6 ⫺4.5 ⫺12.9 ⫺14.8 ⫺15.9 ⫺0.1 3.8 6.0 4.1 ⫺1.3 NA
3 Years (%) ⫺12.8 ⫺4.5 ⫺15.6 ⫺9.8 ⫺16.1 ⫺29.0 ⫺29.0 ⫺30.0 ⫺27.0 ⫺23.2 ⫺30.6 ⫺0.3 7.3 7.1 NA NA NA
Source: Thomson SDC and Citigroup Investment Research.
Tech and Telcos, but have had a better long-term track record in Energy and Health Care (Table 3.2). As mentioned previously, all these numbers represent the average, obscuring the enormous variability in the performance of bidders. In the next section we
58
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 3.2 Performance breakdown by sector
Consumer staples Industrials Health care Materials Consumer discretionary Information technology Utilities Financials Energy Telecommunication services
10 days (%)
1 year (%)
3 years (%)
1.9 1.0 2.6 0.8 1.8 ⫺1.3 ⫺0.6 ⫺0.1 1.4 ⫺0.5
3.2 ⫺7.7 7.6 ⫺3.9 ⫺4.9 ⫺16.7 ⫺1.3 ⫺1.9 6.3 ⫺18.3
0.1 ⫺16.6 2.1 ⫺10.0 ⫺21.1 ⫺47.5 0.0 ⫺4.4 17.3 ⫺41.7
Source: MSCI, Thomson SDC and Citigroup Investment Research.
focus on the extremes and highlight the factors that could help investors discriminate between value-enhancing/outperforming and value-destroying/ underperforming bidders.
3.3
Motives for deals
To assess the characteristics of profitable acquirers, we must first understand the underlying motives for corporate acquisitions. A takeover deal is desirable for acquiring shareholders if the total value created from the deal exceeds the bid premium, the costs of doing the deal, and the transfers to parties other than both firms’ shareholders. There are two possible motives for a deal: to achieve synergies and to reduce agency costs. Value can be created by efficiency increases. Mergers can result in operational improvements by creating economies of scale or scope. They can also create financial synergies by diversifying the business outside a company’s current industry group or by lowering the cost of capital. Acquiring a target with high levels of cash in the balance sheet can result in lower internal financing cost. External financing cost can also be lowered by acquiring a target having lower financial leverage and unused debt capacity (Ghosh and Jain, 2000). Ensuring tax benefits may be another financial synergy that can be achieved through mergers. Such benefits can occur, for example, when a firm with accrued net operating loss (NOL) carry-forwards acquires a profitable target: It can consolidate for tax purposes, accelerating the use of its NOLs. However, successful mergers are based on sound business logic, so tax benefits are more likely to offer reinforcing rather than key motives behind the transaction (Copeland, Weston, and Shastri, 2003). Finally, mergers can increase efficiencies by reducing agency costs. Agency costs arise from a separation of ownership and control (Jensen and Meckling, 1976). When managers have but a fraction of ownership, they may exploit the company by reducing their efficiency because it is the owners who bear the brunt of the costs.
Searching for value-enhancing acquirers
59
The stock market acts as a monitoring system for this phenomenon; the stock price of these companies will fall if managers fail to act in the interest of shareholders (Fama and Jensen, 1983). As a last resort, new owners can use a takeover to circumvent the agency problem by replacing the less effective management team of the target company with a more effective one from the acquiring company. Apart from the previously mentioned motives, some reasons for takeover transactions may be dubious. The hubris hypothesis, for example, suggests that managers seek to acquire firms for personal motives and not necessarily for the benefit of the shareholders of the acquiring firm (Roll, 1986). Overconfident managers often overestimate their own abilities and the potential synergies from a deal. Such managers are likely to overpay for a target. This typically occurs when the target is difficult to evaluate or when competition arises among multiple bidders. The nemesis of the hubristic manager is the “winner’s curse.” The winning bid is cursed because the bid exceeds the true value of the target. In addition, the desire for empire building often leads managers to pursue costly acquisitions. There is also another, darker side to the agency cost theory previously discussed. In this theory, an acquisition—rather than solving agency problems—creates those very problems as a result. In this case, managers are motivated to increase the size of their firms since compensation is usually linked to size. In fact, executive compensation is often implicated as the ultimate motive behind a corporate acquisition.
3.4
Characteristics of value-enhancing acquirers
To determine which factors are highly correlated with M&A success, we concentrated on the two extreme groups (quintiles), examining both short- and long-term performance of the acquirer. We measured the average value of a factor (say market capitalization) for all acquirers in the top-performing group and compared it to the average for the companies in the bottom-performing group (quintile). A t-test is then employed to evaluate whether the difference in the averages is statistically significant. When the factor in question is binary (e.g., cash- or equity-financed deal), we simply compute the performance of the two different categories of the deals separately.
3.4.1
Factors that matter
Short-term market reaction The first and most obvious factor we could use to explain the long-term performance of acquirers is the immediate, short-term market reaction to the bid announcement. It is suggested that long-term negative excess returns may be explained by a slow adjustment to the acquisition announcement. If the market is efficient, an acquirer that outperforms significantly after the bid is announced might enjoy better long-term prospects than does a bidder who is heavily penalized by the market in the short term (Agrawal, Jaffe, and Mandelker, 1992).
60
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
To test for this phenomenon, we ranked all the bidders in our sample by their short-term market excess returns. We used a window of 10 days before to 10 days after the acquisition announcement to measure the short-term market reaction. We then concentrated on the extremes. Those companies that outperformed by more than 7% were placed in the top 20% while those with market relative returns lower than ⫺6% were placed in the bottom quintile. Figure 3.8 shows the longterm performance of these two different groups of acquirers starting from 10 days following the deal announcement. Our results show that acquirers that do well in the short term still underperform, but to a lesser extent. Their cumulative average excess return is just ⫺8.9%, which is almost half the amount of underperformance the group with the worst initial market reaction is suffering. These acquirers that struggle just after the deal is announced continue to underperform the market by 16%, bringing their total underperformance to ⫺6%. We have also calculated the average short-term CAAR of the best and worst performing long-term acquirers. We found that value-enhancing bidders enjoy a more favorable market reaction (⫹2.5%) when the deal is announced in contrast to bidders that struggle in the long run (⫺1.2%). Both of these tests confirm that the market, together with the initial response of investors, is a very good leading indicator for the long-term prospects of the acquirer.
Method of payment The financing structure is one of the most important discriminating factors between value-enhancing and value-destroying acquisitions. In our database we
Average market relative return (%)
2 0 −2 −4
Most favourable shortterm market reaction
−6 −8 −10 −12 −14
Least favourable shortterm market reaction
−16 −18 −20 −10 Days
Yr 1
Yr 2
Number of years post merger announcement
Figure 3.8 Long-term performance of acquirers with different short-term market reaction to the deal announcement. Source: Exshare, Thomson SDC and Citigroup Investment Research.
Yr 3
Searching for value-enhancing acquirers
61
obtained some details on the way the deal was financed. We split our sample into those deals that were financed exclusively by cash (65% of our sample) and those that were not. The latter category would involve equity financing or a combination of equity and cash. Figures 3.9 and 3.10 show the cumulative short- and long-term performance of acquirers around the deal announcement date. The results suggest that acquirers 0.9 Average market relative returns (%)
0.8
Cash deals
0.7 0.6 0.5 0.4
Non-cash mixed deals
0.3 0.2 0.1 0.0 −0.1
−10 −9 −8 −7 −6 −5 −4 −3 −2 −1 0 1 2 3 4 5 6 7 No. of trading days around merger announcement
8
9 10
Figure 3.9 Short-term performance of acquirers paying by cash. Source: Exshare, Thomson SDC and Citigroup Investment Research.
Average market relative returns (%)
5 0 −5 Cash deals
−10 −15 −20 −25 −30 −35 −10 days
Mixed deals Yr 1
Yr 2
No. of years post merger announcement
Figure 3.10 Long-term performance of acquirers paying by cash. Source: Exshare, Thomson SDC and Citigroup Investment Research.
Yr 3
62
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
that pay cash enjoy a much better performance. Cash-financed acquisitions result in a CAAR of 80 basis points for 20 days around the announcement date. Those bidders, however, that finance the transaction with equity or a combination of cash and equity outperform by just 30 basis points. Interestingly, the pre-announcement performance of bidders is similar for the two types of deals. But when the financing details are publicized, a few trading days after the initial announcement, the non-cash bidders start underperforming. The long-term returns also reveal staggering differences between the two different types of deals. A typical cash bidder underperforms by 6% 3 years after the deal announcement. However, firms that do not use cash exclusively to finance the deal do significantly worse. Their cumulative average abnormal return is ⫺28%. So what is behind this difference in performance? An equity-financed deal is associated with share price weakness because it tends to signal that the acquirer’s shares are overvalued and that the buyer is not confident about synergy gains. In contrast, a cash- or debt-financed offer tends to send a positive signal to the market about the buyer’s confidence in its ability to reload its cash balance. Those cash offers that involve a debt issuance can provide a significant additional incentive to make the merger work in order to realize synergy gains quickly.
Size Target companies are typically smaller than bidding firms. An acquisition of a very small company relative to the size of the bidder is, however, unlikely to create the efficiency increases and synergies that are usually desired with a merger. As a proxy for size we use four factors: number of employees, total assets, sales, and market capitalization. We computed the ratio of the target size over the acquirer across all four factors. We then concentrated on the extreme performers (top and bottom 20%) and computed the average relative size for these two groups. Next we tested whether there is a statistically significant difference in these averages between the valueenhancing (top 20% best performers) and value-destroying (bottom 20%) acquirers. We evaluated the success of the deal by looking at the short- and the long-term performance of the bidders separately. Figures 3.11 and 3.12 show the t-statistics that we calculated from these tests for the short-term and the long-term analysis, respectively. A t-statistic higher than 1.96 is considered an indication of statistical significance at the 95% confidence level. A positive and significant value implies that value-enhancing acquirers bid for relatively larger targets compared to value-destroying acquirers. Our analysis suggests that relative size is not an important discriminant for the short-term performance of bidders. Of the relative size factors we tested, none was significant. Relative size can, however, be important for the long-term success of the deal. We find that the best performing acquirers bid for targets that are, on average, 30% their market value. The acquiring group with the worst performance buys firms that are just 20% of their size. The t-statistic on
Searching for value-enhancing acquirers
63
−1.31
Market cap
−1.19
Sales
0.39
Assets
−1.58
−3
−2
Number of employees
−1
0
1
2
3
T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.11 Short-term analysis: difference in relative size (target/bidder) between value-enhancing and value-destroying acquirers. Source: Worldscope and Citigroup Investment Research.
2.31
Market cap
1.11
Sales
0.31
Assets
1.63
−3
−2
−1
0
1
2
Number of employees
3
T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.12 Long-term analysis: difference in relative size (target/bidder) between value-enhancing and value-destroying acquirers. Source: Worldscope and Citigroup Investment Research.
64
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
the market capitalization factor is 2.31, which is statistically significant. The rest of the factors, however, did not confirm that relative size can make a difference. Relative assets and sales were not relevant and number of employees was only marginally significant. We have also tested size factors not in relative but absolute terms2. We find that best performing acquirers (both in the short and long term) have significantly lower market capitalization than do those that underperform. In short, we find that relative size is not an important driver for the short-term performance of bidders. Firms that purchase relatively large targets have more chances to realize meaningful synergies and outperform in the long run than do those that buy smaller targets. The size of the acquiring company is also an important factor. Smaller acquirers have greater capacity to grow and increase their market power than do very big and already well established companies.
Diversified vs. nondiversified mergers (operational synergy) A focused or nondiversified deal is one in which both buyer and target are in the same industry (we use the MSCI GICS industry group classification). Diversified mergers are typically more complicated. It is often more difficult to realize synergies when buying a company from a completely different industry or area of the market. The empirical data confirm this theory. We find that diversifying mergers destroy more value than do focused ones. The short-term average market relative return of acquirers in nondiversified transactions is ⫺0.2%, almost 1% higher than in the case of diversifying deals (see Figure 3.13). In the long term acquirers that buy companies in the same industry still underperform but to a lesser extent than those that diversify (see Figure 3.14). Our findings suggest that firms should focus in their industries and specific areas of expertise rather than diversify.
Valuation Acquirer’s valuation We cannot ignore valuation when evaluating the future prospects of an acquirer. Theory suggests that the management of firms that trade at stretched valuation multiples tends to be overconfident in their abilities to generate synergies from a merger. They therefore tend to overpay for acquisitions (hubris hypothesis). We compared the pre-acquisition valuation multiples between the value-enhancing and value-destroying bidders. We used the following ratios: book-to-price, earnings yield (trailing and 12-month forward) and dividend yield. All multiples are expressed in sector-relative form. Once again, we performed separate tests for the short- and the long-term performance. Our results confirm that the best performing acquirers (in both the short and long term) trade at more attractive valuation multiples. Acquirers that are 2
We standardize the size factors by subtracting the cross-sectional average and dividing with the standard deviation.
Searching for value-enhancing acquirers
65
Average market relative returns (%)
0.4 0.2 0.0 −0.2 Focused deals −0.4 −0.6 −0.8 −1.0 −1.2
Diversified deals −10 −9 −8 −7 −6 −5 −4 −3 −2 −1 0 1 2 3 4 5 6 7 8 No. of trading days around merger announcement
9
10
Figure 3.13 Short-term performance in diversified and focused deals. Source: MSCI, Exshare, Thomson SDC and Citigroup Investment Research.
Average market relative returns (%)
5 0 −5 −10 Focused deals −15 −20 Diversified deals −25 −10 days
Yr 1 Yr 2 No. of years post merger announcement
Yr 3
Figure 3.14 Long-term performance in diversified and focused deals. Source: MSCI, Exshare, Thomson SDC and Citigroup Investment Research.
rewarded by the market are those that have been trading at a significant discount to their sector. Looking at P/E ratios (12-months forward), we find that the median discount to the sector average for the best performing group (long-term) is 16%. The worst performing group, on the other hand, has a P/E ratio similar to the sector average. We reach similar conclusions when looking at other valuation multiples (except dividend yield).
66
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Figures 3.15 and 3.16 plot the t-statistics associated with these valuation multiples. Positive readings with values above 2 on most valuation multiples in both the short- and the long-term analysis confirm that value bidders perform better than firms trading at more demanding multiples. So the valuation of the
2.06
Book to price act
Earnings yield (Last reported)
2.54
2.76 Earnings yield (12 months forward)
0.16
Div yield
0 0.5 1 1.5 2 2.5 3 −1 −0.5 T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.15 Difference in sector relative valuation between value-enhancing and value-destroying acquirers: short-term analysis. Source: Worldscope, Exshare, Thomson SDC and Citigroup Investment Research.
3.37
Book to price act
Earnings yield (Last reported)
5.39
6.76
1.27
Earnings yield (12 months forward)
Div yield
0 1 2 3 4 5 6 7 −1 T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.16 Difference in sector relative valuation between value enhancing and value destroying acquirers: long-term analysis. Source: Worldscope, Exshare, Thomson SDC and Citigroup Investment Research.
Searching for value-enhancing acquirers
67
bidder seems to be a very important factor, serving as a screen for outperforming acquirers both in the short and the long term after the deal is announced.
Target’s valuation We also tested the relevance of the pre-acquisition valuation of the target. We again used sector relative earnings, book value, and dividend multiples. The results are less compelling in this case. The only factor that appears significant in both the short- and the long-term analysis is dividend yield. It seems that the bidders that perform better are those that buy higher yielding targets (relative to their sector). We also compare the target company’s valuation to that of the acquiring firm’s. A popular argument to justify a deal is that the transaction in does not dilute earnings. To test the relevance of EPS impact (dilutive vs. accretive deals), we compare the last 12-months’ P/E ratio of targets to that of acquirers. We found that the best performing group of acquirers (short-term) was trading at a 10% median discount to the target before the deal announcement. The worst performing group of acquirers was trading at a discount of 2%. This was not a statistically significant difference. We find that EPS accretion or dilution is also irrelevant for the long-term track record of acquirers. The best performing acquirers trade at a 8% median discount to the target while the worst performing ones traded at a discount of 4%. This finding contrasts sharply with a popular argument often used to defend or oppose a takeover transaction. This suggests that acquisitions are good if their earnings are accretive and bad if their earnings are dilutive. This is a simplistic and dangerous approach as it is economically unfounded (Lee, 2006). As with any investment, a company creates value only if it is investing at a rate of return greater than the cost of capital. It is obvious that any statement that focuses only on EPS growth does not take into account risk and return and as such it cannot be an indication of a value- enhancing or a value-destroying acquisition.
Bid premium Another important factor that may affect the bidder’s returns is the bid premium. Value-enhancing bidders do not overpay so the bid premium is relatively small. The performance of the target around the bid announcement can therefore be an important leading indicator for the success of the acquirer. We estimate the bid premium by calculating the market-relative performance of the target from 30 trading days prior to and up to the day of the deal completion. We than rank all transactions based on the bid premium and put them into five groups. We again concentrate on the extremes. We focus on the deals where the bid premium was in the bottom 20% (lower than 9%) and the top 20% (higher than 56%) of our sample. We then calculate the performance of the acquirers in these two groups from 10 days before to 3 years after the deal announcement date. Figures 3.17 and 3.18 plot the short- and long-term performance of acquirers that pay the highest and the lowest bid premium.
68
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Average market relative returns (%)
1.5 1.0 0.5 Low bid premium 0.0 −0.5 −1.0 −1.5 −2.0 −2.5 −3.0
High bid premium −10 −9 −8 −7 −6 −5 −4 −3 −2 −1 0
1
2
3
4
5
6
7
8
10
No. of trading days around merger announcement
Figure 3.17 Short-term performance of acquirers that pay the highest and lowest premium. Source: Exshare, Thomson SDC and Citigroup Investment Research.
Average market relative returns (%)
4 2 0 −2
Low bid premium
−4 −6 −8 −10 −12 −14 High bid premium
−16 −18
−10
Yr 1 Yr 2 No. of years post merger announcement
Yr 3
Figure 3.18 Long-term performance of acquirers that pay the highest and lowest premium. Source: Exshare, Thomson SDC and Citigroup Investment Research.
The results suggest that acquirers that overpay underperform. Acquirers that pay the highest bid premium are penalized by the market. Their shares underperform the broad market index by 2% and 15% in 10 days and 3 years after the deal announcement. When the target performance, however, is more modest
Searching for value-enhancing acquirers
69
(bid premium below 9%) the prospects of acquirers is significantly better. They tend to outperform by 51 basis points in the short term and fall below the market index by 8.7% in the long term.
Multiple bidders The number of bidders publicly competing for the same target can also be important. Bidders do not know the true value of the firm they wish to acquire. The intense competition among them is therefore likely to lead to an overpayment (winner’s curse) as management overestimates the importance of and the synergies from this transaction. In our sample we have data on 4645 deals with just one bidder and 220 transactions with multiple bidders. Figures 3.19 and 3.20 plot the shortand long-term performance of acquirers in the case of one and multiple bidders. We find that the firms that win a bidding contest typically perform worse in the short term (⫺0.44%) than the rest of the acquirers (⫹0.60%). In the long term, however, despite the initial underperformance, those firms generate returns similar to those of the average acquirer. Interestingly their share price troughs two years after the deal announcement, then recovers in the third. This could reflect the fact that the winners of the bidding contest may be right to persist and pursue the transaction.
3.4.2
The factors that do not matter
Apart from all of the tests previously discussed, we have also tested a number of other factors; we did not find, however, that they could be useful to screen
Average market relative returns (%)
1.0 0.8 One bidder
0.6 0.4 0.2 0.0 −0.2 −0.4 −0.6 −0.8
Multiple bidder −10 −9 −8 −7 −6 −5 −4
−2 −1 0 31
2
3
4
5
6
7
8
9
No. of trading days around merger announcement
Figure 3.19 Short-term performance of acquirers in the case of one and multiple bidders. Source: Exshare, Thomson SDC and Citigroup Investment Research.
10
70
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Average market relative returns (%)
2 0 −2 −4 −6 −8
One bidder
−10 −12 −14 −16 −18 −10 days
Multiple bidder Yr 1 Yr 2 No. of years post merger announcement
Yr 3
Figure 3.20 Long-term performance of acquirers in the case of one and multiple bidders. Source: Exshare, Thomson SDC and Citigroup Investment Research.
for value-enhancing acquirers. We nevertheless believe that it is important to publish our findings.
Debt capacity Theory suggests that acquirers that are able to use gearing to finance a deal should perform better. The balance sheet structure of an acquirer may be a good leading indicator for the post-deal performance of the combined firm. After all, low levels of debt indicate good prospects for added debt capacity that could be used to finance a takeover bid. To test this proposition we computed the difference in the average short- as well as long-term debt-to-asset ratio between the best and worst performing acquirers.3 We have also looked at whether the leverage of the target, in absolute terms as well as relative to the bidder, could be a key driver of stock market reaction (see Figures 3.21 and 3.22). Our results suggest that leverage is not a major driver of how the equity market reacts to the deal in either the short or the long term.
Management efficiency/operating performance Efficiency improvements can be derived from combining firms with unequal managerial capabilities. A bidder that is run by a more efficient management team can seek to acquire an underperforming target. But also a bidder that has been underperforming can try to turn around the business by buying a very well run 3
We used absolute as well as sector-relative gearing ratios.
Searching for value-enhancing acquirers
71
Total debt/Total capital
1.07
ST debt/ST assets
1.85
Total debt/Total assets
0.04
−3
0 1 2 3 −2 −1 T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.21 Difference in sector-relative debt ratios between value-enhancing and value-destroying acquirers: short-term analysis Source: Worldscope, Exshare, Thomson SDC and Citigroup Investment Research.
1.09
Total debt/Total capital
ST debt/ST assets
Total debt/Total assets
−0.52
0.98
0 1 2 3 −3 −2 −1 T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.22 Difference in sector-relative debt ratios between value-enhancing and value-destroying acquirers: long-term analysis. Source: Worldscope, Exshare, Thomson SDC and Citigroup Investment Research.
and efficient company. We tested whether relative profitability can be an important discriminating factor between value-enhancing and value-destroying deals. We have used last-reported return on equity and margins as well as the 5-year averages in the pre-acquisition period. We concentrated on relative profitability
72
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
ratios by comparing the respective values of the target to that of the bidder. None of the profitability factors we tested was significant either in the short- or the long-run tests. Neither relative operating performance nor profitability seem to be relevant for determining the success of a merger. Our results suggest that leverage is not a major driver of how the equity market reacts to the deal in either the short or the long term. (See Figures 3.23 and 3.24.)
Growth Rapidly growing companies are sometimes the best takeover targets. A firm whose organic growth is slowing might choose to expand through acquisitions and decide to buy a fast growing firm. We tested whether absolute or relative growth can be useful leading indicators for merger success. We used growth rates in earnings and sales for 1, 3, and 5 years prior to the acquisition date. (See Figures 3.25 and 3.26.) Our empirical results did not confirm that growth could be important. We find that the best performing acquirers did not have significantly different earnings growth profiles from the worst performers. Interestingly, acquirers that do well in the long term have significantly lower pre-merger sales growth compared to those that underperform. We find the exact opposite result, however, when evaluating the importance of sales growth for the short-term performance of acquirers. We have also compared the historic growth rates of the target to that of the bidder and used the ratio as a leading indicator for determining merger success.
−0.86
ROE
−0.54
ROE 5yr
0.46
0.35
Op margin 5 yr
Op margin
−3 −2 −1 0 1 2 3 T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.23 Difference in relative profitability ratios between value-enhancing and value-destroying acquirers: short-term analysis. Source: Worldscope, Exshare, Thomson SDC and Citigroup Investment Research.
Searching for value-enhancing acquirers
73
0.81
ROE
0.95
ROE 5 yr
0.82
Op margin 5 yr
0.49
−3
Op margin
0 1 2 3 −2 −1 T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.24 Difference in relative profitability ratios between value-enhancing and value destroying acquirers: long-term analysis. Source: Worldscope, Exshare, Thomson SDC and Citigroup Investment Research.
Sales growth 3 yr
0.53
Sales growth 2 yr
1.12 −0.57
Sales growth 1 yr −1.23
EPS growth 3 yr
−0.25
EPS growth 2 yr
−0.37
EPS growth 1 yr −3
−2
−1
0
1
2
3
T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.25 Difference in growth between value-enhancing and value-destroying acquirers: short-term analysis Source: Worldscope, Exshare, Thomson SDC and Citigroup Investment Research.
74
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Sales Growth 3 yr
−1.31
Sales Growth 2 yr
−1.23
Sales Growth 1 yr
−4.89 −1.62
EPS Growth 3 yr
−0.57
EPS Growth 2 yr EPS Growth 1 yr −6
0.74 0 1 2 3 −5 −4 −3 −2 −1 T-Statistics from a means test between value enhancing and value destroying acquirers
Figure 3.26 Difference in growth between value-enhancing and value-destroying acquirers: long-term analysis. Source: Worldscope, Exshare, Thomson SDC and Citigroup Investment Research.
The results showed very little correlation between relative growth and the equity market reaction to the deal in either in the short or the long term.
Same-country vs. cross-border deals Finally, we have looked at whether cross-border vs. domestic deals can make a difference in performance. Overseas acquisition are often seen as a way to help companies reach a broader geographic market, gain access to local technological expertise, and provide a lower cost production platform. We have 2126 samecountry and 2740 cross-border pan-European deals in our database. Figures 3.27 and 3.28 show the cumulative short- and long-term performance of samecountry and cross-border acquirers around the deal announcement date. The immediate market reaction to cross-border deals is slightly more favorable, with bidders outperforming by 70 basis points, 20 basis points more than the bidders in domestic deals. The long-term performance, however, is very similar between the two groups. We suspect that the potentially more complex nature of cultural, social, and post-merger integration issues may counterbalance the synergies gained from cross-border mergers.
3.5
Putting it all together
To understand which combination of factors is more important in driving the returns of acquirers and what should be the optimal weighting among these
Searching for value-enhancing acquirers
75
Average market relative returns (%)
1.0 Cross border deal 0.8
0.6
0.4 Same border deal 0.2
0.0 −0.2
−10 −9 −8 −7 −6 −5 −4 −3 −2 −1 0
1
2 3
4 5 6
7
8
9
10
No. of trading days around merger announcement
Figure 3.27 Market relative performance of acquirers in domestic and foreign deals: short-term analysis. Source: Exshare, Thomson SDC and Citigroup Investment Research.
Average market relative returns (%)
2 0 −2 −4 −6 −8
Cross border deal
−10 Same border deal −12 −14 −16 −10 Days
Yr 1 Yr 2 No. of years post merger announcement
Yr 3
Figure 3.28 Market relative performance of acquirers in domestic and foreign deals: long-term analysis. Source: Exshare, Thomson SDC and Citigroup Investment Research.
76
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
drivers, we performed a regression analysis. We decided to concentrate only on the long-term performance. The short-term returns are more volatile and difficult to predict. We also think that the majority of investors take a more longterm view when analyzing the prospects of an acquirer. The factors that we used in our model to forecast the 3-year market relative performance4 of acquirers are as follows: ●
●
●
●
●
●
Method of payment (a binary factor that takes the value of 1 if the deal is financed exclusively by cash and 0 otherwise) Relative size (the market capitalization of target relative to the bidder—standardized) Acquirer’s valuation (the acquirers’ sector-relative earnings yield) Bid premium (the target’s market-relative return from 30 days prior to the day of the bid announcement) Short-term Market Reaction (the bidder’s market relative return for 20 days around the bid announcement) Diversification (a binary factor that takes the value of 1 if the deal is for a target in the same industry and 0 otherwise)
The choice of these factors reflects the results of the descriptive/univariate analysis we did in the previous section. Table 3.3 shows the estimated coefficients and the associated t-statistics from these regressions. t-Statistics higher than 1.96 indicate statistical significance at the 95% level. All factors were found to be statistically significant. The goodness of fit from the regression is also quite satisfactory. The six factors we used are capable of explaining about 6% of the return variability. The intercept from the regression is negative, confirming that the average acquirer underperforms the market. The long-term returns of acquirers, however, are positively associated with the shortterm market reaction, the relative size of the target, and the valuation attractiveness of the bidder. We also confirm that cash-financed and focused deals can be very important factors in explaining the acquirers’ returns. Finally, we find a negative relationship between the bid premium and the performance of the bidder, Table 3.3 Regression results from long-term bidders return model
Intercept Short-term market reaction Method of payment Relative size Diversification Acquirer’s valuation Bid premium
Coefficient
t-Statistics
p-Value
⫺0.17 0.34 0.08 0.04 0.09 0.07 ⫺0.13
⫺3.40 2.11 2.65 3.47 2.87 3.62 ⫺3.28
0.00 0.04 0.01 0.00 0.00 0.00 0.00
Source: Citigroup Investment Research 4
We used absolute as well as sector-relative gearing ratios to measure the long-term market relative returns of bidders starting from 10 days after the deal announcement to 3 years after, thereby avoiding the overlapping observations with the short-term market reaction factor.
Searching for value-enhancing acquirers
77
suggesting that firms that overpay tend to struggle to create value for their shareholders in the long term. Our analysis suggests that the most important factors in driving bidders’ returns are the short-term market reaction, the relative size of the target, the bidder’s valuation, and the deal premium.
3.6
Conclusion
Everybody wants to know the next takeover target these days. But hefty returns can also be realized by investing in bidders. Not in all of them, though. Analyzing more than 4000 deals since 1990, we found that the average acquirer underperforms in the long term. Although the average return is negative, the dispersion in their performance is significant. Some acquirers perform extremely well and should remain in investors’ portfolios. In this chapter we try to identify common characteristics among value-enhancing acquirers and attempt to put together a framework that could help investors screen for outperforming bidders. Our work suggests that the short-term market reaction to the deal, together with the bidder’s share price movement in the short term, can be a good leading indicator for the long-term prospects of the transaction. The financing structure of the deal also matters. Deals that are financed exclusively by cash do significantly better. An equity-financed deal is associated with share price weakness because it is often perceived as a signal that the acquirer’s shares are overvalued and that the buyer is not confident about synergy gains. Size also matters. An acquisition of a very small company relative to the size of the bidder is unlikely to create the efficiency increases and synergies that are usually expected with a merger. The best performing acquirers bid for targets that are, on average, 30% of their market value. Focused deals that concentrate on the same industry also enjoy better fortunes than diversifying deals. Valuation is very important. The management of firms that trade at stretched valuation multiples tends to be overconfident of their abilities to generate synergies from a merger. They therefore usually overpay for acquisitions (hubris hypothesis). Acquirers that trade at a discount to their sector do significantly better. Deals in which the bid premium is relatively small also tend to create value. Finally, bidding wars often lead to overpayment and value destruction (winner’s curse). So these are the factors that matter. There are many more that are not significant. Our results suggest that a P/E differential between the bidder and target is not an important factor to screen for value-enhancing acquirers. EPS accretion or dilution is irrelevant for the long-term track record of acquirers. As with any investment, a company creates value only if it is investing at a rate of return greater than the cost of capital. It is obvious that any statement that focuses only on EPS multiples does not take into account risk and return, and hence cannot be an indication of a value-enhancing or a value-destroying acquisition. Leverage is also not a major driver of how the equity market reacts to the deal in either the short or the long term. We have also looked at historic growth
78
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
rates as a leading indicator for determining merger success. The results showed very little correlation between both absolute or relative growth with the equity market reaction to the deal. Finally, relative profitability is not a good discriminating factor between value-enhancing and value-destroying deals either.
References Agrawal, A., Jaffe, J. F., and Mandelker, G. N. (1992). The Post-Merger Performance of Acquiring Firms: A Re-examination of an Anomaly. Journal of Finance, 47:1605–1621. Copeland, T., Weston, J. F., and. Shastri, K (2003). Financial Theory and Corporate Policy, 4th Ed., New York: Reiters. Fama, E., and Jensen, M. (1983). Separation of Ownership and Control. Journal of Law and Economics, 26:301–325. Ghosh, A., and Jain, P. C. (2000). Financial Leverage Changes Associated with Corporate Mergers. Journal of Finance, 50:377–402. Jensen, M. C., and Meckling, M. (1976). Theory of the Firm—Managerial Behaviour, Agency Costs and Ownership Structure. Journal of Financial Economics 3, 305–360. Lee, K. (2006). The Value Driver: Analysing M&A Transactions—The Common Pitfalls, Issue 17. Liodakis, M., and Brar, G. (2005). Searching for Alpha: Spot the Takeover Target, European Quantitative Strategy, Citigroup Investment Research. Loughram, T., and Vijh, A. (1997). Do Long-Term Shareholders Benefit from Corporate Acquisitions? Journal of Finance, 52:1765–1790. Roll, R. (1986). The Hubris Hypothesis of Corporate Takeovers. Journal of Business, 59:197–216.
4 The long-term operating performance in European mergers and acquisitions Marina Martynova, Sjoerd Oosting and Luc Renneboog
Abstract We investigate the long-term operating performance of 155 European corporate mergers and acquisitions completed between 1997 and 2001. Acquiring companies and target companies are from continental European countries and the United Kingdom. Matching with peer companies enables us to adjust for industry, size, and pre-event performance. The first main result is that after adjusting for industry, size, and pre-acquisition, the combined operating performance of the acquirer and target does not change significantly following mergers, whereas their unadjusted “raw” operating performance declines significantly. Second, we find that the acquirer and the target significantly outperform their industry median peers prior to the merger. Third, we find that the post-acquisition performance of the combined firm varies significantly across M&As with different characteristics: hostile versus friendly bids, tender offers versus negotiated deals, and domestic versus cross-border transactions. Furthermore, cash reserves of the acquiring firm prior to the bid and the relative size of the target firm are important determinants of the post-acquisition profitability.
4.1
Introduction
During the last decade, mergers and acquisitions (M&As) involving European companies have occurred in unprecedented numbers. In 1999, the total value of the intra-European M&A activity has peaked at a record level of USD 1.4 trillion (Thomson Financial Securities Data) and for the first time became as sizable as M&A activity in the U.S. Despite these developments, empirical research on M&A activity remains mostly confined to the United Kingdom and the United States and little is known about the effect of continental European takeovers on the operating performance of bidding and target firms. In this chapter, we investigate whether and to what extend European companies improve their profitability subsequent to the completion of takeover transactions. The research question is appealing for the following three reasons. First, empirical evidence on the post-acquisition performance of European firms is virtually nonexistent. To our best knowledge, the literature in this field comprises only two studies: Mueller (1980) and Gugler, Mueller, Yurtoglu, and Zulehner (2003).
80
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
This study contributes to the existing literature by providing new evidence on the long-term performance of the most recent European M&As and by checking the robustness of the results using an up-to-date methodology of measuring improvement or deterioration in post-merger operating performance. Second, even for the United States, research on the improvement of post-merger operating performance is rather limited and its conclusions are often contradictory. Whereas some studies document a significant improvement in operating performance following acquisitions (Healy, Palepu, and Ruback, 1992; Heron and Lie, 2002; Rahman and Limmack, 2004), others reveal a significant decline in post-acquisition operating performance (Kruse, Park, Park, and Suzuki, 2002; Yeh and Hoshino, 2001; Clark and Ofek, 1994). Furthermore, a number of studies demonstrate insignificant changes in post-merger operating performance (Ghosh, 2001; Moeller and Schlingemann, 2005; Sharma and Ho, 2002). A third reason for this study is that we intend to investigate the determinants of the changes in post-merger profitability of bidding and target firms. In particular, we test whether characteristics of the M&A transaction, such as the means of payment, deal hostility, and industry relatedness, have an impact on the long-term performance of the firms. Our analysis is based on a sample of 155 European mergers and acquisitions, completed between 1997 and 2001. We employ four different measures of operating performance: EBITDA and EBITDA corrected for changes in working capital, each scaled by the book value of assets and by sales. Our results can be summarized as follows. First, we find that the post-merger profitability of the combined firm is not significantly different from the aggregate performance of the bidding and target firms prior to the merger. This demonstrates not only that corporate takeovers cannot engender substantial augmentations in operating performance—as is often claimed by the merged company—but also that mergers and acquisitions do not generate poor performance, as was often claimed in earlier academic research. Still, we find that the post-acquisition performance of the combined firm varies significantly across M&As with different characteristics: hostile versus friendly bids, tender offers versus negotiated deals, and domestic versus cross-border transactions. Furthermore, cash reserves of the acquiring firm prior to the bid and the relative size of the target firm are important determinants of the post-acquisition profitability. However, consistent with previous U.S. studies, we find no differences in operating performance of industry-related and diversifying takeovers and deals that involve different means of payment. The outline of this chapter is as follows. Section 4.2 provides an overview of the prior studies on post-acquisition performance. Section 4.3 describes our sample selection procedure and the methodology we used to measure changes in operating performance. The characteristics of our final sample are also given in section 4.3. Section 4.4 presents the main results of our analysis regarding changes in the operating performance of bidding and target firms subsequent to takeovers. Section 4.5 investigates the determinants of the post-acquisition performance. Section 4.6 summarizes the results and offers conclusions.
The long-term operating performance in European mergers and acquisitions
4.2 4.2.1
81
Prior research Post-acquisition performance
Previous empirical studies yield inconsistent results about changes in operating performance following corporate acquisitions. The extant empirical studies can be evenly divided into three groups: studies that report a significant improvement in the post-acquisition performance, those that document a significant deterioration, and those that find insignificant changes in performance. Table 4.1 provides an overview of these studies. The most recent U.S. studies that employ more sophisticated techniques to measure changes in the post-merger performance tend to show that the profitability of the bidding and target firms remain unchanged (Moeller and Schlingemann, 2005; Ghosh, 2001) or significantly improves after the takeover (Heron and Lie, 2002; Linn and Switzer, 2001). The conclusions of U.K. studies are more contradictory; for example, Dickerson, Gibson, and Tsakalotos (1997) find a significant decline in post-acquisition performance, whereas Powell and Stark (2005) show a significant growth. Like the U.K. studies, Asian studies also yield contradictory results. Evidence suggests that Japanese M&As incur a decrease in post-acquisition operating performance of the merged firm (Kruse, Park, Park, and Suzuki, 2002; Yeh and Hoshino, 2001), Malaysian takeovers are associated with an increase in the post-acquisition performance (Rahman and Limmack, 2004), while Australian M&As lead to insignificant changes in the profitability of bidding and target firms after the takeover (Sharma and Ho, 2002). For continental Europe, Gugler, Mueller, Yurtoglu, and Zulehner (2003) document a significant decline in post-acquisition sales of the combined firm, but an insignificant increase in post-acquisition profit.
4.2.2
The determinants of the post-acquisition performance
Method of payment: cash versus stock Empirical evidence suggests that the means of payment is an important determinant of the long-term post-acquisition performance: cash offers are associated with stronger improvements than are takeovers involving other forms of payment (Linn and Switzer, 2001; Ghosh, 2001; Moeller and Schlingemann, 2005). A possible explanation is that cash deals are more likely to effect replacement of (underperforming) target management, conceivably improving performance (Denis and Denis, 1995; Ghosh and Ruland, 1998; Parrino and Harris, 1999). An alternative explanation is that a cash payment is frequently financed with debt (Ghosh and Jain, 2000; Martynova and Renneboog, 2006). Debt financing restricts the availability of corporate funds at the managers’ disposal, minimizing the scope for free cash flow problems (Jensen and Meckling, 1976). Following these arguments, one can conclude that cash-financed takeovers are more likely to bring out more managerial discipline. However, nothing in the empirical
Table 4.1 Sample selection procedure Author(s)
Sample period
Market
Sample Performance size measure
Scaled by
Matched by
Studies that document an improvement in post-acquisition operating performance (1) Industry Powell and 1985–1993 U.K. 191 (1) Pre-tax CF (1) MV Stark (2005) (2) Pure CF assets (2) Industry, (incl. changes (2) Adj. MV size, and in WC) assets pre-event (3) BV assets performance (4) Sales
Pre-merger Change Conclusion performance (C) or intercept (I) Model A and T
C⫹I
Median postacquisition operating perform ance increases
A and T
C⫹I
Operating cash flow performance improves
Rahman and 1988–1992 Malaysia 113 Limmack (2004)
Pure CF (incl. changes in WC)
BV assets
Industry and size
Heron and Lie (2002)
1985–1997 U.S.
859
Operating income
Sales
(1) Industry Only A (2) Industry and pre-event performance
C
Operating performance improves after M&As
Linn and Switzer (2001)
1967–1987 U.S.
413
Pre-tax CF
MV assets
Industry
A and T
C
Post-acquisition cash flow increases
Parrino and Harris (1999)
1982–1987 U.S.
197
Pre-tax CF
Adj. MV assets
Industry
None
Other
Post-acquisition operating performance improves
Switzer (1996)
1967–1987 U.S.
324
Pre-tax CF
MV assets
Industry
A and T
C⫹I
Median performance improves over 5 years following the acquisition
Healy, Palepu, 1979–1984 U.S. and Ruback (1992)
50
Pre-tax CF
Adj. MV assets
Industry
A and T
C⫹I
Post-merger operating cash flow returns increase
Moeller and 1985–1995 U.S. Schlingemann acquirers (2005)
2362*
Pre-tax CF
MV assets
Industry
Only A
I
Negative (but insignificant) change in cash flow performance after mergers. Crossborder acquirers underperform domestic acquirers.
Gugler, Mueller, Yurtoglu, and Zulehner (2003)
2753
(1) EBIT (2) Sales
No scaling
Industry
None
Other
Post-acquisition profits are higher than predicted (mostly significantly). Sales are lower than predicted (mostly significantly).
A and T
C⫹I
Insignificant change in post-acquisition performance.
Industry, size, A and T and Pre-event performance
C⫹I
No significant changes in operating performance following M&As
1981–1998 World
Studies that document no significant changes in post-acquisition operating performance Sharma and 1986–1991 Australia 36 Pure CF (incl. (1) BV assets Industry and Ho (2002) changes (2) BV equity Size in WC) (3) Sales (4) # shares Ghosh (2001) 1981–1995 World
315
Pre-tax CF
Adj. MV assets
(continued)
Table 4.1 (continued) Author(s)
Sample period
Market
Herman and 1975–1983 U.S. hostile Lowenstein offers (1988)
Sample size
Performance measure
56
(1) Net income (1) BV equity (2) EBIT (2) Capital
Belgium, Different Profit after tax Germany, per U.K., U.S., country France, Netherlands, Sweden
Scaled by
BV assets
Matched by
Pre-merger Change Conclusion performance (C) or intercept (I) Model
Unmatched
Only A
Industry and None size
C
Bidders’ return on capital (ROC) decreases; ROE increases.
Other
Belgium, Germany, U.K., and U.S.: an increase in postmerger profitability. France, Netherlands, and Sweden: a decline in profitability.
Mueller (1980)
1950s, 1960s, 1970s
Lev and Mandelker (1972)
1952–1963 U.S.
69
(1) Net income (1) BV assets Industry and A and T (2) Op. income (2) BV equity size (3) # of shares (4) Sales
C
NI/assets significantly increase for acquiring firms. Other performance measures exhibit no significant changes
Lev and Mandelker (1972)
1952–1963 U.S.
69
(1) Net income (1) BV assets Industry and A and T (2) Op. income (2) BV Equity size (3) # of shares (4) Sales
C
NI/assets significantly increase for acquiring firms. Other performance measures exhibit no significant changes
Studies that document a deterioration in post-acquisition operating performance Kruse, Park, 1969–1992 Japan 46 Pre-tax CF (1) Adj. MV Industry Park, and assets and size Suzuki (2) Sales (2002) Yeh and Hoshino (2001)
1970–1994 Japan
Dickerson, 1948–1977 U.K. Gibson, and Tsakalotos (1997) Clark and 1981–1988 U.S. Ofek (1994) distressed targets Meeks 1964–1972 U.K. (1977)
Hogarty (1970)
1953–1964 U.S.
A and T
C⫹I
(1) BV equity (2) BV assets
Industry
Only A
Other
1443** Pre-tax profits
Net assets
Industry
Only A
Other
38
EBITD
Sales
Industry
A and T
C
223
Pre-tax profits
Net assets
Industry
A and T
C
43
EPS and capital # of shares gains
Industry
Only A
Other
86
(1) Net income (2) Op. income
Overall decline in cash flow; however, mergers lead to a significant improvement in the performance Significant decline in ROA and ROE following a merger; however, only M&As that involve keiretsu are followed by a significant decline in ROE and ROA. M&As involving independent firms do not. A significant decline in acquirer’s ROA
Operating performance declines over 3 years following M&As Post-merger profitability is significantly lower than the pre-merger profitability. Investment performance of the acquirers (acquirer’s perspective) deteriorates after M&As
* While the total sample consists of 4,430 acquisitions, the regression model used to estimate changes in operating performance is based on 2,362 observations. ** More specifically, the sample includes 2,941 companies, of which 613 (21%) companies were involved in 1,443 acquisitions.
86
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
literature fails to find a significant relationship between the method of payment and post-merger operating performance (Healy, Palepu, and Ruback, 1992; Powell and Stark, 2005; Heron and Lie, 2002; Sharma and Ho, 2002).
Deal atmosphere: friendly versus hostile Hostility in corporate takeovers may also be associated with better long-term operating performance of the merged company. Hostile bids are more expensive for the bidding firms, such that only takeovers having high synergy potential are likely to succeed (Burkart and Panunzi, 2006). But, again, the empirical literature finds no evidence in support of this conjecture (Healy, Palepu, and Ruback, 1992; Ghosh, 2001; and Powell and Stark, 2005). In addition to the atmosphere of takeover bid the acquisition method (a tender offer or a negotiated deal) may also be an important determinant of the post-merger performance. However, the empirical evidence does not unveil this such relation either (Switzer, 1996; Linn and Switzer, 2001; Heron and Lie, 2002; Moeller and Schlingemann, 2005).
The acquirer’s leverage and cash reserves The activities of highly leveraged acquirers may be subject to severe monitoring by creditors such that unprofitable M&As would be effectively prevented ex-ante. Empirical evidence on this relationship is mixed. Whereas Ghosh and Jain (2000), Kang, Shivdasani, and Yamada (2000), and Harford (1999) provide evidence in line with the conjecture,1 Linn and Switzer (2001), Switzer (1996), and Clark and Ofek (1994) find no significant relation between acquirer’s leverage and post-merger operating performance. Acquirer’s cash reserves are also expected to affect post-merger operating performance.’ before the first sentence starting with ‘According to Jensen’s … According to Jensen’s (1986) free cash flow theory, acquirers with excessive cash holdings are more likely to make poor acquisitions and hence experience significant post-merger underperformance relative to their peers with limited cash holdings. Harford (1999) and Moeller and Schlingemann (2005) confirms this conjecture empirically.
Industry relatedness: focused versus diversifying acquisitions Although diversifying (or conglomerate) acquisitions are expected to create operational and/or financial synergies, the creation of diversified firms is also associated with a number of disadvantages such as rent-seeking behavior by divisional managers (Scharfstein and Stein, 2000), bargaining problems within the firm (Rajan et al., 2000), or bureaucratic rigidity (Shin and Stulz, 1998). These disadvantages 1
Ghosh and Jain (2000) find that an increase in financial leverage around M&As is significantly positively correlated with the announcement of abnormal stock returns. Kang, Palepu, and Ruback (2000) show for 154 Japanese mergers between 1977 and 1993 that the amount of bank debt is positively and significantly related to the acquirer abnormal returns. Harford (1999) shows that cash-rich firms experience negative stock price reactions following acquisition announcements, which is more negative when higher amounts of excess cash are involved.
The long-term operating performance in European mergers and acquisitions
87
of diversification may outweigh the alleged synergies and result in poor postmerger performance of the combined firm. Furthermore, diversifying M&As may be an outgrowth of the agency problems between managers and shareholders (Shleifer and Vishny, 1989), which is also likely to result in the deterioration of corporate performance after the takeover. While earlier studies confirm these conjectures (Healy, Palepu, and Ruback, 1992; Heron and Lie, 2002), later studies find the relationship between diversifying takeovers and poor post-merger performance insignificant (Powell and Stark, 2005; Linn and Switzer, 2001; Switzer, 1996; Sharma and Ho, 2002). Furthermore, Kruse, Park, Park, and Suzuki (2002) and Ghosh (2001) document that diversifying acquisitions significantly outperform their industry-related peers.
Relative size of the target Takeovers of relatively large targets are more likely to achieve sizeable operating and financial synergies and economies of scale than small acquisitions, thereby leading to stronger post-acquisition operating performance. However, the acquirer of a relatively large target may face difficulties in integrating the target firm, which could lead to a deterioration of performance. There is empirical evidence in support of both conjectures. Linn and Switzer (2001) and Switzer (1996) provide evidence that acquisitions of relatively large targets outperform those of small targets. Clark and Ofek (1994) document that, in large acquisitions, the difficulties with managing a large combined firm outweigh the operating and financial synergies, leading to deterioration of operating performance. However, most empirical evidence reports no significant relationship between the relative target size and post-merger performance (Powell and Stark, 2005; Moeller and Schlingemann, 2003; Heron and Lie, 2002; Sharma and Ho, 2002; Kruse, Park, Park, and Suzuki, 2002; Healy, Palepu, and Ruback, 1992).
Domestic versus cross-border deals In cross-border acquisitions, bidding and target firms are likely to benefit by taking advantage of imperfections in international capital, factor, and product markets, by internalizing the R&D capabilities of target companies (Eun et al., 1996) and by expanding their businesses into new markets (as a response to globalization trends). In this case, cross-border acquisitions are expected to outperform their domestic peers. However, regulatory and cultural differences between the bidder and target countries may lead to complications in managing the postmerger process, contributing to the failure to achieve the anticipated merger synergies. As a result of such difficulties in cross-border bids, the post-merger performance of the combined firm may deteriorate (Schoenberg, 1999). Moeller and Schlingemann (2003), Goergen and Renneboog (2004), and Martynova and Renneboog (2006b) show that that firms acquiring foreign targets experience significantly lower takeover announcement returns than their counterparts acquiring domestic targets. Gugler, Mueller, Yurtoglu, and Zulehner (2003) report a significant effect of cross-border deals on post-acquisition operating performance.
88
4.3 4.3.1
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Data and methodology Sample selection
Our sample of European acquisitions, which were completed between 1997 and 2001, is selected from the Mergers and Acquisitions Database of the Securities Data Company (SDC) and Zephyr.2 Only intra-European domestic and cross-border takeovers are included, in which both the acquirer and the target are from continental Europe or the United Kingdom. We retain the takeover deals in which at least one of the participants is a publicly traded company. We exclude from the sample deals in which the acquirer is the management or the employees, or the target is a subsidiary of another company. Furthermore, we exclude M&As in which either a bidder or a target or both are financial institutions (banks, savings banks, unit trusts, mutual funds, and pension funds). We also removed 15 takeovers in which the target was re-sold or the acquirer was acquired by a third party within 3 years after the deal completion. This selection results in 858 European M&A deals (see Table 4.2). We further require profit-and-loss accounts and balance sheet data to be available for acquirers and targets for at least 1 year prior and 1 year after the acquisition. Accounting data is collected from Amadeus Extended database.3 In our Table 4.2 Sample selection procedure Total number of completed deals (1997–2001) Removed deals Net number of deals Number of deals in which A and T have at least 1 year pre- and 1 year post-acquisition financials available in Amadeus Number of deals in which A and T have at least 3 years pre- and post-acquisition financials available
873 ⫺15a 858 (100%) 155 (18%) 81 (9%)
a
Fifteen deals were removed from the sample for the following reasons: target (or part of the target) was sold again within 1 year after the acquisition (8⫻), more than 1 acquirer was involved (2⫻), acquirer was taken over within 1 year after acquisition (2⫻), other (3⫻).
2
The reason for selecting deals that were launched and completed between 1997 and 2001 is that accounting data for European firms are available in the Amadeus database only from 1995 onward. For this study, we require that at least 2 years of pre-acquisition accounting data be available. Therefore, we were forced to restrict our sample only to M&As completed as of 1997. The upper-time bound of our sample comes from another restriction of the Amadeus database: the latest accounting data available in Amadeus refer to 2004. Thus, we were also forced to restrict our sample to M&As completed by 2001, as this allows us to analyze the post-merger operating performance over 3 years after the bid completion. 3 Amadeus Extended is an online database that contains information on 8,600,000 public and private companies in 38 European countries. Initially, we employed Amadeus Standard database which contains 250,000 public and private companies in Europe. However, we find that Amadeus Standard covers just 40% of bidding and target firms from our sample. Using Amadeus Extended, we find data for 73% of our acquirers and targets (624 M&As).
The long-term operating performance in European mergers and acquisitions
89
analysis, we focus on the year of the transaction’s completion, rather than the year of the announcement of the bid. Our sample includes a number of takeovers for which the year of announcement and year of completion do not coincide. For 89% of deals, a completion date is available. When a completion date is not available, we take the announcement year as a proxy for the completion year. Table 4.2 summarizes the overall sample selection procedure which results in the final sample of 155 deals.
4.3.2
Sample description
Our final sample of intra-European M&As comprises 155 deals (see Table 4.3) and includes 144 (93%) friendly and 11 (7%) hostile takeovers; an acquisition is considered hostile if the board of directors of the target firm rejects the offer, or multiple acquirers are competing for the same target. All-cash acquisitions account for 70% of the sample, whereas the remainder comprises mixed (20%) and equity-paid (10%) deals. About one-third of the sample are acquisitions that involve bidding and target firms operating in the same industry, defined based on the 2-digit NACE industry code classification.4 Most of the acquisitions involve relatively small target companies. The median relative size to the target firm, defined as the ratio of target’s to acquirer’s sales in the year prior to the takeover, does not exceed 9%.5 However, acquisitions of relatively large targets are not rare, either: in one-third of our M&As the relative size of the target firm exceeds 20%. The largest transaction in our sample is the mega-acquisition of Mannesmann by Vodafone in 2000, with a deal value of USD 203 billion. Other large transactions are the acquisition of Elf Aquitaine by TotalFina in 1999 (USD 50 billion), the merger between Germany’s Hoechst and France’s Rhone-Poulenc in 1999 (USD 22 billion), and the acquisition of Airtel by Vodafone in 2000 (USD 14 billion). The smallest transaction was the acquisition of C.K. Coffee by Coburg Group in 2001 (both British companies), with deal value of only USD 140,000.
4.3.3
Selection of peer companies
To measure changes in operating performance following a takeover, we compare the realized performance with the benchmark performance that would be generated in case the takeover bid had not taken place. However, while performing this comparison, one should take into account that operating performance is affected not only by the takeover but also by a host of other factors. To isolate the takeover effect, the literature suggests to adjust the performance for the industry trend (see, e.g., Healy, Palepu, and Ruback, 1992). Alternatively, one could match the 4
Changing to a 4-digit NACE classification does not materially change the results in the remainder of our study. In 51 deals (33%), acquirer and/or target had NACE industry code 7415 (holding company), in which case we assumed industry relatedness to be unknown. 5 When the relative size of the target firm is calculated as the ratio of target’s and acquirer’s book values and of total assets, the median relative size is 8.81%.
Table 4.3 Sample description No of deals Panel A: Completion year 1997 7 1998 26 1999 38 2000 54 2001 30 TOTAL
155
Percent (%)
5% 17% 25% 35% 19% 100%
No of Percent deals (%) Panel E: Pre-acquisition acquirer leverage(a) Leverage ⬍ 15% 65 Leverage 15%–30% 41 Leverage 30%–45% 22 Leverage ⬎ 45% 20 Unknown 7 TOTAL
Panel B: Acquirer country Austria 1 Belgium 3 Czech Republic 1 Finland 3 France 36 Germany 9 Italy 4 Netherlands 3 Norway 3 Portugal 1 Spain 8 Sweden 9 Switzerland 4 United Kingdom 70 TOTAL
1% 2% 1% 2% 23% 6% 3% 2% 2% 1% 5% 6% 3% 45%
155
100%
Panel C: Method of payment Cash 108 Stock 16 Mix 31
70% 10% 20%
TOTAL
100%
Panel D: Deal atmosphere Friendly 144 Hostile 11
93% 7%
TOTAL
155
100%
Tender offer Negotiated deal
54 101
35% 65%
TOTAL
155
100%
100%
(median leverage ⫽ 16.81%) Panel F: Pre-acquisition acquirer cash reserves(b) Cash reserves ⬍ 5% 66 43% Cash reserves 5%–10% 30 19% Cash reserves 10%–15% 17 11% Cash reserves ⬎ 15% 40 26% Unknown 2 1% TOTAL
155
100%
(median cash reserves ⫽ 6.41%) Panel G: Industry relatedness(c) Focused 49 Unfocused 55 Unknown 51 TOTAL
155
155
42% 26% 14% 13% 5%
155
32% 35% 33% 100%
Panel H: Relative size of target(d) Target size ⬍ 10% 82 Target size 10%–20% 25 Target size ⬎ 20% 47 Unknown 1 TOTAL
155
53% 16% 30% 1% 100%
(median target size ⫽ 8.28%) Panel I: Cross-border deals Domestic 104 Cross-border 51 TOTAL
155
67% 33% 100%
(a)
Defined as long-term debt plus loans divided by book value of total assets; all measures are one year prior to the year of acquisition. (b) Defined as cash and cash equivalents divided by book value of total assets; all measures are one year prior to the year of acquisition. (c) Defined by a 2-digit NACE industry code. (d)
Defined as the target’s sales divided by the acquirer’s sales; all measures are one year prior to the year of acquisition.
The long-term operating performance in European mergers and acquisitions
91
sample of firms involved in M&As by industry, asset size, and a performance measure (typically the market-to-book ratio or EBIT) with nonmerging companies (as suggested in Barber and Lyon, 1996), and examine whether merging companies outperform their nonmerging peers prior and subsequent to the bid. In our analysis we employ both adjustment methodologies in order to check whether the choice of the adjustment model affects our conclusions. As a proxy for industry trends, we consider the performance of a median company that operates in the same industry as a company involved in M&As from our sample. The industry median is identified from the pool of all companies recorded in the Amadeus database that have the same 4-digit industry code as our sample firm in the year prior to the acquisition. The firm with the median EBITDA-to-assets ratio is then selected as the industry median peer. We also used the Amadeus database to identify the industry, size, and performance-matched peer company for each bidder and target from our sample. For each bidding (target) firm, the list of Amadeus companies with the same industry code and available EBITDA-to-assets ratio has been further filtered down to the list of firms that fall within the same size quartile (as measured by total assets) as the bidding (target) firm. From this list, we select the company with an EBITDA-to-assets ratio that is closest to the ratio of the analyzed bidder (target). The selected firm constitutes our industry, size, and performance-matched peer.6 Caution is taken to select peer companies that were not engaged in M&A activity over the period studied.
4.3.4
Measures of operating performance
Most studies on post-acquisition operating performance define operating performance as “pre-tax operating cash flow” which is the sum of operating income, depreciation, interest expenses, and taxes (see, e.g., Healy, Palepu, and Ruback, 1992; Ghosh, 2001; Heron and Lie, 2002). It is typically argued that such a performance measure is unaffected either by the accounting method employed to compute depreciation or by nonoperating activities (interest and tax expenses). However, this measure is not a “pure” cash flow performance measure, as it does not take into account changes in working capital (changes in receivables, payables, and inventories). In this study, we employ two measures of cash flow: (1) EBITDA-only and (2) EBITDA minus changes in working capital.7 To adjust for the differences in size across companies, we divide these 6
Using the Peer Group function in Amadeus, we could usually gather an international (European) list of peer companies within the same industry. However, this function returned an error message when the number of peer companies in a specific industry exceeded 500 (this phenomenon happened often when a company had an NACE code 7415 (holding company), returning a huge number of industry peers). In that case we downloaded a national list of companies within the same industry, instead of an international list. 7 The number of our observations for cash flow measure (2) is lower than that for measure (1) because changes in working capital are not available for several bidders, targets, and/or their peers from our sample.
92
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
cash flow measures by (1) book value of assets and (2) sales.8 Overall, we consider four following measures of operating performance: (a) (b) (c) (d)
(EBITDA – ΔWC)/BV assets (EBITDA – ΔWC)/Sales EBITDA/BV assets EBITDA/Sales
The first measure of operating performance shows how effectively a company is using its assets to generate cash. The second measure shows how much cash is generated for every dollar of sales. The third and fourth measures do not include changes in working capital, but are comparable to the pre-tax cash flow as is used in most of the past empirical research. Figure 4.1 summarizes our methodology to estimate changes in operating performance following the takeover. Since our analysis focuses on the changes in profitability of the combined firm for the period preceding the takeover, we sum the cash flows of acquirer and target and scale it by the sum of their total assets or sales. That is, we
Firm A + T pre-acquisition
minus
Peers (A + T) pre-acquisition1
Firm (AT) post-acquisition
Median pre-acquisition adjusted performance
minus
Peers (A + T) post-acquisition2
Median post-acquisition adjusted performance
Improvement? ⫺3
⫺2
⫺1
Year of acquisition
1
2
3
1 Pre-acquisition
performance of acquirer's and target's peer company is combined with acquirer's and target's relative asset or sales weights in the years ⫺3, ⫺3 and ⫺1. Peers are selected on (1) industry median or (2) industry, size and pre-event performance. 2 Post-acquisition, combined performance of peer companies is calculated in a similar way as in pre-acquisition years. The only difference is that performance of peer companies is combined with fixed asset or sales weights of acquirer and target in year –1 (the reason is that T does not separately report assets values after the acquisition anymore).
Figure 4.1 Methodology employed to measure changes in post-merger operating performance.
8
Some U.S. research uses a third variable to scale cash flows: the market value of assets. Market value is insensitive to whether purchase or pooling of interest accounting is used in acquisitions. Until recently, U.S. companies under U.S. GAAP were free to choose between the purchase
The long-term operating performance in European mergers and acquisitions
93
compute the raw pre-acquisition profitability of the combined firm as follows: CFfirm, t ⫽
CFA, t ⫹ CFT , t BASEA, t ⫹ BASET , t
The peer pre-acquisition profitability of the combined firm is then computed as a weighted average of the profitability of the acquirer’s and the target’s peer companies, where the relative size of the acquirer’s and target’s relative asset (sales) are the weights: ⎛ ⎞⎟ CFpeerA, t BASEA, t ⎜ ⎟⎟ ⫻ CFpeer , t ⫽ ⎜⎜ ⎟ ⎜⎝ BASEA, t ⫹ BASET , t ⎠⎟ BASEpeerA, t ⎛ ⎞⎟ CFpeerT , t BASET , t ⎜ ⎟⎟ ⫻ ⫹⎜⎜ ⎟ ⎜⎝ BASEA, t ⫹ BASET , t ⎟⎠ BASEpeerT , t
For the years following the acquisition, the raw profitability of the combined firm is the realized cash flow of the merged company scaled by its total assets or sales: CFfirm, t ⫽
CFAT BASEAT
The peer post-acquisition profitability of the combined firm is calculated in a way similar to that for the pre-acquisition years: a weighted average of the profitability of the acquirer’s and target’s peers. However, the weights used to compute the peer post-acquisition profitability of the combined firm are the ones that we also method and the pooling of interest method to account for their acquisitions. In the former method, the acquirer records the target’s assets at fair value. If the amount paid for a company is greater than fair market value, the difference is reflected as goodwill. The pooling of interests does not require the target’s assets to be recorded at fair value and no goodwill is booked. The problem with the use of the market value of assets to scale cash flows is that the market value can hide operating improvements: The market value may already incorporate possible improvements (or declines) in operating performance in the denominator on the day of the takeover announcement. Hence, possible changes in the numerator (cash flows) can be neutralized by the change in the market value of the denominator. Healy, Palepu, and Ruback (1992) solve this problem by excluding the changes in market capitalizations at the merger announcement. However, even after excluding changes in equity value around the announcement, performance may still be biased because acquiring firms’ market values decline systematically over 3 to 5 years following acquisitions (Agrawal, Jaffe, and Mandelker, 1992). Another reason we did not scale by market value is that the European accounting regulation (IAS) allows only the purchase method of accounting in corporate acquisitions. Finally, in order to be able to use market values we require both acquirer and target to be listed—a requirement that reduces our sample by half.
94
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
used to compute the peer pre-acquisition profitability. That is, the peer postacquisition profitability of the combined firm is calculated as follows: ⎛ ⎞⎟ CFpeerA, t BASEA, t⫺1 ⎜ ⎟⎟ ⫻ CFpeer , t ⫽ ⎜⎜ ⎜⎝ BASEA, t⫺1 ⫹ BASET , t⫺1 ⎟⎟⎠ BASEpeerA, t ⎛ ⎞⎟ CFpeerT , t BASET , t⫺1 ⎜ ⎟⎟ ⫻ ⫹ ⎜⎜ ⎟ ⎜⎝ BASEA, t⫺1 ⫹ BASET , t⫺1 ⎠⎟ BASEpeerT , t
A company’s profitability adjusted for industry trend is calculated as a difference between the company’s raw and peer profitability: CFind⫺adjusted , t ⫽ CFfirm, t ⫺ CFpeer
industry , t
an nd
CFind , size , perf ⫺adjusted , t ⫽ CFfirm, t ⫺ CFpeer
ind , size , perf ⫺ adjusted
,t
In order to assess the changes in the profitability of the combined firm caused by the takeover, we employ two models: the change model and the intercept model. The change model calculates the change in profitability for each firm such that the median profitability of the 3 years prior to the takeover is compared to the median profitability over the 3 years subsequent to the merger. With a Wilcoxon signed rank examine, we then examine whether the median postacquisition performance is significantly different from the median pre-acquisition performance.9 An analysis of changes in operating performance is also performed using means (over the 3 years before and after the acquisition) instead of medians. The results are qualitatively similar and are available upon request. The intercept model estimates changes in operating performance with the intercept (α0) from the following regression: post pre medianCFadjusted ⫽ α0 ⫹ α1 ⋅ medianCFadjusted ⫹ε
Factor α1 reflects a relation between pre- and post-acquisition profits, whereas changes in profitability are captured by the intercept (α0). To test for the significance of the changes we apply a standard t-test.
9
The use of medians has the disadvantage that median differences are sometimes counterintuitive. For example, the post-acquisition performance minus the pre-acquisition performance can be a negative, whereas one would expect a positive difference (e.g., when the median pre-performance is ⫺0.04% and the median post-performance is 0.84%). This difference arises from the fact that median differences are not calculated simply by subtracting median pre-performance from median post-performance, but are calculated as the median of the differences.
The long-term operating performance in European mergers and acquisitions
4.4
95
Changes in corporate performance caused by M&As: results
4.4.1
Does operating performance improve following acquisitions?
Table 4.4 exhibits insignificant changes in profitability of the combined firm after the takeover. Of our four performance measures, none reveals significant changes: measures scaled by book value indicate an insignificant decrease in performance (⫺0.01% for measure 3 and ⫺0.62% for measure 1) while measures scaled by sales point to an insignificant increase (⫹0.15% for measure 2 and ⫹0.16% for measure 4). This result is in line with the previous empirical studies that document insignificant changes in operation performance following mergers (Mueller, 1980; Sharma and Ho, 2002; and Ghosh, 2001). On the other hand, the result contradicts the findings of a number of other studies showing a significant improvement or deterioration in post-acquisition
Table 4.4 Changes in operating performance following acquisitions Raw performance
Industry-adjusted
Industry-, size-, and performance-adjusted
Median # of (%) obs.
Median % # of (%) positive obs.
⫺3 ⫺2 ⫺1
10.23 11.46 10.75
56 91 123
⫺1.04 46% ⫺0.05 50% 3.77b 64%
37 76 110
1.16 1.82 0.01
51% 56% 50%
39 71 105
Median pre-acquisition performance
10.82
125
1.58a 56%
114
⫺0.04
49%
110
1 2 3
10.16 8.95 9.53
121 120 110
3.15 59% 1.77 57% 3.55c 74%
102 99 92
0.87 –0.60 1.87
55% 49% 63%
103 101 88
9.16
125
3.27c 68%
114
0.84
55%
110
Median % (%) positive
# of obs.
MEASURE 1 (EBITDA – ΔWC) BVassets
Median post-acquisition performance Median difference
⫺0.90** 125
⫹0.05
114
⫺0.62
110
% positive differences
42%
52%
114
49%
110
125
(continued)
Table 4.4 (contined) Raw performance
Industry-adjusted
Industry-, size-, and performance-adjusted
Median # of (%) obs.
Median % # of (%) positive obs.
median % # of (%) positive obs.
8.65 10.12 9.78
57 91 122
3.56b 3.92a 3.23b
67% 62% 60%
35 71 106
1.57 ⫺0.69 0.45
59% 46% 51%
39 69 101
8.98
125
3.34b
60%
112
⫺0.51
50%
107
11.24 9.46 10.41
122 120 112
3.56b 3.58c 4.79c
65% 65% 71%
99 95 91
2.98 ⫺0.35 1.75
58% 49% 53%
101 97 88
9.46
125
3.94c
66%
112
1.05
54%
107
MEASURE 2 (EBITDA – ΔWC) Sales ⫺3 ⫺2 ⫺1 Median pre-acquisition performance 1 2 3 Median post-acquisition performance Median difference
⫺0.03
125 ⫹1.69
112
⫹0.15
107
% positive differences
49%
125 56%
112
51%
107
MEASURE 3 EBITDA BV assets ⫺3 ⫺2 ⫺1
12.48 12.51 11.70
92 122 155
2.73c 2.37c 1.89c
64% 60% 62%
85 116 154
1.07c 1.59b 0.15a
67% 59% 58%
75 108 153
Median 12.27 pre-acquisition performance
155
2.12c
61%
154
0.39c
61%
154
1 2 3
10.17 10.02 9.82
153 149 142
0.83 0.52 1.34b
55% 55% 58%
148 140 132
0.54 0.67 0.83
52% 55% 55%
151 146 131
Median 10.23 post-acquisition performance
155
0.67a
59%
154
0.43
54%
154
Median difference
–1.73*** 155
–1.02**
154
–0.01
154
% positive differences
37%
44%
154
50%
154
155
The long-term operating performance in European mergers and acquisitions
Raw performance
Industry-adjusted
Median (%)
# of obs.
Median % (%) positive
11.02 10.82 11.10
90 119 153
4.35c 4.01c 3.59c
Median 11.26 pre-acquisition performance
154
1 2 3
10.75 10.67 10.48
Median 11.44 post-acquisition performance
97
Industry-, size-, and performance-adjusted
# of obs.
Median (%)
% positive
# of obs.
68% 65% 70%
78 108 145
2.39a 0.50 0.90
63% 52% 58%
71 103 146
3.34c
70%
147
0.99
58%
148
153 150 149
3.11c 3.23c 2.75c
63% 67% 67%
140 135 133
1.49 1.57 1.48
57% 56% 56%
146 141 134
154
3.22c
69%
147
1.72
55%
148
MEASURE 4 EBITDA Sales ⫺3 ⫺2 ⫺1
Median difference
⫺0.65*
154
⫺0.12
147
⫹0.16
148
% positive differences
45%
154
48%
147
52%
148
*** / ** / * Significant at the 1% / 5% / 10% level. Wilcoxon signed rank test shows that median post-acquisition performance is significantly different from median pre-acquisition performance. a b c / / Significant at the 10% / 5% / 1% level. Wilcoxon signed rank test shows that the firm’s performance is significantly different from those of peer performance in the same year. # Working capital is defined each year as stocks plus accounts receivable minus accounts payable. ## For measures 1 and 2, cash flow in 1995 equals EBITDA (changes in working capital are not included) because changes in working capital are not available in Amadeus for that year. Our conclusions do not change materially when we exclude year 1995 from the analysis. ### For measures 2 and 4, one deal is removed from the sample because the acquirer in this deal has an EBITDA-to-sales ratio of ⫺26, 200% in the year following the acquisition (i.e. almost zero sales are recorded in that year).
performance (Kruse, Park, Park, and Suzuki, 2002; Yeh and Hoshino, 2001; Dickerson, Gibson, and Tsakalotos, 1997; Clark and Ofek, 1994; Rahman and Limmack, 2004). However, this difference in the results may be driven by the fact that none of the previous studies uses a pure cash flow measure (which includes changes in working capital). This omission may engender downward bias in their profitability measures. Furthermore, most of the U.S. studies that find significant increases in cash flow performance employ market value of
98
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
assets to scale cash flows (Linn and Switzer, 2001; Parrino and Harris, 1999; Switzer, 1996; Healy, Palepu, and Ruback, 1992), whereas our analysis is based on the book value of assets and sales. The comparison of the raw performance (without adjusting for the industry) reveals that the post-acquisition cash flow declines substantially, with 3 out of the 4 performance measures showing significant decreases ranging between ⫺0.65% and ⫺1.73% (see Table 4.4). The result is in line with Powell and Stark (2005), who show that the raw operating performance of the combined firm generally deteriorates following U.K. takeovers. Another important result presented in Table 4.4 is that bidding and target companies significantly outperform the median companies in their respective industries prior to the takeover. This suggests that companies undertake corporate acquisitions in periods when they are performing better than their median peers in the industry. We further investigate whether the inclusion of changes in the working capital into our performance measure has a significant impact on the overall results. Table 4.5 reports the changes in the use of working capital over the postacquisition period compared to those over the pre-acquisition period. Our evidence suggests that the use of working capital does not change significantly following the takeover.
Robustness checks In this section, we investigate whether our results are robust with respect to different specifications of the profitability measures. First, we recalculate changes in the operating performance using means rather than medians (see Section 4.3.4). That is, for each combined firm, we calculate mean annual pre- and post-acquisition
Table 4.5 Do takeovers lead to a better management of working capital? MEASURE
Raw changes in working capital
Obs.
Industryadjusted
Obs.
Working capital; adjusted for industry, size and pre-event performance
Obs.
(1) Working capital
⫺2.18***
151
⫺1.97**
146
⫺0.43
144
⫺0.80
150
⫺0.72
139
⫹0.77
137
BV assets (2) Working capital Sales *** / ** Significant at the 1% / 5% level. Wilcoxon signed rank test shows that the median level of post-acquisition working capital is significantly different from the median level of pre-acquisition working capital.
The long-term operating performance in European mergers and acquisitions
99
performance and adjust it to the mean pre- and post-operating performance of the combined peer companies. Unsurprisingly, we find that the results based on means are more volatile than those based on medians because of the influence of outliers. Nonetheless, our initial conclusion remains unchanged, as we find no statistically significant changes in the operating performance following acquisition. Second, we employ the market value of assets as an alternative scale factor for our cash flow measure, as applied in previous U.S. studies. Following Healy, Palepu, and Ruback (1992), we define the market value of assets as the market capitalization of equity plus the book value of net debt. In some cases, new peers have to be selected, as some original peers are not listed and hence lack market capitalization data. The results indicate that, when the market value is used to scale the performance measures, changes in the operating performance following takeovers are positive.10 However, as none of the changes is significant, our initial conclusion remains unchanged. Third, we examine whether the intercept model yields conclusions different from the change model. Panel A of Table 4.6 exhibits that, consistent with Powell and Stark (2005), Ghosh, (2001), Linn and Switzer (2001), and Switzer (1996), the intercept model gives structurally higher estimates of the performance improvements than the change model. The explanation is that the change model is based on medians and is therefore less sensitive to outliers, whereas the intercept model is based on means. Panel B shows that the slope coefficients are significant in 7 out of 8 regressions, which suggests that the post-acquisition performance is related to pre-acquisition performance. Strikingly, when we adjust operating performance only by industry, we find that high pre-acquisition profitability is associated with higher post-acquisition profitability. However, when the adjustment is made on the basis of industry, size, and performance, we observe a significant negative relationship: High pre-acquisition profitability is followed by lower post-acquisition results. This negative relationship highlights the importance of the adjustment approach employed and may explain the contradictory results across many studies.
4.5
The determinants of the post-acquisition operating performance
In this section, we investigate whether the characteristics of the takeover deal predict the post-acquisition performance of the combined firm. We test whether post-acquisition performance of the merged firm varies across takeovers with different means of payment, degree of hostility, business expansion strategy (focus versus diversification), and geographic scope of the deal (domestic
10
Summary tables of the robustness checks are available upon request.
100
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 4.6 The change model versus the intercept model: comparison of results Panel A: Median change in operating performance (%) Measure
Industry-adjusted
Industry-, size-, and performance-adjusted
Change model
Intercept model
Change model
Intercept model
1. (EBITDA ⫺ ΔWC) BV assets
⫹0.1
⫹0.3 (a)
⫺0.6
⫺0.2 (e)
2.
(EBITDA ⫺ ΔWC) Sales
⫹1.7
⫹12.0** (b)
⫹0.2
⫹9.3 (f)
3.
EBITDA BV assets
⫺1.0**
⫹0.5 (c)
⫺0.01
⫹0.5 (g)
4.
EBITDA Sales
⫺0.1
⫹24.3 (d)
⫹0.2
⫹0.9 (h)
Panel B: Regression models related to the Intercept model post pre (a) medianCFind ⫽ 0.003 ⫹ 0.261 ⭈medianCFind
R2 ⫽ 0.06
(0.18) (2.76***) (b)
post ⫽ medianCFind
pre 0.120 ⫹ 0.786 ⭈medianCFind
(c)
post ⫽ medianCFind
(d)
post ⫽ medianCFind
(e)
post medianCFind size, pref
(f)
post medianCFind size, pref
(g)
post medianCFind size, pref
(h)
post medianCFind size, pref
R2 ⫽ 0.61
(2.49**) (13.06***) pre 0.005 ⫹ 0.506 ⭈medianCFind
R2 ⫽ 0.04
(0.21) (2.51**) pre 0.243 ⫹ 1.937 ⭈medianCFind
R2 ⫽ 0.05
(1.03) (2.62***) post ⫽ ⫺0.002 ⫺ 0.423 ⭈medianCFind size, pref
R2 ⫽ 0.09
(⫺0.10) (⫺3.26***) post ⫽ 0.093 ⫺ 0.206 ⭈medianCFind size, pref
R2 ⫽ 0.01
(1.06) (⫺0.93) post ⫽ 0.005 ⫺ 0.506 ⭈medianCFind size, pref
R2 ⫽ 0.04
(0.21) (⫺2.51**) post ⫽ 0.009 ⫺ 0.380 ⭈medianCFind size, pref
R2 ⫽ 0.15
(0.76) (⫺5.01***) *** / ** Significant at 1% / 5%, using a two-tailed test
versus cross-border M&As). We also investigate whether the relative size of the target firm and the pre-acquisition leverage and cash holdings of the acquirer influence the post-acquisition performance of the combined firm. Our primary focus in this section is on measures of the raw performance and the performance
The long-term operating performance in European mergers and acquisitions
101
adjusted for industry, size and pre-event performance.11 Also, we present results only for the profitability measures that are corrected for the changes in working capital: (EBITDA – ΔWC)/BV assets and (EBITDA – ΔWC)/Sales.
4.5.1
Method of payment: cash versus equity
Table 4.7 shows the post-acquisition performance of the merged firms for the subsamples of takeovers partitioned by means of payment: all-cash offers, cashand-equity offers, and all-equity offers. The results presented in the table suggest that no significant differences are displayed in the profitability of corporate takeovers employing different methods of payment. The results are in accord with previous studies (Healy, Palepu, and Ruback, 1992; Powell and Stark, 2005; Heron and Lie, 2002; Sharma and Ho, 2002).
4.5.2
Deal atmosphere: friendly versus hostile takeovers
Panel A of Table 4.8 shows that hostility in corporate takeovers is associated with lower post-merger profitability. However, the effect is not statistically significant. Therefore, we conclude that no evidence adequately supports the notion that hostile takeovers create more (long-term) synergistic value than do friendly ones. The result is consistent with previous empirical findings for the United States (see, e.g., Ghosh, 2001; Louis 2004). The lack of significant differences in the performance of hostile and friendly offers may arise from the fact that the sample of friendly acquisitions includes a high number of deals conducted in a form of tender offers, which are almost as expensive as hostile takeovers.12 Therefore, we further test whether the form of the acquisition (tender offer versus negotiated deal) has an impact on the post-merger profitability of the combined firm. Panel B of Table 4.8 reports that the difference in profitability of tender offers and negotiated deals is statistically insignificant. However, the difference seems to be significant in economic terms, as we find that the combined profitability of the bidding and target firms somewhat declines following a tender offer, whereas it increases following a negotiated M&A deal. The overall results are similar to those of Switzer (1996), Linn and Switzer (2001) and Moeller and Schlingemann (2003), who find no statistical difference in the long-term performance of hostile and friendly acquisitions in the United States.
4.5.3
Pre-acquisition leverage and cash holdings of the acquirer
In this section, we investigate whether highly leveraged acquirers outperform lowleveraged acquirers owing to better creditor monitoring. We divide our sample into 11 12
The results of the analysis based on the performance adjusted for industry is available upon request. Grossman and Hart (1980) show that small shareholders may hold out their shares in a tender offer until the bidder increases the offer price, hence rendering tender offers among the most expensive forms of acquisition.
Table 4.7 Median changes in operating performance (% point) by means of payment Obs.
Stock
Obs.
Mix
Obs.
Statistical significance of difference H0: Cash ⫽ Stock ⫽ Mix (chi-square)
Statistical significance of difference H0: Cash ⫽ Stock (Mann–Whitney)
88 88
⫺2.79 ⫹1.18
11 11
⫺0.50 ⫺0.22
26 26
no no
no no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/BV assets ⫹0.95 78 ⫺1.21 10 ⫺1.86 22 2. (EBITDA – ΔWC)/Sales ⫹0.08 76 ⫹2.23 10 ⫺1.27 21
no no
no no
MEASURE
Cash
Raw performance 1. (EBITDA – ΔWC)/BV assets ⫺0.83** 2. (EBITDA – ΔWC)/Sales ⫺0.31
** Significant at the 5% level; Wilcoxon sign rank test shows that the median post-acquisition performance is significantly different from median pre-acquisition performance.
The long-term operating performance in European mergers and acquisitions
103
Table 4.8 Median changes in operating performance (% points) by the attitude of the target’s board towards the bid (hostile versus friendly) and by the form of takeover (tender offer versus negotiated deal) Panel A: Attitude of the target’s board toward the bid (hostile versus friendly) MEASURE
Friendly
Obs. Hostile Obs. Statistical significance of difference (Mann–Whitney)#
Raw performance 1. (EBITDA – ΔWC)/BVassets 2. (EBITDA – ΔWC)/Sales
⫺0.83** 118 ⫺0.02 118
⫺2.67 ⫺1.94
7 7
no no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/BVassets ⫹0.60 104 ⫺6.31 6 no 2. (EBITDA – ΔWC)/Sales ⫹0.31 101 ⫺5.81 6 no Panel B: Form of takeover (tender offer versus negotiated deal) MEASURE
Raw performance 1. (EBITDA – ΔWC)/BVassets 2. (EBITDA – ΔWC)/Sales
Negotiated Obs. Tender Obs. Statistical significance deal offer of difference (Mann–Whitney)# ⫺1.51** ⫺0.65
79 79
⫹0.03 46 ⫹0.86 46
no no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/BVassets ⫹0.95 68 ⫺1.28 42 no 2. (EBITDA – ΔWC)/Sales ⫹0.65 65 ⫺1.53 42 no ** Significant at the 5% level. Wilcoxon signed rank test shows that median post-acquisition performance is significantly different from median pre-acquisition performance. # To test for significance of the difference in profitability of friendly and hostile deals we employ a Mann–Whitney test.
quartiles by the acquirers’ pre-acquisition leverage and test for significance of the differences in post-acquisition profitability of the combined firms across the subsamples. We define leverage as the ratio of the book value of total debt (long-term and short-term debt) to the book value of total assets.13 Panel A of Table 4.9 shows that higher levels of pre-acquisition leverage do not lead to higher postacquisition profitability. Therefore, we conclude that pre-acquisition acquirer’s 13
The first quartile subsample includes companies with leverage lower than 5.55%; the leverage of companies in the second quartile subsample ranges between 5.55% and 16.79%; in the third quartile it is between 16.79% and 32.44%; and in the fourth quartile it is above 32.44%. For 7 acquirers, the pre-acquisition leverage is not available and hence these companies are excluded from the analysis.
Table 4.9 Median changes in operating performance (%-point) by leverage and cash reserves of the acquirer Panel A: Leverage MEASURE
Lev Q1
Obs.
Lev Q2
Obs.
Lev Q3
Obs.
Lev Q4
Obs.
Statistical significance of difference H0: Q1 ⫽ Q2 ⫽ Q3 ⫽ Q4 (chi-square)#
Statistical significance of difference H0: Q1 ⫽ Q4 (Mann–Whitney)#
Raw performance 1. (EBITDA – ΔWC)/BV assets 2. (EBITDA – ΔWC)/Sales
⫺2.67 ⫺0.53
29 30
⫺1.11 ⫹0.73
28 27
⫺1.28 ⫺0.66
32 32
⫺0.10 ⫺0.00
29 29
no no
no no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/BV assets ⫺1.12 26 ⫹0.00 22 ⫺0.32 2. (EBITDA – ΔWC)/Sales ⫹0.18 26 ⫹2.73 21 ⫹2.87
28 27
⫹1.57 ⫺1.57
27 26
no no
no no
* None of the changes in post-acquisition performance is significantly different from pre-acquisition performance, using the Wilcoxon signed rank test # There are no significant differences among the leverage quartiles, using both a chi-square test and a Mann–Whitney test to see whether the median values differ.
Panel B: cash reserves MEASURE
Cash Q1
Obs.
Cash Q2
Obs.
Cash Q3
Obs.
Cash Q4
Obs.
Statistical significance of difference H0: Q1 ⫽ Q2 ⫽ Q3 ⫽ Q4 (chi-square)#
Statistical significance of difference H0: Q1 ⫽ Q4 (Mann–Whitney)#
Raw performance 1. (EBITDA – ΔWC)/BV assets 2. (EBITDA – ΔWC)/Sales
⫺0.10 ⫹1.90
29 29
⫺0.96 ⫹0.04
29 29
⫺3.28** ⫺1.01
33 33
⫺0.19 ⫹0.02
32 32
no no
no no
27 25
⫺2.45 ⫺3.79
30 29
no no
no no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/BV assets ⫹1.57 25 ⫺0.10 26 ⫺0.63 2. (EBITDA – ΔWC)/Sales ⫹3.32 25 ⫹1.03 26 ⫹0.31
** Significance at the 5% level. Wilcoxon signed rank test shows that median post-acquisition performance is significantly different from median pre-acquisition performance. # To test for statistical significance of the differences in performance measures across the sub-groups, we employ a chi-square test when 2 subgroups are compared and a Mann–Whitney when more than 2 sub-groups are compared.
The long-term operating performance in European mergers and acquisitions
105
leverage has no impact on the operating performance of the combined firm following the takeover. Likewise, none of the U.S. studies finds a significant relation between leverage and post-acquisition operating performance (e.g., Linn and Switzer, 2001; Switzer, 1996; Clark and Ofek, 1994). Acquirer’s cash reserves may be another important determinant of the postacquisition performance of the combined firm, as Jensen’s (1986) free cash flow theory predicts that managers of cash-rich firms are more likely to be involved in poor takeovers. We test this conjecture by comparing the post-acquisition profitability of combined companies across the quartiles based on the relative amount of cash reserves held by the acquirer prior to the acquisition. We define an acquirer’s cash reserves as the firm’s cash and cash equivalents scaled by book value of total assets one year prior to the acquisition.14 Panel B of Table 4.9 shows that, even though none of the changes in post-acquisition profitability is statistically significant, the trend is clear toward better long-term performance of the takeovers by acquirers with lower cash reserves. Acquirers with the lowest level of cash holdings (first quartile) experience an increase in the post-acquisition profitability by 1.57%, whereas acquirers with the highest level of cash reserves (fourth quartile) experience a decline by 2.46%. The results are in line with the findings of Harford (1999), who shows that acquisitions by cash-rich companies lead to significant deteriorations in the operating performance of the combined firm.
4.5.4
Industry relatedness: focus versus diversification strategy
A number of empirical studies are dedicated to the analysis of whether the relatedness of the merging firms’ businesses is associated with higher post-merger profitability (see, e.g., Powell and Stark, 2005; Linn and Switzer, 2001; Switzer, 1996; Sharma and Ho, 2002). These studies find no significant differences in the postmerger profitability of related and unrelated acquisitions, of takeovers with a focus strategy and diversifying mergers, of horizontal and vertical takeovers, of takeovers that aim at product expansion and those that do not. Similarly, Table 4.10 unveils no significant relationship between takeover strategy (diversification vs. focus) and the post-acquisition performance of the combined firms.15 We consider an acquisition to be diversifying if the acquiring and target companies operate in unrelated industries as defined by their 2-digit NACE industry classification.16 We conclude that a takeover strategy based on industry-relatedness has no impact on the post-acquisition performance of the combined firm. 14
The first quartile subsample includes companies with cash reserves lower than 2.47% of total assets, the cash reserves of companies in the second quartile subsample range between 2.47% and 6.39% of total assets, in the third quartile cash is between 6.39% and 15.37%, and in the fourth quartile it is above 15.37%. For 2 acquirers, the pre-acquisition data on cash reserves are not available, and these companies are excluded from the analysis. 15 We exclude 51 M&As from the analysis when either the acquirer or the target or both have 7415 (holding company) as NACE industry code or when the NACE code for one of the parties involved in the deal is not available. 16 As a robustness check, we also perform the analysis based on the 4-digit NACE industry code classification. We find that employing a 4-digit industry code does not materially affect the results.
106
International M&A Activity Since 1990: Recent Research and Quantitative Analysis Table 4.10
Median changes in operating performance (%-point) by takeover strategy (focus versus diversification)
MEASURE
Diversification
Obs.
Focus
Obs.
Statistical significance of difference (Mann–Whitney)
Raw performance 1. (EBITDA – ΔWC)/BV assets 2. (EBITDA – ΔWC)/Sales
⫺0.65 ⫹0.36
47 47
⫺3.03 ⫺1.93
40 39
Yesa no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/BV assets ⫺0.88 40 ⫺0.02 34 no 2. (EBITDA – ΔWC)/Sales ⫹2.73 39 ⫺1.35 33 no * None of the changes in post-acquisition performance are statistically significantly different from preacquisition performance (the conclusion is based on the results of the Wilcoxon signed rank test) a Significantly different at the 10% level. Mann-Whitney test shows whether focus acquisition strategy leads to a significantly different performance of the combined firm than the diversification strategy.
4.5.5
Relative size of the target
The long-term M&A performance literature yields contradictory conclusions about whether or not the size of the takeover transaction matters for the postacquisition profitability of the combined firm. To test whether size matters in European M&As, we partition our sample into two subsamples by the relative size of the target firm. The subsample of large targets includes deals that involve target companies with pre-acquisition sales of at least 20% of the sales of their acquirers (44 deals). The rest of takeovers comprise the subsample of small target transactions (82 deals). Table 4.11 documents that relatively large takeovers significantly outperform their smaller peers. Combined firms experience an increase in profitability by 3.36% following the acquisition of a relatively large target and a decrease by 1.35% following the takeover of smaller targets. A possible explanation is that larger M&As have a greater scope to explore financial and operating synergies, thereby resulting in a sizable improvement in profitability for the combined firm. When we split our sample into quartiles by the relative size of the target firm, we find that, although the post-merger profitability increases with size, this increase is nonlinear. The very large M&As (fourth quartile) tend to be less profitable than the medium-sized M&As (third quartile), but only the smallest M&As (first quartile) tend to have a significant negative impact on the operating performance of the combined firms.17 The result confirms our conjecture that problems of managing a very large newly created firm may outweigh the alleged benefits of the takeover and hence worsen profitability of the combined firm.
4.5.6
Domestic versus cross-border deals
Table 4.12 examines whether the post-acquisition performance evolves differently following domestic and cross-border M&As. Panel A shows that the profitability 17
Table is available upon request.
The long-term operating performance in European mergers and acquisitions
107
Table 4.11 Median changes in operating performance (%-point) by the relative size of the target firm MEASURE
Large Obs. Small Obs. Statistical significance targets# targets of difference (Mann–Whitney)
Raw performance 1. (EBITDA – ΔWC)/ ⫺2.17 BV assets 2. (EBITDA – ΔWC)/ ⫹1.18 Sales
44
⫺0.76 81
no
43
⫺0.40 82
no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/ ⫹0.63 39 ⫺1.12 71 no BV assets 2. (EBITDA – ΔWC)/ ⫹3.36 38 ⫺1.35 69 yesa Sales * None of the changes in performance is significant at least at the 10%-level. a Significantly different at the 10% level. Mann–Whitney test shows whether the profitability of relatively large M&As is significantly different from that of the relatively small M&As. # The subgroup includes acquisitions of relatively large targets with sales of at least 20% of bidder sales in the year prior to the acquisition. The ‘Small targets’ subgroup includes deals that involve relatively small targets with the relative (to the acquirer) size of sales of less than 20%.
of the combined firm increases by 0.5% following domestic takeovers and decreases by 1.8% following cross-border deals. Although the difference in the changes in performance is not statistically significant, it is notable in economic terms. We further investigate whether the performance of domestic takeovers across countries shows a difference. We therefore divide our sample of domestic M&As into three subsamples: U.K., French, and other deals. Panel B of Table 4.12, in presenting a comparison of U.K. and French M&As, yields inconclusive results, as the conclusion depends on the analyzed profitability measure. However, none of the performance measures shows statistically significant differences in the profitability of U.K. and French M&As. In contrast, independent of the analyzed profitability measure, takeovers undertaken in other European countries systematically outperform their U.K. and French counterparts (the difference is still not statistically significant). We conclude that the profitability of corporate takeovers is similar across all continental European countries and the United Kingdom.
4.5.7
Multivariate analysis
Table 4.13 summarizes the results of our univariate analysis of the determinants of post-merger profitability. In this section, we explore the combined effect of the determinants of the profitability in a multivariate framework. Table 4.14
Table 4.12 Median changes in operating performance (%-point) in domestic and cross-border M&As Panel A: Domestic versus cross-border deals MEASURE
Raw performance 1. (EBITDA – ΔWC)/BV assets 2. (EBITDA – ΔWC)/Sales
Domestic M&As
Obs.
Cross-border M&As
Obs.
Statistical significance of difference (Mann–Whitney)
⫺0.90 ⫺0.24
85 85
⫺0.99* ⫹0.02
40 40
no no
37 37
no no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/BV assets ⫹0.57 73 ⫺1.81 2. (EBITDA – ΔWC)/Sales ⫹0.48 70 ⫺1.79
* Significant at the 10% level. Wilcoxon signed rank test shows that median post-acquisition performance is significantly different from median pre-acquisition performance. # None of the differences in profitability of domestic and cross-border deals are significant at least at the 10% level (based on the results of the Mann–Whitney test).
Panel B: Domestic deals for the UK, France and other European countries (%-point) MEASURE
Deals within the UK
Obs.
Deals within France
Obs.
Deals within Other Countries
Obs.
Statistical significance of difference H0: UK ⫽ FR ⫽ OTH (chi-square)
Statistical significance of difference H0: UK ⫽ FR (Mann–Whitney)
⫺0.86 ⫹0.47
42 42
⫹0.35 ⫹1.32
22 22
⫺2.33 ⫺2.95
21 21
no no
no no
Adjusted for the performance of industry-, size-, and performance-matched peer 1. (EBITDA – ΔWC)/BV asset ⫺0.87 34 ⫹1.83 19 ⫹2.02 2. (EBITDA – ΔWC)/Sales ⫹0.26 32 ⫺0.14 18 ⫹1.00
20 20
no no
no no
Raw performance 1. (EBITDA – ΔWC)/BV assets 2. (EBITDA – ΔWC)/Sales
#
None of the changes in the performance is significant at least at the 10% level (the conclusion is based on the results of the Wilcoxon signed rank test) To test for statistical significance of the differences in performance measures across the subgroups, we employ a chi-square test when 2 subgroups are compared and a Mann–Whitney when more than 2 subgroups are compared.
##
Table 4.13 Summary of the results1 Section
Variables
5.1
Cash versus equity payment
Is there a significant change in adjusted operating performance for each subgroup individually?2 ●
●
5.2
Friendly versus hostile takeovers
●
●
5.2
Tender offer versus negotiated deal
●
●
5.3
Pre-acquisition leverage of the acquirer
●
●
5.3
Pre-acquisition cash reserves of the acquirer
●
●
Are the differences across the subgroups statistically significant?3
Cash offers: an increase in operating profitability by 0.95% (statistically insignificant); Stock and mixed offers: a decrease by 1.21% and 1.86% respectively (statistically insignificant).
No
Friendly M&As: an increase in profitability by 0.60% (statistically insignificant); Hostile bids: a decrease by 6.31% (statistically insignificant).
No
Tender offers: a decline in profitability by 1.28% (statistically insignificant); Negotiated deals: an increase in profitability by 0.95% (statistically insignificant).
No
Acquirers with lowest level of leverage (Q1): a decline in profitability by 1.12% (statistically insignificant); Acquirers with highest level of leverage (Q4): an increase in profitability by 1.57% (statistically insignificant).
No
Acquirers with lowest level of cash reserves (Q1): an increase in profitability by 1.57% (statistically insignificant); Acquirers with highest level of leverage (Q4): a decline in profitability by 2.45% (statistically insignificant).
No
(continued)
Table 4.13 (continued) Section
Variables
5.4
Focus versus diversification takeover strategy
Is there a significant change in adjusted operating performance for each subgroup individually? 2 ●
●
5.5
Relative size of the target
●
●
5.6
Domestic versus cross-border M&As
●
●
1
Are the differences across the subgroups statistically significant? 3
Industry focus: a decline in profitability by 0.02% (statistically insignificant); Industry diversification: a decline in profitability by 0.88% (statistically insignificant).
No
Relatively small targets (Q1): a decline in profitability by 1.12% to 1.35% (statistically insignificant); Relatively large targets (Q4): an increase in profitability by 0.63% to 3.36% (statistically insignificant).
Yes
Domestic M&As: an increase in profitability by 0.57% (statistically insignificant); Cross-border M&As: a decline in profitability by 1.81% (statistically insignificant).
No
Note that the main conclusions are based on the performance measures adjusted for the performance of a peer company matched by industry, size and pre-acquisition performance (and not on the raw performance measure). 2 To test for statistical significance of the results we employ a Wilcoxon sign rank test. 3 To test for statistical significance of the differences across the subgroups we employ a Mann–Whitney test when comparing 2 subgroups and a chi-square test when comparing more than 2 subgroups.
Table 4.14 The determinants of post-merger operating performance: multivariate analysis Dependent variable: performance adjusted for industry-, size-, and pre–acquisition performance (EBITDA – ΔWC)/BV assets
Independent variables Intercept Pre-acq. adj. performance Cash payment (dummy)
(1)
(2)
(3)
(4)
⫺0.018 (⫺0.26)
⫺0.025 (⫺0.48)
⫺0.006 (⫺0.17)
⫺0.002 (⫺0.10)
⫺0.184 (⫺0.88)
⫺0.380** (⫺2.57)
⫺0.449*** ⫺0.423*** (⫺3.28) (⫺3.26)
0.034 (0.61)
0.002 (0.04)
⫺0.111 (⫺0.94)
⫺0.039 (⫺0.50)
⫺0.029 (⫺0.38)
0.073 (1.31)
0.038 (0.97)
⫺0.022 (0.59)
Acquirer pre-acq. leverage
⫺0.135 (⫺0.87)
0.041 (0.38)
Acquirer pre-acq. cash reserves
⫺0.218 (⫺1.05)
⫺0.037 (⫺0.23)
Hostile bid (dummy) Tender offer (dummy)
Industry focus (dummy)
(EBITDA – ΔWC)/Sales
0.040 (0.75)
⫺0.048 (⫺0.33)
(5)
(6)
(7)
(8)
Expected sign
0.146 (0.36)
0.044 (0.16)
⫺0.005 (0.03)
0.093 (1.06)
⫺0.153 (⫺0.14)
⫺0.223 (⫺0.92)
⫺0.242 (⫺1.06)
⫺0.206 (⫺0.93)
0.230 (0.68)
0.039 (0.17)
⫺0.458 (⫺0.65)
⫺0.310 (⫺0.75)
⫺0.262 (⫺0.66)
Negative
0.718** 0.308 (2.12) (1.47)
0.243 (1.27)
Negative
⫺1.387 (⫺1.43)
⫺0.395 (⫺0.67)
⫺1.672 (⫺1.34)
⫺0.953 (⫺1.09)
⫺0.099 (⫺0.30)
Zero
Zero
Zero ⫺0.734 (⫺0.97)
Negative Zero (continued )
Table 4.14 (continued) Dependent variable: performance adjusted for industry-, size-, and pre–acquisition performance (EBITDA – ΔWC)/BV assets (1)
(2)
(3)
(EBITDA – ΔWC)/Sales (4)
(5)
(6)
(7)
⫺0.001 (⫺0.26)
⫺0.001 (⫺0.18)
⫺0.011 (⫺0.49)
0.000 (0.46)
0.000 (0.47)
0.012 (0.09)
Cross-border M&A (dummy)
0.410 (0.73)
0.020 (0.51)
0.006 (0.17)
0.610* (1.79)
0.342 (1.62)
0.292 (1.52)
Number of deals F-statistic p-value R2
66 0.548 0.833 0.081
101 0.998 0.443 0.080
63 1.134 0.356 0.159
97 0.855 0.558 0.071
104 0.961 0.456 0.056
Relative size of the target
**/** Significant at the 1%/5% level.
107 1.905* 0.087 0.102
109 10.628*** 0.001 0.090
(8)
Expected sign Positive Negative
107 0.859 0.356 0.008
The long-term operating performance in European mergers and acquisitions
113
reports the results of the OLS regressions for different profitability measures and model specifications. Overall, the regression results are consistent with our univariate analysis findings: none of the takeover characteristics have significant power to explain the post-merger profitability of combined firms.18 The intercept is also insignificant in each regression model, regardless of its specification, suggesting that the operating performance does not change significantly following takeovers. Strikingly, a systematic negative relationship exists between pre- and post-acquisition performance: Better performance prior to the takeover is associated with poorer performance after the deal’s completion.
4.6
Conclusion
Although numerous research papers have been written on the stock price performance following mergers and acquisitions, the empirical evidence on changes in post-acquisition operating performance is relatively sparse and their conclusions are inconsistent. The differences in the measurement of profitability and in the benchmarking (the choice of the correct peer companies) is partly responsible for the inconsistency in conclusions across studies. Whereas most of the research focuses on U.S. deals, we have little empirical evidence on the long-term operating performance following European mergers and acquisitions. In this chapter, we investigate the long-term profitability of 155 European corporate takeovers completed between 1997 and 2001, where the acquiring and target companies are from continental Europe and the United Kingdom. We employ four different measures of operating performance that allow us to overcome a number of measurement and statistical limitations of the previous studies and to test the manner in which the conclusions vary across the measures. We find that, in general, profitability of the combined firm decreases significantly following the takeover. However, this decrease becomes insignificant after we control for the performance of the peer companies which are chosen in order to control for industry, size and pre-event performance. This suggests that the decrease is caused by macroeconomic changes unrelated to takeovers. We also find that both acquiring and target companies significantly outperform the median peers in their industry prior to the takeovers. We also reveal that the conclusions regarding changes in post-merger profitability critically depend on the model applied to estimate the changes. Generally, an increase in profitability following M&As is higher when the intercept model is applied, whereas the change model returns lower estimates of the increase in post-acquisition profitability. Our analysis of the determinants of the post-acquisition operating performance reveals that none of the takeover characteristics such as means of payment,
18
Our results do not change when the multivariate analysis comprises dummies instead of continuous variables for pre-acquisition leverage of the acquiring firm, pre-acquisition cash reserves held by the acquirer and relative target size.
114
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
geographical scope, and industry-relatedness have significant explanatory power. However, we find an economically significant difference in the long-term performance of hostile and friendly takeovers, and of tender offers and negotiated deals: the performance deteriorates following hostile bids and tender offers. The acquirer’s leverage prior to takeover seems to have no impact on the post-merger performance of the combined firm, whereas its cash holdings are negatively related to the performance. This suggests that companies with excessive cash holdings suffer from free cash flow problems (Jensen, 1986) and are more likely to make poor acquisitions. Acquisitions of relatively large targets result in better profitability of the combined firm subsequent to the takeover, whereas acquisitions of small target lead to the profitability decline.
References Agrawal, A., Jaffe, J. F., and Mandelker, G., N. (1992). The Post-Merger Performance of Acquiring Firms: A Re-examination of an Anomaly. Journal of Finance, 47:605–1621. Asquith, P., and Mullins, D. W., 1986). Equity Issues and Offering Dilution. Journal of Financial Economics, 15:61–89. Barber, B. M. and Lyon, J. D. (1996). Detecting Abnormal Operating Performance: The Empirical Power and Specification of Test Statistics. Journal of Financial Economics, 41:359–399. Bradley, M., Desai, A. S., and Kim, E. H. (1988). Synergistic Gains from Corporate Acquisitions and Their Division between the Stockholders of Target and Acquiring Firms. Journal of Financial Economics, 21:340. Clark, K., and Ofek, E. (1994). Mergers as a Means of Restructuring Distressed Firms: An Empirical Investigation. Journal of Financial and Quantitative Analysis, 29:541–566. Coffee, Jr., J. C., Lowenstein, L., and Rose-Ackerman, S. (eds.) (1988). Knights, Raiders and Targets. New York: Oxford University Press. Cornett, M., and Sankar, D. (1991). Medium of Payment iIn Corporate Acquisitions: Evidence from Interstate Bank Mergers. Journal of Money, Credit, and Banking, 23:767. Denis, D. J., and Denis, D. K. (1995). Performance Changes following Top Management Dismissals. The Journal of Finance, 50:1029–1057. Dickerson, A. P., Gibson, H. D., and Tsakalotos, E. (1997). The Impact of Acquisitions on Company Performance: Evidence from a Large Panel of U.K. Firms. Oxford Economic Papers, 49:344–361. Eun, C., Kolodny, R., and Scheraga, C. (1996). Cross-border acquisitions and shareholder wealth: tests of the synergy and internalization hypotheses. Journal of Banking and Finance, 20:1559–82. Ghosh, A. (2001). Does Operating Performance Really Improve following Corporate Acquisitions? Journal of Corporate Finance, 7:151–178. Ghosh, A., and Jain, P. J. (2000). Financial Leverage Changes Associated with Corporate Mergers. Journal of Corporate Finance, 6:377–402.
The long-term operating performance in European mergers and acquisitions
115
Ghosh, A., and Ruland, W. (1998). Managerial Ownership, Method of Payments for Acquisitions, and Executive Job Retention. Journal of Finance, 53: 785–798. Goergen, M., and Renneboog, L. (2004). Shareholder Wealth Effects of European Domestic and Cross-Border Takeover Bids. European Financial Management Journal, 10:9–45. Grossman and Hart, (1980). Disclosure law and takeover bids. Journal of Finance, 35:323–34. Gugler, K., Mueller, D. C., Yurtoglu, B. B., and Zulehner, C. (2003). The Effects of Mergers: An International Comparison. International Journal of Industrial Organization, 21:625–653. Harford, J. (1999). Corporate cash reserves and acquisitions. Journal of Finance, 54:1969–1998. Healy, P. J., Palepu, K. G., and Ruback, R. S. (1992). Does Corporate Performance Improve after Mergers? Journal of Financial Economics, 31:135–175. Herman, E., and Lowenstein, L. (1988). The efficiency effects of hostile takeovers. In Knights, Raiders and Targets (Coffee, Jr., J. C., Lowenstein, L., and Rose-Ackerman, S. eds.). New York: Oxford University Press, pp. 211–240. Heron, R., and Lie, E. (2002). Operating Performance and the Method of Payment in Takeovers. Journal of Financial and Quantitative Analysis, 37:137–156. Hogarty, T. F. (1970). The Profitability of Corporate Mergers. The Journal of Business, 33:317–327.Jensen, M. C. (1986). Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review, 76:323–329. Kang, J. K., Shivdasani, A., and Yamada, T. (2000). The Effect of Bank Relations on Investment Decisions: An Investigation of Japanese Takeover Bids. Journal of Finance, 55:2197–2218. Kruse, T. A., Park, H. Y., Park, K., and Suzuki, K. I. (2002). The Value of Corporate Diversification: Evidence from Post-Merger Performance in Japan. AFA 2003 Washington, DC Meetings. Lev, B., and Mandelker, G. (1972). The Microeconomic Consequences of Corporate Mergers. The Journal of Business, 45:85–104. Linn, S. C. and Switzer, J. A. (2001). Are Cash Acquisitions Associated with Better Postcombination Operating Performance Than Stock Acquisitions? Journal of Banking and Finance, 6:1113–1138. Loughran, T., and Vijh, A. (1997). Do Long Term Shareholders Benefit from Corporate Acquisitions? The Journal of Finance, 52:1765–1790. Louis, H. (2004). Earnings Management and the Market Performance of Acquiring Firms. Journal of Financial Economics, 74:121–48. Maloney, M. T., McCormick, R. E., and Mitchell, M. L. (1993). Managerial Decision Making and Capital Structure. Journal of Business, 66:189–217. Martynova, M., and Renneboog, L. (2006a). Takeover Waves: Triggers, Performance and Motives. Working Paper. European Corporate Governance Institute. Martynova, M., and Renneboog, L. (2006b). Mergers and Acquisitions in Europe. In Advances in Corporate Finance and Asset Pricing (Renneboog, L., ed.). Amsterdam: Elsevier.
116
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Masulis, R. W., and Korwar, A. N. (1986). Seasoned Equity Offerings: An Empirical Investigation. Journal of Financial Economics, 28:91–118. Meeks, G. (1977). Disappointing Marriage: A Study of the Gains from Merger. Cambridge: Cambridge University Press. Moeller, S. B., and Schlingemann, F. P. (2005). Are Cross-Border Acquisitions Different from Domestic Acquisitions? Evidence on Stock and Operating Performance for U.S. Acquirers. Journal of Banking and Finance, 29:533–64. Mueller, D. C. (ed.) (1980). The Determinants and Effects of Mergers: An International Comparison. Cambridge, Mass.: Oelgeschlager, Gunn & Hain. Parrino, J. D., and Harris, R., S. (1999). Takeovers, Management Replacement, and Post-Acquisition Operating Performance: Some Evidence from the 1980s. Journal of Applied Corporate Finance, 11:88–97. Powell, R. G., and Stark, A. W. (2005). Does Operating Performance Increase Post-Takeover for UK Takeovers? A Comparison of Performance Measures and Benchmarks. Journal of Corporate Finance, 11:293–317. Rahman, R. A., and Limmack, R. J. (2004). Corporate Acquisitions and the Operating Performance of Malaysian Companies. Journal of Business Finance & Accounting 31:359–400. Rajan, R., Servaes, H., and Zingales, L. (2000), The cost of diversity: the diversification discount and inefficient investment. Journal of Finance, 55:35–80. Ravenscraft, D. J., and Scherer, F. J. (1987). Mergers, Selloffs, and Economic Efficiency. Washington, D.C.: The Brookings Institution. Scharfstein, D., and Stein, J. (2000). The dark side of internal capital markets: divisional rent-seeking and inefficient investment. Journal of Finance, 55:2537–64. Schoenberg, R. (1999). Cultural Compatibility in International Acquisitions, in: Burton, F., Chapman, M. and Cross, A. (eds.), International business organization: Subsidiary management, entry strategies and emerging markets. Academy of International Business series. New York: St. Martin’s Press; London: Macmillan Press, 294–306. Sharma, D. S., and Ho, J. (2002). The Impact of Acquisitions on Operating Performance: Some Australian Evidence. Journal of Business Finance & Accounting, 29:155–200. Shin, H.-H., and Stulz, R. (1998). Are internal capital markets efficient? Quarterly Journal of Economics, 113:513–551. Spiess, D. K., and Affleck-Graves, J. (1995). Underperformance in Long-Run Stock Returns following Seasoned Equity Offerings. Journal of Financial Economics, 38:243–267. Switzer, J. A. (1996). Evidence on Real Gains in Corporate Acquisitions. Journal of Economics and Business, 48:443–460. Yeh, T. M., and Hoshino, Y. (2002). Productivity and Operating Performance of Japanese Merging Firms: Keiretsu-Related and Independent Mergers. Japan and the World Economy, 14:347–366.
5 How do bondholders fare in mergers and acquisitions? Luc Renneboog and Peter G. Szilagyi
Abstract We show firm evidence that bondholders fare better in the stakeholder-oriented governance regimes of Continental Europe, where creditor interests are better represented than in the market-oriented Anglo-American world. Bondholders tend to respond less favorably to cross-border deals. However, they can benefit enormously from spillovers of governance structures and creditor protection that improve the position of creditors vis-à-vis the firm. Both U.S. and European evidence shows that the relative risk profiles of bidding and target firms have a strong impact on bondholder wealth, whether asset risk related to business operations or financial risk related to financing operations is considered. However, almost no evidence supports the earliest theories on the bondholder wealth effects of M&As: that bondholders benefit more from diversifying deals, and that bidding firms reverse bondholder gains by financing their acquisitions with leverage.
5.1
Introduction
While bondholders are among the most important corporate stakeholders, academic research on how mergers and acquisitions (M&As) affect them has been fairly limited. Until recently, only three main hypotheses have been tested in the empirical literature. First, bondholders may benefit more from diversifying deals because the cash flow streams of the merging firms are less well correlated (Levy and Sarnat, 1970). Second, firms may reverse any bondholder gains by paying for acquisitions with borrowed cash, thereby inducing leverage (Dennis and McConnell, 1986). And third, the actual changes in bondholder wealth should be influenced by how the pre-merger risk profiles of bidder and target compare (Shastri, 1990). That the wealth effects of M&As may also vary across countries owing to variations in governance and legal standards was first proposed by Renneboog and Szilagyi (2006). Creditor interests vis-à-vis shareholders are better represented in Continental Europe’s stakeholder-oriented governance regimes than in the AngloAmerican market-oriented world. Thus, the authors argue that bondholders in the former should benefit relatively more from M&As. The authors also observe that cross-border M&As are different from domestic deals, and should benefit
118
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
bondholders less because they introduce greater informational asymmetries as well as legal uncertainties and inefficiencies. However, cross-border deals also provide a platform for interactions between governance and legal systems. Thus, if they expose firms to jurisdictions with better creditor protection, they may even allow creditors to strengthen their legal position. Similar spillovers in governance structures have already been documented in a number of areas, and have been shown to influence the choice of target firms (Rossi and Volpin, 2004), the method of payment (Faccio and Masulis, 2005), and even the valuation of industries where the cross-border deals occur (Bris and Cabolis, 2002). This chapter discusses each of these hypotheses in detail and reviews the empirical literature on how bondholders are in fact affected by M&A activity. Regrettably, empirical evidence is largely confined to U.S. domestic deals, with recent prominent studies including those by Maquieira, Megginson, and Nail (1998) and Billett, King, and Mauer (2004). The only non-U.S. study remains that by Renneboog and Szilagyi (2006), who investigate both domestic and crossborder European M&As. The results show that cross-country variations in governance and legal standards do influence the bondholder wealth effects of M&As. In domestic deals, there is little evidence of bondholder gains for U.S. and U.K. firms, whereas bondholders reap remarkably strong wealth benefits in Continental Europe. Bondholders tend to respond less favorably to cross-border deals, but can benefit enormously from spillovers of governance structures and creditor protection that improve the representation of creditor interests. The relative risk profiles of bidding and target firms also have a strong impact on bondholder wealth, whether asset risk related to business operations or financial risk related to financing operations is considered. Increasingly, however, the oldest hypotheses on the wealth effects of M&As are being challenged; bondholders are sometimes found to benefit less rather than more from diversifying deals, and are largely unaffected by the payment method. The remainder of this paper is organized as follows. Section 5.2 reviews basic theory on how M&As should affect bondholder wealth, and describes the existing U.S. evidence. Section 5.3 explains the potential impact of cross-country variations in governance and legal standards, and discusses the results of Renneboog and Szilagyi’s (2006) European study. Finally, Section 5.4 allows for some concluding remarks.
5.2 5.2.1
The theory and empirics of bondholder wealth in M&As Background
Finance theory suggests that M&As can have many different effects on bondholders. Early studies postulate that bondholders benefit from a co-insurance of cash flows. If two firms with imperfectly correlated cash flow streams merge, their combined cash flow volatility becomes lower, thereby reducing default risk
How do bondholders fare in mergers and acquisitions?
119
and increasing debt capacity (Levy and Sarnat, 1970). The co-insurance effect is likely to be stronger in diversifying or conglomerate mergers in which the merging parties have little or no economic relationship between them. Thus, it is customarily conjectured that bondholders gain more from diversifying than from non-diversifying M&As. However, diversifying mergers tend not to create new wealth, neither providing operating efficiencies nor increasing product or factor market power (Berger and Ofek, 1995). Then, any bondholder gains must come from mere redistributions of shareholder wealth, whereby an increase in bond prices is accompanied by an offsetting reduction in share prices (Higgins and Schall, 1975; Galai and Masulis, 1976). Dennis and McConnell (1986) argue that bidding firms may reverse such wealth shifts by financing their acquisitions with leverage. Cash offers generally require debt financing because most bidders have limited cash and liquid assets (Faccio and Masulis, 2005). Thus, they tend to increase default risk in the combined firm while reducing the collateral available to bondholders. If the bidder offers equity, no assets leave the firm, and financial distress costs are reduced. Ultimately, this suggests that bondholders benefit more from equity-financed acquisitions. Still, we cannot discount the agency and signaling effects associated with equity financing. In the spirit of Myers and Majluf (1984), DeAngelo, DeAngelo, and Rice (1984) point out that the managers of a bidding firm prefer to make an equity offer if they believe that their firm is overvalued. If the market interprets an equity offer as bad news on the firm’s future expected cash flows, as Mitchell and Stafford (2000) indeed find, bondholder sentiment may deteriorate. It is notable that the preceding conjectures intuitively separate asset risk effects associated with business operations and financial risk effects associated with financing operations. From the bondholders’ perspective, this distinction is formalized by Shastri (1990). The author derives predictions for the risk effects of M&As by comparing the pre-merger risk profiles of the bidder and the target. Asset risk in the combined firm can differ from the asset risks of the merging parties because they have different levels of asset risk to start with, and/or because their unleveraged stock returns are imperfectly correlated. Overall, a reduction in asset risk should increase, while an increase in asset risk should decrease bondholder wealth. The impact of the asset risk change depends not only on the size of the risk change, but also on the pre-merger risk of debt. Thus, relatively risky bonds should benefit the most from a risk reduction, and relatively safe bonds should lose the most from a risk increase. Shastri (1990) relates financial risk effects specifically to leverage. Obviously, other factors also contribute to the risks associated with financing operations. For example, interest coverage better captures the immediate probability of default. Whatever the measure used, financial risk in the combined firm will differ from the financial risks of the merging parties unless they are identical premerger. Then, a reduction in financial risk should increase, while an increase in financial risk should decrease bondholder wealth. Of course, this financial risk effect does not account for expected risk changes arising from post-merger financing operations or as a result of the payment method.
120
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Although the relative size of bidding and target firms has no direct risk implications, it may still have an indirect impact by affecting projected efficiency gains that influence the combined firm’s ability to service its fixed debt obligations. On the one hand, larger targets may induce a greater co-insurance of cash flows and contribute more assets to the combined firm, thus adding debt capacity (Hovakimian, Opler, and Titman 2001). On the other hand, the absorption capacity of bidding firms should be limited. Large deals are hard to implement successfully; thus the efficiency gains associated with the acquisition of smaller targets should be relatively larger (Bhagat, Dong, Hirshleifer, and Noah, 2005). Target bondholders may also gain more when the bidding firm is relatively large, to the extent that large bidders are generally more diversified; hence their credit risk tends to be lower at a given leverage ratio (Faccio and Masulis, 2005).
5.2.2
Empirical evidence from U.S. domestic deals
Although M&As have been among the hottest topics of academic research, empirical evidence on how they affect bondholders is relatively sparse and remains almost exclusively restricted to the United States. The results of the existing U.S. literature, summarized in Table 5.1, are varied but do suggest that the bondholders of U.S. firms do not benefit from M&As at all, except under specific conditions. The early studies took a fairly narrow approach in identifying the drivers of bondholder wealth changes. Kim and McConnell (1977) and Asquith and Kim (1982) considered completed diversifying deals only to examine the presence of co-insurance effects. Neither study shows evidence of statistically significant bondholder gains; Kim and McConnell (1977) find insignificantly negative, while Asquith and Kim (1982) report insignificantly positive, abnormal returns for both bidding and target firms. Dennis and McConnell (1986) were the first to consider that the merging firms’ pre-merger risk profiles, as measured by Standard and Poor’s credit ratings, can have an impact on the wealth changes their bondholders incur. The authors stop short of testing for differences in the abnormal bond returns of junk-grade versus investment-grade firms. Nonetheless, they report marginal evidence that the bondholders of junk-grade targets indeed incur higher wealth gains, though the gains themselves fall short of statistical significance. The authors find that bidder bondholders tend to lose rather than gain from M&As. Surprisingly, however, it is the bondholders of junk-grade bidders who suffer significant losses, a finding that is inconsistent with Shastri’s (1990) argument that relatively safe bonds should lose the most from a risk increase. That bidding firms can use leveraged financing to reverse any wealth shifts from shareholders to bondholders, is already acknowledged, but not tested, by Dennis and McConnell (1986). Eger (1983) omitted any effect the payment method may have by focusing on pure stock-for-stock M&As. Remarkably, she found that bidder bondholders earn highly significant abnormal gains in these deals. Subsequent studies failed to find unanimous support for this result, however.
Table 5.1 The bondholder wealth effects of US M&Asa Study
Sample period
Deal type
Event window
Merging party
Mean change
Kim and McConnell (1977)
1960–1973
Completed conglomerate
[0] month
All
⫺0.45
44
Market model using equal-weighted bond index and the NYSE stock index
Asquith and 1960–1978 Kim (1982)
Completed conglomerate
[0] month
All Bidder Target
1.07 1.08 1.05
62 38 24
Bond matched by rating, maturity, coupon, and industry
Eger (1983)
1958–1980
Completed Stock-for-stock
[⫺30,0] days
Bidder
1.01***
33
Bond portfolio matched by rating and maturity
Dennis and McConnell (1986)
1962–1980
Completed
[⫺1,0] days
Bidder BBB or below Target BBB or below
⫺0.17 ⫺0.51* 0.03 0.35
67 31 27 19
Dow Jones Industrial Bond Index
Walker (1994)
1980–1988
Completed
[0] month
All Target All All
0.31 0.83 ⫺0.73 1.39
92 33 35 12
U.S. Treasury matched by duration
Cash-for-stock Stock-for-stock
N
Benchmark/ methodology
(continued)
Table 5.1 (continued) Study
Sample period Deal type
Event window
Merging party
Maquieira, 1963–1996 Megginson, and Nail (1998)
Diversifying, [⫺2,2] months completed, stock-for-stock Nondiversifying, completed, stock-for-stock
All Bidder Target All Bidder Target
Billett, King, 1979–1997 and Mauer (2004)
All mergers
Bidder (i) Junk-grade Investment-grade (ii) Rated lower than target Rated higher than target (iii) Asset risk reduced Asset risk increased (iv) Leverage reduced Leverage increased (v) Relative size ⬍ median
[⫺1,0] months
Mean change
N
Benchmark/ methodology
0.44 0.33 1.22 1.44*** 1.90*** 0.50
253 222 31 282 189 93
U.S. Treasury matched by maturity and coupon, using mean valuation prediction errors (VPE)
⫺0.17*** ⫺0.55* ⫺0.09* ⫺0.74**
831 151 680 48
Lehman Brothers Corporate Bond Index matched by rating and maturity
⫺0.22
112
⫺0.20**
583
⫺0.16
89
⫺0.25**
420
⫺0.17
310
⫺0.03
415
a
Relative size ⬎ median
⫺0.32***
416
Target (i) Junk-grade Investment-grade (ii) Rated lower than bidder Rated higher than bidder (iii) Asset risk reduced Asset risk increased (iv) Leverage reduced Leverage increased (v) Relative size ⬍ median Relative size ⬎ median
1.09** 4.30*** ⫺0.80*** 1.64
265 98 167 96
⫺1.19***
65
1.53**
202
⫺1.56**
25
2.00**
136
0.04 2.79*** ⫺0.61**
94 132 133
This table shows the estimated abnormal bond returns of firms involved in M&As. Abnormal returns are in percentages and are calculated using the event study methodologies described in the table. N is the number of firms; where a firm has multiple bonds outstanding, it is treated as a value-weighted portfolio of its bonds. ***/**/* indicate significance at the 1, 5 and 10% level, respectively.
124
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Walker (1994) compared cash-for-stock and stock-for-stock M&As, and found that while bidder bondholders do benefit more from stock-for-stock deals, their wealth gains are statistically insignificant. Later, Maquieira, Megginson, and Nail (1998) reexamined pure stock-for-stock deals. The authors compared diversifying and nondiversifying deals to determine whether bondholders benefit more from wealth redistributions from shareholders to bondholders or simple synergistic wealth creation. The results play down the role of co-insurance effects arising from diversification, suggesting that bondholders are unlikely to benefit from M&As unless they are synergistic. In fact, the authors showed that only bidder bondholders gain even in nondiversifying deals; in all other cases, abnormal bond returns are positive but statistically insignificant in both bidding and target firms. Walker (1994) qualified his findings further by performing a multivariate analysis of bondholder wealth changes. He reported strong evidence that both bidder and target bondholders earn higher abnormal returns when their firm is junk-grade, thus indeed benefiting relatively more from a risk reduction. The author found that the bondholders of targets, but not bidders, are also sensitive to financial risk changes in the combined firm as measured by changes in leverage. Otherwise, he found no indication that bondholders would respond better to either diversifying or nondiversifying deals, or that covenants included in the bond contract would be an important determinant of abnormal bond returns. The drivers of bondholder wealth changes are revisited by Billett, King, and Mauer (2004), using a vast, 940-strong sample of U.S. M&As. Remarkably, the results reported in this study often represent a considerable departure from previous findings. These authors show strong evidence that the bondholders of junk-grade targets tend to gain from takeover bids, but the bondholders of bidders and investment-grade targets tend to suffer considerable losses. Shastri’s (1990) predictions on how the relative pre-merger risk profiles of bidder and target affect the risk effects of M&As are supported for targets, but not for bidders. Target bondholders benefit considerably more when the overall risk profile of the bidding firm, as captured by its credit rating assigned by Standard and Poor’s, is superior. They respond to changes in asset risk in particular, which Billett, King, and Mauer (2004) measure by the standard deviation of unleveraged stock returns; changes in financial risk, measured by leverage, affects only bondholders in junk-grade targets. Billett, King, and Mauer (2004) also reinvestigate whether abnormal bond returns do actually differ in diversifying versus nondiversifying, and cash- versus stock-financed deals. The authors find that bondholders do not respond to either consideration. On the other hand, they show for the first time that the merging firms’ relative size is an imperative determinant; the larger the target relative to the bidder, the lower the abnormal bond returns in both firms. The authors also consider the implications of other deal characteristics such as type, attitude, status, and timing. They find that both bidder and target bonds fare better when the deal is classified as friendly (versus hostile), and when it occurs in the 1990s, coinciding with an increase in the incidence of event risk covenants. In investment-grade
How do bondholders fare in mergers and acquisitions?
125
bidders and targets, bondholders also benefit more when the bidder approaches the target by negotiations with management rather than through a tender offer. Finally, some evidence supports the notion that target bonds perform better at the announcement than when the deal is completed.
5.3 5.3.1
Do cross-country differences in governance and legal standards matter? Background
The relevance of governance and legal standards That the economic implications of M&As can vary considerably across countries with different governance and legal standards is well-documented in the corporate governance literature. As a recent example, Goergen and Renneboog (2004), who investigate how European deals affect shareholder wealth, find that abnormal stock returns are strongly influenced by whether the merging firms are based in a common-law (United Kingdom) or a civil-law (Continental European country). From the perspective of bondholders, a similar distinction between commonand civil-law countries is equally warranted. In the common-law AngloAmerican countries, strong shareholder rights vis-à-vis managers and stringent disclosure requirements encouraged the emergence of market-oriented corporate governance systems. These regimes basically view creditors and other stakeholders as independent parties that maintain arm’s-length contractual arrangements with the firm (Jensen and Meckling, 1976). In the civil law-based, more stakeholder-oriented governance systems of Continental Europe and Japan, the dynamics of the firm–creditor relationship are very different. Banks act as concentrated lenders and delegated monitors, playing a key role in mitigating informational asymmetries and agency problems (Diamond, 1991), and reducing the marginal utility of external market mechanisms. Other stakeholders also develop long-term relationships with the firm, and closely held equity and pyramidlike group memberships are in place. The greater influence of banks and other risk-averse stakeholders on corporate decision-making dictates that bondholders in stakeholder-oriented governance regimes should benefit more from M&As. Conflicts of interest cannot be ruled out between senior banks and junior bondholders, especially if bondholder claims are unsecured (La Porta, Lopez-deSilanes, Shleifer, and Vishny, 1998). Nonetheless, close bank monitoring should prevent managers from excessive risk-taking, which by itself should make M&As more bondholder-friendly. La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998) find that countries of the same legal origin also bear some resemblance to the extent that they offer regulatory protection to creditors. A notable observation is that, on average, common-law countries are more protective of creditor rights than are civillaw—especially French civil-law—countries. Still, enormous variation persists,
126
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
even within particular families of legal origin. For example, English insolvency law strictly enforces creditor rights, whereas the softer U.S. approach puts them under judicial discretion (Sussman, 2005). La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998) report similar differences within the French legal family, such as those between the more pro-creditor Netherlands and the more pro-debtor France. The extent to which creditors enjoy regulatory protection should also influence managerial risk-taking regardless of the legal and governance regime in place. These factors show that the representation of bondholder interests in M&As is likely to be affected by a variety of institutional conditions, and suggest that the results on how U.S. deals affect bondholders can be extended to other countries only to a limited extent.
The interaction of governance and legal standards in cross-border deals Cross-country differences in governance and legal standards can have particularly strong economic implications in cross-border M&As. That cross-border deals provide a platform for interactions between governance systems has been well documented in the recent literature: Considerable spillovers have been shown to occur in governance structures, accounting standards and disclosure practices. The evidence suggests that such considerations affect shareholder returns (Moeller and Schlingemann, 2005), also influencing the choice of target firms (Rossi and Volpin, 2004), the method of payment (Faccio and Masulis, 2005), and even the valuation of industries where the cross-border deals occur (Bris and Cabolis, 2002). From the perspective of bondholders, cross-border M&As exhibit some distinctive peculiarities relative to domestic deals. Denis, Denis, and Yost (2002) draw a parallel between global and industrial diversification, observing that the two induce a similar diversification discount in share prices. Accordingly, Moeller and Schlingemann (2005) find that U.S. firms that acquire cross-border targets achieve lower abnormal stock returns and fewer improvements in operating performance. In the spirit of these findings, the implications of crossborder takeovers for bondholders are twofold. On the one hand, the cash flow streams of bidder and target are likely to be less well correlated in cross-border deals; thus bondholders should benefit—and shareholders should lose—from reduced cash flow volatility. On the other hand, even if the projected efficiency gains are considerable, capturing them is complicated. In cross-border M&As, informational asymmetries are greater and corporate culture may clash. Bondholders may also suffer directly from the added legal uncertainty and inefficiency associated with the default of an internationally diversified firm. If a cross-border deal involves firms from countries with different governance regimes, the representation of creditor interests in the combined entity may be altered considerably. Like governance structures, accounting standards, and disclosure practices, the strong influence of creditors can be passed on, but the quality of the transition will depend on the change in the relative power and monitoring incentives of banks and other stakeholders. Intuitively, cross-regime
How do bondholders fare in mergers and acquisitions?
127
M&As may import creditor influence from the perspective of one firm while diluting creditor influence from the perspective of another. Thus, the bondholders of a common-law firm should gain more from a merger with a civil-law firm, while the bondholders of a civil-law firm should gain less from a merger with a commonlaw firm. For bondholders, another critical issue involves how creditor rights and their enforcement compare in the national jurisdictions of the merging firms. La Porta, Lopez-de-Silanes, Shleifer, and Vishny (2000) argue that functional spillover of creditor protection have limitations, because corporate assets remain under the jurisdiction of the country where they are located. However, the exposure of a firm to multiple insolvency jurisdictions facilitates legal arbitrage by its creditors through jurisdiction (or forum) shopping, whereby they race to choose a supposedly friendly jurisdiction to strengthen their legal position and obtain maximum satisfaction for their claims. The BIS (2002) points out that jurisdiction shopping may, in general, be more relevant for common-law jurisdictions, which follow the “incorporation doctrine,” than for civil-law jurisdictions, which adhere to the “real seat doctrine.” The United States is cited as an example, in which jurisdiction shopping is a well known phenomenon, thus explaining the popularity of pro-creditor bankruptcy courts in Delaware and New York. While jurisdiction shopping is less prevalent across civil law-based legal regimes, it is often facilitated by international treaties on cross-border insolvency, such as the Nordic Bankruptcy Convention of 1933 and the Montevideo and Bustamente Conventions in South America. The European Union (EU) has also pioneered cooperation among its national jurisdictions in insolvency proceedings and harmonized conflict of law issues. In marked contrast with the United States, insolvency law in the European Union is still a matter of national law, and this territoriality principle was indeed the only principle regulating cross-border insolvencies until 2000. In that year, however, the European Union passed the European Insolvency Regulation (EIR),1 introducing a modified form of the universalism principle into cross-border insolvency proceedings. The essence of the EIR is that it puts one main jurisdiction in charge of insolvency proceedings. It does not, however, clearly specify how the main jurisdiction should be selected, allowing creditors some discretion in choosing where to petition for bankruptcy proceedings (Franken, 2005). It is possible that the debtor pre-empts creditors in filing for bankruptcy under the universalism rule in a national jurisdiction that is not advantageous to creditors. However, when this latter case arises, creditors may still open secondary or territoriality-based proceedings in other jurisdictions provided that the firm has assets in those countries (Enriques and Gelter, 2006). Overall, bondholders can clearly benefit from improved creditor protection brought about by cross-border M&As for three reasons. First, new exposure to a jurisdiction with better creditor rights is likely to force management to avoid excessive risk-taking. Second, a cross-border acquisition (of substantial foreign 1
European Council Regulation No. 1346/2000 on insolvency proceedings.
128
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
operations) may lead to jurisdiction shopping in case of insolvency such that the best possible law from the perspective of the creditors may apply. And third, in case the jurisdiction adopted does not maximize creditor rights, the creditors may still open territoriality-based proceedings whereby all creditors (including those from the jurisdiction with the worse creditor rights) would also have a claim on the assets of the country that protects creditor rights best. Thus, if a crossborder takeover allows access to a jurisdiction with better creditor rights (through the principle of universalism or territoriality), it should induce relatively higher bondholder wealth gains for the bondholders in the country with the lower creditor protection.2 Of course, an improvement in creditor protection is also conditional on the general effectiveness and predictability of the judiciary and the enforcement of debt contracts.
5.3.2
Bondholder wealth in European M&As: the impact and spillover of governance and legal standards
Renneboog and Szilagyi (2006) are the first to investigate how cross-country variations in institutional conditions influence bondholder wealth changes around M&A announcements. The authors examine the wealth effects of 238 domestic and cross-border European deals launched between 1995 and 2004 using euro- and sterling-denominated Eurobonds. The use of Eurobonds rather than domestic bonds is a notable break with the approach of the prior U.S. literature, and inherently restricts the scope of investigation to investment-grade issuers. However, Gabbi and Sironi (2005) argue that only Eurobonds should be used for such a cross-country study because they are highly standardized and thus much more comparable across borrowers of different nationalities.3 Renneboog and Szilagyi (2006) also point out that Eurobonds constitute the only European corporate bond market that is sufficiently large and liquid to allow for the construction of robust pricing benchmarks. 2
In the special cases where the governing law of the debt contract is expressly specified, as is the case with Eurobonds, jurisdiction shopping is not directly applicable. Nonetheless, the improved bargaining power of the other creditors should still put pressure on management to moderate risk-taking even at the expense of shareholder value, thus also benefiting the bondholders (Esho, Sharpe, and Tchou, 2004). 3 Eurobonds are mostly unsecured and carry fewer covenants than do domestic bonds. This explains why most Eurobonds are de facto of high credit quality; investors are unwilling to accept such illprotected securities from low-quality borrowers. Eurobonds also tend to be issued in relatively large amounts, typically in bearer form, and are normally exempt from withholding tax. This ensures that they are held predominantly by institutional investors, making the Eurobond market highly liquid and competitive, and ensuring efficiency and a minimal risk of price anomalies. Finally, Eurobonds are typically governed by English law and listed on the Luxembourg Stock Exchange. English law is generally preferred because it permits the inclusion of collective action clauses in the bond contract, and allows greater scope for the bond trustee to negotiate with the issuer (Smith and Walter, 1997). The Luxembourg Stock Exchange was among the first to relax Eurobond issuing procedures in 1990, and is preferred because it has low fees, no withholding tax, and quickly approves new listings.
How do bondholders fare in mergers and acquisitions?
129
The empirical part of the study mainly focuses on bidding firms, because the size of the target sample is limited, with only 24 observations. Table 5.2 summarizes some of the more important univariate results. The first remarkable finding is that acquisitions by European bidding firms are generally more bondholder-friendly than are U.S. domestic deals. Bidder bondholders earn wealth gains that are highly significant, both statistically and economically. In target firms, bondholder returns are also positive but insignificantly different from zero. The results in Table 5.2 show only marginal evidence that bidder bondholders benefit less from cross-border M&As, and suggest that these bondholders accrue economically significant wealth gains both in civil-law Continental European countries and the common-law United Kingdom. However, further stratification of the sample reveals that the main source of bondholder gains tends to be domestic deals in Continental Europe and cross-border deals in the United Kingdom. The lower cross-border returns in Continental Europe are likely to be motivated by governance concerns. In stakeholder-oriented governance regimes, creditors and other stakeholders enjoy considerable authority vis-à-vis the firm, and a cross-border acquisition may deteriorate their strong domestic position (Köke and Renneboog, 2005). In the United Kingdom, bidder bondholders seem to benefit from domestic deals not at all, a finding that corresponds well with the empirical evidence reported for the United States. Conversely, they reap substantial wealth gains from cross-border deals where the target country’s legal regime is civil law-based, and especially when it offers a relative improvement in creditor rights. That bidder bondholders generally benefit from improved creditor protection in cross-border deals is demonstrated in Table 5.2. Renneboog and Szilagyi (2006) verify this phenomenon using a self-constructed creditor rights index based on La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998), and the credit contract enforcement index of Djankov, McLiesh, and Shleifer (2004). They find that bidder bondholders benefit from cross-border M&As only when the deal provides access to a jurisdiction with better creditor rights. Improvements in the efficiency of credit contract enforcement, measured as the number of days needed to resolve claims disputes, induce a similar but more muted impact. That the source of these bondholder gains is a simultaneous reduction in shareholder wealth is also confirmed by the authors in their analysis of abnormal stock returns. As in the prior U.S. literature, Renneboog and Szilagyi (2006) find no evidence that bidder bondholders benefit more from diversifying M&As, or that they are adversely affected if a deal is cash-financed. Rather, they respond to the relative characteristics of the target firm, and the effect these have on the risk profile of the combined entity. As in Billett, King, and Mauer (2004), abnormal bond returns are sensitive to asset risk considerations in particular; they are significantly positive when asset risk is lower in the combined firm and insignificantly negative when it is higher. Changes in financial risk, measured either by leverage or interest coverage, have a similar but less pronounced impact. The size of the target firm is another key driver of bondholder wealth changes, in that the
130
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 5.2 The bondholder wealth effects of European M&Asa Grouping criteria Bidding firms Reference: Target firms
Mean change
N
0.56*** 0.62
225 24
Governance characteristics Geographic focus
Domestic Cross-border Difference
0.84*** 0.41** ⫺0.43
79 146
Legal regime of bidder country
Civil law Common law Difference
0.48*** 0.71*** 0.23
146 79
Target country scores better than bidder country in cross-border deals in: Creditor rights No 0.12 Yes 0.88*** Difference 0.77*
70 37
Credit contract enforcement
0.22 0.72** 0.50
89 53
0.55*** 0.58** 0.04
152 73
0.61*** 0.34 ⫺0.28
142 23
No Yes Difference Deal characteristics
Industry focus
Nondiversifying deal Diversifying deal Difference
Payment method
Cash financing Equity or mixed financing Difference Firm characteristics
Asset risk
Lower in combined firm Higher in combined firm Difference
0.74*** ⫺0.71 ⫺1.45**
51 17
Leverage
Lower in combined firm Higher in combined firm Difference
0.59** 0.07 ⫺0.51
44 27
Target/bidder relative size
Smaller than sample median Greater than sample median Difference
0.85** ⫺0.07 ⫺0.93**
36 35
a
This table shows the estimated abnormal Eurobond returns of firms involved in European M&As during 1995–2004. Abnormal bond returns are defined as the sum of monthly abnormal returns in the two months surrounding the deal announcement (i.e., months –1 and 0) and are expressed in percentages. The abnormal monthly returns are calculated using equal-weighted benchmark bond portfolios matched by currency (euro and sterling), credit rating (BBB, A, AA and AAA) and duration (1–3, 3–5, 5–7, 7–10 and 10⫹ years). N is the number of firms; where a firm has multiple Eurobonds outstanding, it is treated as a value-weighted portfolio of its bonds, with the weights based on the market value of each bond 2 months before the announcement. The difference in means t-test assumes unequal variances across groups when a test of equal variances is rejected at the 10% level. ***/**/* indicate significance at the 1, 5, and 10% level, respectively. Source: Renneboog and Szilagyi (2006).
How do bondholders fare in mergers and acquisitions?
131
larger the target relative to the bidder, the smaller the gains that bidder bondholders accrue. Though not reported in Table 5.2, the authors also perform multivariate regressions to corroborate their univariate findings. The results provide even stronger evidence that cross-country variations in governance and legal standards affect bondholder wealth changes in M&As. The regressions show that ceteris paribus, bidder abnormal bond returns are considerably higher when the bidding firm is from Continental Europe, the deal is domestic, and when the deal is cross-border but the target country offers better creditor rights and enforces credit contracts more efficiently. The results also confirm that bidder bondholders respond to changes in both asset risk and financial risk, and benefit less if the target firm is relatively large. The authors use the same multivariate framework to analyze the abnormal returns incurred by target firm bondholders. The scope of their investigation is limited owing to the small number of observations. However, the regressions still provide evidence that target abnormal bond returns are higher when the target firm is from Continental Europe, and when the bidder is based in a country with better creditor rights. Target bondholders also respond unfavorably to increases in asset risk, but the impact of financial risk changes is more ambiguous. The study concludes by investigating whether such deal characteristics as type, attitude, status, and timing have wealth implications for bidder and target bondholders. Contrary to Billett, King, and Mauer (2004), the results suggest that bidder bondholders actually benefit less when the target is approached through negotiations with management rather than by a tender offer. The authors also document for the first time that bidder bondholders earn higher returns when the target firm is privately held rather than publicly listed. A similar listing effect has been previously documented for shareholder returns, but the academic literature has yet to explain why, despite several recent research efforts (Faccio, McConnell, and Stolin, 2006).
5.4
Conclusion
This chapter has contributed to the comparative corporate governance literature by reviewing the empirical evidence on how bondholder wealth is affected by M&As. The focus of existing studies is largely confined to U.S. domestic deals, and they show very limited evidence that either bidding or target bondholders would benefit from M&As at all (Eger, 1983; Maquieira, Megginson, and Nail, 1998; Billett, King, and Mauer, 2004). Renneboog and Szilagyi (2006) are the first to examine whether this result also holds for cross-border M&As and across European countries with different governance and legal regimes. The authors show firm evidence that bondholders fare better in the stakeholderoriented governance regimes of Continental Europe, where creditor interests are better represented than in the market-oriented Anglo-American world. Bondholders tend to respond less favorably to cross-border deals. However,
132
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
they can benefit enormously from spillovers of governance structures and creditor protection that improve the position of creditors vis-à-vis the firm. Both U.S. and European evidence shows that the relative risk profiles of bidding and target firms have a strong impact on bondholder wealth, whether asset risk related to business operations or financial risk related to financing operations is considered. However, almost no evidence supports the earliest theories on the bondholder wealth effects of M&As: that bondholders benefit more from diversifying deals, and that bidding firms reverse bondholder gains by financing their acquisitions with leverage. The remarkable conclusion we can draw from the literature is that crosscountry differences in governance and legal standards induce variations as great those in the bondholder wealth effects of M&As as any firm- or deal-level characteristic. This shows that the corporate governance literature must continue to test its findings in an international context. United States studies on the implications of M&As should provide reasonable guidance for research on other countries and governance regimes; however, their results are unlikely to apply unconditionally, underlining the strong need for comparative studies.
References Asquith, P., and Kim, E. H. (1982). The Impact of Merger Bids on the Participating Firms’ Security Holders. Journal of Finance, 37:1209–1228. Berger, P. G., and Ofek, E. (1995). Diversification’s Effect on Firm Value. Journal of Financial Economics, 37:39–65. Bhagat, S., Dong, M., Hirshleifer, D., and Noah, R. (2005). Do Tender Offers Create Value? New Methods and Evidence. Journal of Financial Economics, 76:3–60. Billett, M. T., King, T.-H. D., and Mauer, D. C. (2004). Bondholder Wealth Effects in Mergers and Acquisitions: New Evidence from the 1980s and 1990s. Journal of Finance, 59:107–135. BIS (2002). Insolvency Arrangements and Contract Enforceability: Report of the Contact Group on the Legal and Institutional Underpinnings of the International Financial System. Basel: Bank for International Settlements. Bris, A., and Cabolis, C. (2002). Corporate Governance Convergence by Contract: Evidence from Cross-Border Mergers. Working Paper. Yale International Center for Finance. DeAngelo, H., DeAngelo, L., and Rice, E. M. (1984). Going Private: Minority Freezeouts and Stockholder Wealth. Journal of Law and Economics, 27:367–401. Denis, D. J., Denis, D. K., and Yost, K. (2002). Global Diversification, Industrial Diversification, and Firm Value. Journal of Finance, 57:1951–1979. Dennis, D. K., and McConnell, J. J. (1986). Corporate Mergers and Security Returns. Journal of Financial Economics, 16:143–187.
How do bondholders fare in mergers and acquisitions?
133
Diamond, D. W. (1991). Monitoring and Reputation: The Choice between Bank Loans and Directly Placed Debt. Journal of Political Economy, 99:689–722. Djankov, S., McLiesh, C., and Shleifer, A. (2004). Private Credit in 129 Countries. Working Paper. National Bureau of Economic Research. Eger, C. E. (1983). An Empirical Test of the Redistribution Effect in Pure Exchange Mergers. Journal of Financial and Quantitative Analysis, 18:547–572. Enriques, L., and Gelter, M. (2006). How the Old World Encountered the New One: Regulatory Competition and Cooperation in European Corporate and Bankruptcy Law. Law Working Paper. European Corporate Governance Institute. Esho, N., Sharpe, I. G., and Tchou, N. (2004). Moral Hazard and Collective Action Clauses in the Eurobond Market. Working Paper. University of New South Wales. Faccio, M., and Masulis, R. W. (2005). The Choice of Payment Method in European Mergers and Acquisitions. Journal of Finance, 60:1345–1388. Faccio, M., McConnell, J. J., and Stolin, D. (2006). Returns to Acquirers of Listed and Unlisted Targets. Journal of Financial and Quantitative Analysis, 41:197–220. Franken, S. (2005). Creditor- and Debtor-Oriented Corporate Bankruptcy Regimes Revisited. European Business Organization Law Review, 5:645–676. Fuller, K., Netter, J. M., and Stegemoller, M. (2002). What Do Returns to Acquiring Firms Tell Us? Evidence from Firms That Make Many Acquisitions. Journal of Finance, 57:1763–1793. Gabbi, G., and Sironi, A. (2005). Which Factors Affect Corporate Bonds Pricing? Empirical Evidence from Eurobonds Primary Market Spreads. European Journal of Finance, 11:59–74. Galai, D., and Masulis, R. W. (1976). The Option Pricing Model and the Risk Factor of Stock. Journal of Financial Economics, 3:53–81. Goergen, M., and Renneboog, L. (2004). Shareholder Wealth Effects of European Domestic and Cross-Border Takeover Bids. European Financial Management, 10:9–45. Higgins, R. C., and Schall, L. D. (1975). Corporate Bankruptcy and Conglomerate Merger. Journal of Finance, 30:93–111. Hovakimian, A., Opler, T., and Titman, S. (2001). The Debt–Equity Choice. Journal of Financial and Quantitative Analysis, 36:1–24. Jensen, M. C., and Meckling, W. H. (1976). Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics, 3:305–360. Kim, E. H., and McConnell, J. J. (1977). Corporate Mergers and the Co-insurance of Corporate Debt. Journal of Finance, 32:349–365. Köke, J. F., and Renneboog, L. (2005). Do Corporate Control and Product Market Competition Lead to Stronger Productivity Growth? Evidence from Market-Oriented and Blockholder-Based Governance Regimes. Journal of Law and Economics, 48:475–516.
134
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R. W. (2000). Investor Protection and Corporate Governance. Journal of Financial Economics, 58:3–27. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R. W. (1998). Law and Finance. Journal of Political Economy, 106:1113–1155. Levy, H., and Sarnat, M. (1970). Diversification, Portfolio Analysis and the Uneasy Case for Conglomerate Mergers. Journal of Finance, 25:795–802. Maquieira, C. P., Megginson, W., and Nail, L. (1998). Wealth Creation versus Wealth Redistributions in Pure Stock-for-Stock Mergers. Journal of Financial Economics, 48:3–33. Mitchell, M. L., and Stafford, E. (2000). Managerial Decisions and Long-Term Stock Price Performance. Journal of Business, 73:287–329. Moeller, S. B., and Schlingemann, F. P. (2005). Global Diversification and Bidder Gains: A Comparison between Cross-Border and Domestic Acquisitions. Journal of Banking and Finance, 29:533–564. Myers, S. C., and Majluf, N. S. (1984). Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have. Journal of Financial Economics, 13:187–221. Renneboog, L., and Szilagyi, P. G. (2006). How Do Mergers Affect Bondholders in Europe? Evidence on the Impact and Spillover of Governance and Legal Standards. Working Paper. European Corporate Governance Institute. Rossi, S., and Volpin, P. F. (2004). Cross-Country Determinants of Mergers and Acquisitions. Journal of Financial Economics, 74:277–304. Shastri, K. (1990). The Differential Effects of Mergers on Corporate Security Values. Research in Finance, 8:179–201. Smith, R. C., and Walter, I. (1997). Risks and Rewards in Emerging Market Investment. Working Paper. New York University. Sussman, O. (2005). The Economics of the EU’s Corporate-Insolvency Law and the Quest for Harmonisation by Market Forces. Working Paper. Oxford Financial Research Centre. Walker, M. M. (1994). Determinants of Bondholder Wealth Following Corporate Takeovers (1980–1988). Quarterly Journal of Business and Economics, 33:12–29.
6 Mix-and-match facilities and loan notes in acquisitions1
Marc Goergen and Jane Frecknall-Hughes
Abstract This chapter deals with acquisitions that offer target shareholders a choice of different types of consideration, including the possibility of mixing and matching these considerations. It also discusses why bidders may want to issue loan notes and why target shareholders may want to take up loan notes rather than cash or shares. The study presented in this chapter provides three major contributions to the literature. First, contrary to existing studies, we do not take the eventual payment for an acquisition as a given, but rather take into account the choice of different types of considerations offered to the target shareholders. Second, whereas most of the existing literature concentrates on the bidder’s point of view, we focus on the target shareholders’ point of view. Third, in contrast with most other studies, which consider loan notes to be equivalent to cash, we clearly distinguish between the two. We make the following conclusions. First, the popularity of loan notes over the late 1990s and early years of the 21st century has increased. Second, contrary to the suggestions of theories on asymmetric information and findings from empirical studies on wealth gains, target shareholders do not always choose (exclusively) cash instead of shares. Third, loan notes issues offer clear benefits to both the target shareholders and bidders. More precisely, using the tax models developed, we show that, in certain circumstances, the tax position of target shareholders may make loan notes a more attractive choice than other types of consideration. Furthermore, bidders derive benefits from loan notes in the form of improved cash management and cheap financing of acquisitions.
6.1
Introduction
This chapter deals with acquisitions that offer target shareholders a choice of different types of consideration including—in some cases—the possibility of mixing and matching these considerations. It also discusses why bidders may want to issue loan notes and why target shareholders may want to take up loan notes rather than cash or shares. The pilot study presented in this chapter provides three major contributions to the literature. First, contrary to existing studies, we do not take the eventual payment for an acquisition as a given, but we 1
We are grateful to Luc Renneboog and Suzette Rouch-Cordot for comments and suggestions.
136
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
take into account the choice of different types of considerations offered to the target shareholders. Second, whereas most of the existing literature concentrates on the bidder’s point of view, we focus on the target shareholders’ point of view. Third, unlike most other studies, which treat loan notes as equivalent to cash, we clearly distinguish between the two. The remainder of the chapter is organized as follows. Section 6.2 reviews the limited literature on the different types of consideration offered in mergers and acquisitions (M&As) and their impact on wealth effects for target and bidder shareholders. Section 6.3 discloses the data sources and the sample selection criteria. Section 6.4 describes the characteristics of acquisition deals that offer a choice of considerations. The next section, Section 6.5, discusses the advantages of loan notes both from the target’s perspective and from the bidder’s. Section 6.6 deals with the accounting treatment and the disclosure of loan notes. The often complex accounting treatment of loan notes may explain why previous studies have used the shortcut of treating these as equivalent to cash. Section 6.7 explains how loan notes may be appealing to shareholders with particular tax positions. Finally, Section 6.8 summarizes our conclusions.
6.2
Literature review
A vast literature describes and analyzes the long- and short-term wealth effects that target and bidding shareholders incur from M&As. Goergen and Renneboog (2004) review the empirical literature. They report unanimous support for the fact that target shareholders earn substantial abnormal returns from takeover transactions. For the United States, abnormal returns earned on the announcement day ranged from 21% during the 1990s (Mulherin and Boone, 2000) to 29% during 1963–1986 (Jarrell and Poulsen 1989). For the United Kingdom, Franks and Harris (1989) report abnormal returns of 23% over the month preceding the announcement. Conversely, no consensus has emerged as to whether the bidding firms derive any wealth effects. Roughly half of the studies reports small negative and significant abnormal returns for the bidders (see, e.g., Walker, 2000; Mitchell and Stafford, 2000; Sirrower 1997; Healy, Palepu, and Ruback 1992) and the other half finds small positive or zero abnormal returns (see, e.g., Eckbo and Thorburn, 2000; Maquiera, Megginson, and Nail 1998; Schwert, 2000; Loderer and Martin, 1990). Despite the quantity of literature on the wealth effects accruing to the target and bidding shareholders, few studies focus on the different types of consideration offered to the target shareholders. Unfortunately, the few studies that do so tend to equate loan notes to cash and to consider the final package of considerations paid to the target shareholders as a given. They therefore fail to account for any choice of consideration that was available to the target shareholders. Sudarsanam (2003) discusses the historical patterns in the payment methods for the case of the United Kingdom. He reports that cash has been the most popular consideration over the last three decades, followed by equity, then debt, or
Mix-and-match facilities and loan notes in acquisitions
137
combinations of different types of considerations. However, the popularity of cash as a means of payment decreases during bullish stock markets to the benefit of equity and vice versa. Sudarsanam (1995) finds that, in the case of the United Kingdom, the use of cash as a means of payment increases the success of hostile bids. He argues that, given the valuation problems involved in equity offers, the target’s managers can easily criticize their terms as inadequate if they are opposed to a bid. Franks, Harris, and Mayer (1988) find that the abnormal returns earned by the target shareholders depend on the means of payment. For the United Kingdom, they find that, during the announcement month, target shareholders gained on average 30% in all cash offers compared with only 15% in all equity offers. Jensen and Ruback (1983) review the evidence from seven U.S. studies. Computing a weighted average of the abnormal returns obtained from these different studies, they find that target shareholders earn abnormal returns of 29% in tender offers compared with only 16% in mergers.2
6.3
Data sources and sample selection
The list of takeovers, together with the characteristics of each deal,3 was obtained from the Deals module in Thomson One Banker, which is based on the SDC Platinum database. Further details were collected from the offer documents sent to the target shareholders and the company reports of the bidders. The electronic versions of these documents were downloaded from Thomson Research. The deal characteristics obtained from Thomson One Banker were cross-checked by consulting the offer documents. The details on the different types of consideration—or means of payment—offered to the target shareholders were sourced from the offer documents. Target shareholders were offered up to three different types of consideration: cash, new ordinary shares in the bidder, and/or loan notes. The company reports of the bidders provided information on the take-up of the different types of consideration offered. In order to ensure the availability of the required information, we focus on deals in which (1) both the target and bidder are U.K. firms and are listed on the London Stock Exchange and (2) the deals are completed and result in the bidder’s owning 100% of the target’s equity. There were 519 such deals during 1985–2004. Of these, 233 deals offered a choice of considerations to the target shareholders. We were able to obtain both the offer document and the bidder’s company report for the financial year covering the deal for 133 of the 233 deals. Most of the deals for which these documents were not available were 2
Event windows range from 20 to 60 days and are centred on the announcement date or cover different periods before and after the announcement date. 3 These include among others the names of the target and bidder, the industries in which they operate, the announcement date, the date when the deal became effective, the synopsis of the deal, the main events during the deal negotiation as well as their dates, and the total value of the deal.
138
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
from the 1980s or early 1990s. These 133 deals form the sample for the pilot study, which is the subject of this chapter.
6.4
Characteristics of mix-and-match facilities
Of the 133 deals forming the sample, 128 offered a mix–and-match facility. A mix-and-match facility gives individual target shareholders the choice to vary the proportions of the different types of consideration they have been offered. In one case, for example, target shareholders who had been offered a mix-andmatch facility consisting of cash and equity (i.e., new shares in the target firm) could choose between 100% cash or 100% equity or any combination of the two. However, 62 of the 128 deals with a mix–and-match facility offered only a limited choice to the target shareholders. WM Morrison Supermarkets plc, which took over Safeway plc in 2004, offered such a limited mix-and-match facility. Under the basic terms, Safeway’s shareholders were offered one new Morrisons share plus 60 pence in cash for each Safeway share. Safeway’s shareholders could also elect to vary the proportions of Morrisons shares and cash as consideration for their shares. However, these elections were only satisfied as long as enough other Safeway shareholders made off-setting elections. Overall, Morrisons still issued the same total number of shares and paid the same total amount of cash as under its basic offer. In only five deals were the target shareholders offered what we call a “combination” in addition to at least one other means of payment. A combination offered at least two types of consideration, but in fixed proportions, such as £2.80 in cash and 11 new shares of the bidder for every 5 shares in the target. As all 286 deals of the initial population of 519 deals over 1985–2004 that we excluded from our sample offered just one fixed combination, these 5 deals were included in the sample only because they were offered along with a choice of other alternative considerations.4 Finally, a total of 140 alternative considerations (a mix-and-match facility counting for a single alternative) was offered by the sample of 133 deals. In what follows, we concentrate on the 127 deals that offered a mix-and-match facility, excluding all the deals that offered combinations.5 Table 6.1 provides information on the different types of mix-and-match facilities. Panel A shows the frequency of the different types of mix-and-match facilities. The most frequently offered mix-and-match facility was cash and ordinary shares (46 out of the total of 127 mix-and-match facilities), followed by cash, 4
Two of the five deals offered more than one fixed combination (e.g., £3 in cash and ten new shares in the target or £3 in loan notes and ten new shares); one deal offered a fixed combination and a (limited) mix and match facility; and the remaining two deals offered a fixed combination of cash and shares as well as an all-share alternative. 5 As stated in the previous footnote, one deal offered both a fixed combination and a mix and match facility. This is excluded from the main sample. None of the deals offered more than one (limited) mix and match facility.
Mix-and-match facilities and loan notes in acquisitions
139
Table 6.1 The different types of mix-and-match facilities Panel A: Frequency of the different types of mix-and-match facilities Type of consideration
Frequency
Percentage
Cash ⫹ ordinary shares Cash ⫹ loan notes Cash ⫹ loan notes ⫹ ordinary shares Others Total
46 38 41 2 127
36.2 29.9 32.3 1.6 100.0
Panel B: Size of deals using the three main types of mix-and-match facilities and deals not offering any choice
Deal size in £m pounds sterlinga
Average Median
Cash plus ordinary shares
Cash plus loan notes
Cash plus loan notes plus ordinary shares
Deals without a choice
274.8 37.2
340.0 137.3
172.1 69.6
650.5 51.5
Panel C: t-Statistics (median tests)b for differences in means (medians) Column minus row
Cash plus loan notes
Cash plus ordinary shares Cash plus loan notes Cash plus loan notes plus ordinary shares
⫺0.523 (10.812) – –
Cash plus loan notes plus ordinary shares
Deals without a choice
1.089 (0.030) 1.689 (1.524) –
⫺1.567 (0.023) ⫺1.231 (0.049) ⫺2.231 (0.031)
The deal size is adjusted for inflation using the OECD Consumer Price Index (year 2000 ⫽ 100). The test statistic for the difference between two medians is a chi-square with 1 degree of freedom.
a
b
loan notes, and ordinary shares (41), and then cash and loan notes (38). Although the number of deals is low in some years, making it difficult to draw reliable conclusions, we see from Figures 6.1 and 6.2 that the popularity of mix-and-match facilities offering cash, loan notes, and ordinary shares, as well as those offering cash and loan notes, has increased since 1998, to the detriment of cash and equity. In most deals, the loan notes are unsecured and are not listed on the Stock Exchange. Furthermore, they are floating-rate notes, with the benchmark rate typically being the LIBOR.6 Loan notes frequently earn LIBOR ⫺0.5% or ⫺1%. 6
London Inter-Bank Offered Rate.
140
International M&A Activity Since 1990: Recent Research and Quantitative Analysis 30 25 20 15 10 5 0 1990
1992
1994
1996
1998
Cash & ordinary shares Cash & loan notes & ordinary shares
2000
2002
2004
Cash & loan notes
Figure 6.1 Incidence of deals with different types of consideration, 1988–2004.
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1990
1992
1994
1996
1998
Cash & ordinary shares Cash & loan notes & ordinary shares
2000
2002
2004
Cash & loan notes
Figure 6.2 Incidence of deals with different types of consideration (1988–2004), expressed in percentage terms.
Panel B of Table 6.1 reports the average and median deal sizes in real pounds sterling, i.e., pounds sterling at year 2000 prices. The tests for the differences in means as well as those for the differences in medians can be found in Panel C. Focusing on the average size (Panel B), the 286 deals that do not offer a choice of considerations to the target shareholders—and are not part of our sample— are the largest ones, with an average size of £651 m. They are followed in size
Mix-and-match facilities and loan notes in acquisitions
141
by the deals with mix-and-match facilities consisting of cash and loan notes, then by those with cash and equity mix-and-match facilities. The average deal size for mix-and-match facilities offering cash, ordinary shares, and loan notes is the smallest. Concerning the median size, the ranking by size is different, with cash and loan notes mix-and-match facilities ranking at the top, followed by mix-and-match facilities offering the three types of consideration, deals without a choice, and then mix-and-match facilities consisting of cash and ordinary shares. The differences in ranking and the difference between the mean and median values in general suggest that the distributions for the deal size are skewed and non-normal. In particular, we see some extremely large transactions—as compared with the average transaction size—among deals without a choice and those offering mix-and-match facilities of cash and ordinary shares. For example, the former category includes the takeover of SmithKline Beecham plc by Glaxo Wellcome plc, which had a deal value of £46 billion (expressed in year 2000 pounds sterling). If this deal is excluded, the average deal value for this category drops down to £486 m. Panel C of Table 6.1 shows that the average size for deals offering no choice is significantly larger—at the 5% level of significance for the two-tailed t-test— than that for mix-and-match facilities consisting of cash, ordinary shares, and loan notes. However, the difference in medians between these two types of deals is not significantly different from zero. Furthermore, we see a significant difference— albeit only at the 10% level—between the average deal size for mix-and-match facilities with cash and loan notes and the average size for mix-and-match facilities offering all three types of considerations. Finally, although the difference between the average size for mix-and-matches involving cash and ordinary shares and that for mix-and-matches involving cash and loan notes is insignificant, the difference between the two median sizes is significant at the 1% level. At first sight, it is difficult to interpret these results. Figure 6.3 sheds further light on the frequency distribution of deal sizes for each of the four types of deals. For all four types, the frequency distribution is U-shaped, showing that the majority of acquisitions are relatively small, with a sizeable percentage of very large deals, but very few deals fall between the two extremes. The only slight exception to this pattern is the size distribution for mix-and-match facilities offering cash, loan notes, and shares. Here the curve is not only flatter, but it also has a peak in the middle. More importantly, however, the two types of mix-and-match facilities involving loan notes have a flatter, U-shaped distribution than do the two other types of deals. Table 6.2 shows the averages for the maximum possible take-up and the actual take-up for the three most frequent types of consideration in mix-and-match facilities—cash, ordinary shares and loan notes—for all the deals (Panel A)7
7
Panel A includes the deals with unrestricted mix and match facilities which, by definition, have a maximum possible take-up of 100.
142
International M&A Activity Since 1990: Recent Research and Quantitative Analysis 60.0%
Frequency
50.0% 40.0% 30.0% 20.0% 10.0% 0.0% 50
100
150
200
250
300
350
400
450
500 Higher
Real deal value (£m) Cash & ordinary shares Cash & loan notes Cash, ordinary shares & loan notes Deals with no choice
Figure 6.3 Frequency distribution of real deal value. Table 6.2 Take-up of alternatives within mix-and-match facilitiesa Panel A: All the deals Cash
Average Median Percentage of firms paying out maximum
Ordinary shares
Loan notes
Maximum possible take-up
Actual take-up
Maximum possible take-up
Actual take-up
Maximum possible take-up
Actual take-up
87.5% 100.0% 13.4%
58.2% 66.7% –
76.5% 100.0% 11.5%
52.1% 49.6% –
92.7% 100.0% 0.0%
8.7% 6.0% –
Panel B: Deals with an upper limit on at least one type of consideration Average Median Percentage of firms paying out maximum a
73.9% 75.8% 11.5%
49.9% 50.4% –
64.8% 59.7% 15.5%
48.2% 47.7% –
83.4% 100.0% 0.0%
7.5% 5.6% –
The maximum (actual) possible take-up is the total value (actual value) of a certain type of consideration expressed as a percentage of the value of the total consideration.
Mix-and-match facilities and loan notes in acquisitions
143
and the deals with an upper limit on at least one of the types of considerations offered (Panel B). The maximum possible (actual) take-up is the total value (actual value) of a certain type of consideration expressed as a percentage of the value of the total consideration. Panel A shows that, for all the 128 deals, the average for the maximum possible take-up is highest for loan notes, followed by cash, then shares. The median is 100%, reflecting the fact that 66 of the deals have unrestricted mixand-match facilities. The actual take-up is highest for cash, closely followed by ordinary shares. Conversely, the take-up of loan notes is very low. However, the difference between the maximum possible take-up and the actual take-up (not reported in the table) is lowest for ordinary shares, followed by cash, and then by loan notes. The maximum possible amounts of cash and equity are taken up in 13% and 12% of deals, respectively. Panel B reports the equivalent figures for the 62 deals with an upper limit on at least one of the considerations included in the mix-and-match facility.8 The average for the maximum possible take-up for cash is about 74% with a median of 76%. In about 12% of the deals, the maximum possible amount of cash is paid out. Concerning ordinary shares, the maximum possible take-up is lower, with 65%, and a median of 60%. However, the percentage of deals in which the target shareholders receive the maximum number of bidder shares on offer is higher, with 16%. Finally, the average maximum possible take-up for loans notes is highest, with 83%. The median of 100% shows that in one-half of the deals we find no upper limit on the number of loan notes that can be taken up. As expected, in no deal do the target shareholders take up all the loan notes on offer. The actual take-up is closest to the maximum amount on offer for shares, followed by cash, then loan notes. This is an interesting result for two reasons. First, the average target shareholder has a stronger preference for shares than cash, despite the fact that the empirical evidence suggests that the gains to target shareholders are higher in cash deals than in equity swaps. Second, loan notes seem to appeal to a minority of target shareholders only. The relatively low popularity of loan notes seems to explain why most mix-and-match facilities do not specify an upper limit on the loan notes, despite the fact that virtually all require a minimum amount of applications (in terms of their pounds-sterling value) for loan notes to be issued. Overall, Table 6.2 suggests that target shareholders value choice because they do not exhaust the maximum possible take-up for a particular type of consideration in most deals. Furthermore, in only 12–16% of the deals, the mix-andmatch facility seems to restrain their choice.
8
For a handful of deals, the actual take-up for a certain type of consideration exceeded the maximum amount offered by a few percentage points. The bidders behind these deals may have decided to exceed the maximum amount if the amount applied for slightly exceeded the maximum to keep down the overall transaction costs caused by the deal.
144
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
6.5
Loan notes
6.5.1
Introduction
Mix-and-match facilities involving loan note offers have increased in popularity in the last 20 years, although acceptance of loan notes has always been less popular in general than shares or cash (see Section 6.4). The value of this element of consideration, too, is always lower than cash or shares (as reported in Table 6.2). Table 6.3 shows the incidence of deals involving loan notes in the period 1988–2004, and Figures 6.1 and 6.2 show the comparison with deals involving other types of consideration. As can be seen, their popularity peaked in the late 1990s and early years of the 21st century. Of the 98 deals in this period, 64 took place in the years 1998–2001. The obvious question, then, is twofold: why the upsurge in popularity at that time, and why has that popularity subsequently declined? One answer is strictly behavioral: something becomes popular not because it is necessarily the best option, but because everyone else is doing it. Financial instruments or devices, as is anything else involving human activity, are subject to fashion. Some few years ago, for example, following an initiative taken by Debenhams (a chain of department stores), it became very popular for store card and credit-card companies to distinguish some part of charged costs as “service charges,” and thus not subject to Value-Added Tax (VAT). While this practice saved tax in the short term, it was eventually quashed—the result of a court case brought (and won) by Her Majesty’s Revenue and Customs, who had been suffering the loss of VAT revenue. The financial press had roundly criticized this practice, offering much adverse commentary. In a sense, many card providers were impelled to adopt the practice, despite disagreeing with it and despite damaging their reputations, because all their peers were following it. Such issues apart, loan notes are especially flexible and useful devices, both for acquiring companies and target shareholders. Another possible reason for the increased popularity of loan notes is advanced in Section 6.5.2.
Table 6.3 Numbers of deals involving loan note transactions, 1988–2004 Year
Number of deals
Year
Number of deals
1988 1989 1990 1991 1992 1993 1994 1995 1996
2 1 2 3 1 1 4 5 4
1997 1998 1999 2000 2001 2002 2003 2004 Total deals
6 15 21 17 11 3 1 1 98
Mix-and-match facilities and loan notes in acquisitions
145
The purpose of this section is to examine the data on offers involving loan notes and consider the advantages and otherwise of loan notes (versus other types of consideration) for (a) the target shareholder and (b) the bidder, as well as to determine how finalized offers are reported and disclosed in acquiring company accounts.
6.5.2
The target shareholders’ perspective
One of the attractions of loan notes is that they provide the target shareholders with a way to manage the tax consequences of an asset disposal. Loan notes will therefore appeal to target shareholders who are tax payers and have already exhausted all or part of their tax credit in terms of capital gains for the current tax year.9 The facility for tax management holds true whether the target shareholder is an individual, a company, or some other corporate body (except for tax-exempt bodies), although the situation resulting from such a disposal is often more complex and serious for the individual. An individual may suffer more owing to the nature of the tax calculations, the affordability of tax advice (companies often have tax advisers as a matter of course), and the sums of tax that may arise. If loan notes are taken up as consideration for the disposal of shares, the disposal is deemed to take place when the actual cash is received, that is, when the loan notes are redeemed. This means that, within the parameters permitted by the loan note redemption policy, a target shareholder can determine when a chargeable gain, if any, may arise, and thus if or when a tax liability arises and must be paid. More precisely, some loan notes give the holders the option to request repayment either on demand or on 30 days’ notice. This is advantageous to the holders, as it facilitates receipt of consideration so as to manage the tax liability arising (see succeeding discussions). Most companies have a final redemption date, however, so that such loan stock is not left unredeemed for an indefinite period. Given that loan notes enable their holders to time their tax liabilities via the option of early redemption or at least to postpone them (as compared with cash payments, in which case no such option may be offered), they are valuable to certain types of target shareholders. As a result, most loan notes earn interest below LIBOR (typically, 0.5% or 1% below LIBOR). In turn, the low interest rate makes loan notes unattractive to target shareholders who would not benefit from delaying their tax liabilities or are not subject to capital gains tax (such as charities). 9
Shareholders may not have wanted and, indeed, may have voted not to accept proposed takeovers. There is some evidence of resistance to takeover deals in general, as in the case of Laporte plc’s 1993 acquisition of Evode plc, where Note 17 in the company accounts refers to £8.4 m being paid for “the remaining Evode ordinary and convertible cumulative redeemable preference shares”, which occurs after the main consideration has been paid and which is a clear suggestion that this was a compulsory purchase after the deal had been declared final. These shareholders were not in favour of the transaction.
146
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Examples to show how such disposals may affect target shareholders are given in Section 6.7. The numbers of small shareholders rose significantly following the substantial privatization of utility companies in the late 1970s and 1980s. This growth was also generally encouraged by advances in technology, such as by Internet trading. It is therefore not surprising to find the development of a device such as loan notes to allow shareholders, particularly individuals, to cope better with changes that might not be of their choosing, but do present a potentially enormous impact on taxes. The reason for the increased popularity of loan notes during 1998–2001 may indeed have been the peak in Stock Exchange prices during the late 1990s. The hefty rise in share prices may have exhausted the tax credits of lots of investors and, as a result, generated a need for postponing the tax liabilities caused by acquisitions.
6.5.3
The bidder’s perspective
From the acquiring company’s perspective, loan notes present three advantages. First, unlike cash (which may come from different sources, e.g., via an equity issue), loan notes offer equivalence between the means of payment and the source of financing for the acquiring company. However, the two main advantages of loan notes are concerned with cash management and the low costs of issuing loan notes.
Deferring cash payable The use of loan notes defers the amount of cash that is payable on completion of a deal until some future date. This is useful if a company cannot easily or immediately borrow sufficient capital to finance any deal. In some cases, bidders issue loan notes with staggered redemption periods to coincide with their particular cash flows. In conclusion, then, loan notes are a very flexible finance device, which can be made shorter or longer in term by the acquiring company. Moreover, they can offer cash management possibilities to the acquirer while maintaining a flexible redemption facility to the loan note holder.
Low costs of issuing loan notes As loan notes typically offer interest rates below LIBOR in exchange for tax savings, they represent a cheap way of financing an acquisition. Bidders also frequently further reduce the issuing costs of loan notes via the imposition of a minimum subscription. It is common to find offer documents stating that loan notes will not be issued unless target shareholders subscribe for a minimum specified value. Table 6.4 shows the range of minimum specified values for some of the deals during the sample period. Sometimes the restriction is a maximum (see also Section 6.4). In St Ives Group plc’s 1998 acquisition of Hunters Armley Group plc, a loan note issue was capped at a maximum of £3 m. Imperial Metals Industries later (IMI) plc’s
Mix-and-match facilities and loan notes in acquisitions
147
Table 6.4 Range of minimum specified loan note values for deals, 1988–2004 Minimum specified loan note application (total £)
Year of deal
Target company
Acquiring company
£100,000
1998 2000 2000
Plasmec plc A&C Black plc Sports and Outdoor Media International plc
Arlen plc Bloomsbury Publishing plc Sportsworld Media Group plc
£250,000
1998
Gibbs Mew plc
Enterprise Inns plc
£1,000,000
1998 1999 1999 2001 2001 2001
EIS Group plc Provend Group plc Hodder Headline plc Old English Inns plc Exchange FS Group plc Country Gardens plc
TI Group plc Bunzl plc WH Smith Group plc Greene King plc Marlborough Stirling plc Wyevale Garden Centres plc
£2,500,000
1998
Independent Parts Group plc Tarmac plc
Finelist Group plc
2000
Anglo American plc
£5,000,000
1997 1998 2001
Chubb Security plc Eurotherm plc Wates City of London Properties plc
Williams Holdings plc Siebe (later Invensys) plc Pillar Property plc
£10,000,000
1998 2000
Dennis Group plc Swallow Group plc
Mayflower Corp plc Whitbread plc
£20,000,000
1994
Trans World Communications plc
Emap plc
1999 acquisition of Polypipe plc required no lower or upper limit on nominal amount of loan notes, although loan notes would be guaranteed by Lloyds Bank plc only up to a nominal amount of £100 m. In conclusion, loan notes are not only a flexible finance alternative facilitating the bidder’s cash management, but they are also a cheap way to finance an acquisition. The question that arises is why, then, are loan notes not the main means of payment for acquisitions. One answer has already been mentioned in Section 6.5.2. Although loan notes may be attractive to certain shareholders, enabling them to reduce their tax liabilities, the low interest rates they offer may make them unattractive to shareholders who are not subject to tax on capital gains. Thus, although loan notes clearly formed part of the offer, the options were not taken up in some cases. Table 6.5 shows instances of transactions in which loan notes were available in an offer document, but the options were not taken up.
148
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 6.5 Transactions where loan notes were offered, but not taken up Year
Target
Acquiring company
Notes
1995 1996 1998 1999 1999
Eastern Group plc Innovations Group plc Eurotherm plc Cirqual plc Mansfield Brewery plc
1
1999 1999 2000 2000 2000 2000
Vickers plc Wyndham Group plc Breedon plc Dorling Kindersley Holdings plc Rugby Group plc Sports and Outdoor Media International plc Beazer Homes plc Exchange FS Group plc Old English Inns plc Tempus plc Tay Homes plc
Hanson plc Burton plc Siebe (later Invensys) plc L. Gardner Group plc The Wolverhampton and Dudley Breweries plc Rolls-Royce plc Ryland Group plc Ennstone plc Pearson plc RMC Group plc Sportsworld Media Group plc
2001 2001 2001 2001 2002
Persimmon plc Marlborough Stirling plc Greene King plc WPP Group plc Redrow plc
2 3 4
1. Some detective work is required in looking at both the 1995 and 1996 accounts to determine the absence of loan notes, as the 1995 accounts show £2,495 m accrued purchase consideration, which is not immediately identifiable as cash or loans—it is not evident as cash. 2. Only cash is evident as consideration, although share and loan notes were offered. 3. No evidence shows that the loan note option again was taken up, but here, too, some detective work is necessary. Page 43 of the 2001 accounts, the consolidated cash flow, gives £325,389 k as cash paid out for the acquisition of subsidiaries. Therefore, the loan note option cannot have been taken up. 4. The only consideration evident is cash, although both loan notes and shares appear as options in the offer document.
6.6 6.6.1
Accounting disclosure and treatment Introduction
Various issues arise when examining the accounts of acquiring companies to determine how deals have been accounted for in annual reports. These are considered in the following sections. The fact that, in some cases, it is fairly complicated to determine the actual take-up of loan notes may explain why most existing studies have used the shortcut of equating loan notes to cash.
6.6.2
Tracing details of the deal
It is not always clear what has happened in any given deal, and it can be difficult and (sometimes virtually impossible) to trace a deal in all its component
Mix-and-match facilities and loan notes in acquisitions
149
parts from offer document through the way it is accounted for and disclosed in an acquiring company’s annual accounts. In their accounts, many companies do not distinguish between cash and loan notes at all in describing types of consideration for deals. Greene King plc, for instance, combined them, with respect to consideration paid for its 1999 acquisition of Morland plc (Greene King plc, Report and Accounts for the period ended 29 April 2000, Note 26). To trace the split between cash and loan notes requires further reference to Note 20, where loans are detailed. If loan notes are on offer, then they are usually offered in the same amount as the cash alternative. This is evident in Anglian Water (later AWG) plc’s 2000 takeover of Morrison plc, where shareholders could opt for shares for up to 50% of the cash consideration (one Anglian Water share for every 620p of cash consideration). They could opt for loan notes for part or all of the cash consideration (£1 nominal loan note for every £1 of cash consideration). In Marlborough Stirling plc’s 2001 acquisition of Exchange FS Group plc, there was a loan note alternative to the cash element of the consideration. Shareholders were also offered a mix-and-match facility whereby they could elect to vary the proportion of shares and cash received, subject to availability. The component parts of Anglo American plc’s acquisition of Tarmac plc are untraceable in the acquiring company’s 2000 annual report and accounts. It is similarly difficult to trace details of Prudential plc’s 1999 acquisition of M&G Group plc in the acquiring company’s 1999 accounts. Note 27 gives the fair value of consideration (including expenses) as £1,943 m, but it is unclear what this includes, and whether the loan notes of £168 m (as shown in Note 21) form part of the consideration, even though loan notes were on offer here. Sometimes difficulty in tracing details may be characteristic of the time at which deals occurred, especially if the deals bridge the acquiring company’s accounting year end or the acquiring company obtains a controlling stake in the target after having previously owned a smaller number of shares for a period of time (which may vary in length). Such appears to be the case with Allied Leisure plc’s 2000 acquisition of Waterfall Holdings plc. Note 12a discloses an earlier investment in the target, but the value of that investment is not analyzed into its component parts, so it is impossible to know overall how much of each type of consideration was offered. In the case of The Berkeley Group plc’s 1991 acquisition of James Crosby Group plc, Berkeley’s 1991 accounts include under Accruals in Note 14 a sum of £8,508,000 with respect to this acquisition. This sum represents amounts due but not yet paid, as the deal was being finalized at the year end, and will thus include the offered loan notes, which are subsequently shown separately in the 1992 accounts. It could have been that, while the ultimate liability was known in 1991, the form it might take (cash/loan notes) was not clear at that point. The conclusion to be drawn from the preceding discussion is that, in some instances at least, details of takeover deals are difficult to trace in the accounts of the acquiring company. This may be a feature of when they occurred (e.g., bridging an accounting period end), or it may represent the fact that a deal is “small” in terms of the acquiring company’s overall figures (as in the case of
150
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Prudential), in which case figures are not separately itemized. Also, companies may wish to keep deals confidential from rivals. This may also affect other issues, which we discuss in succeeding sections.
6.6.3
Depth of detail disclosed
Some companies disclose a wealth of detail on acquisitions, whereas others disclose an absolute minimum.10 Generally, in the wake of accounting scandals such as Enron and subsequent changes in reporting practices, one might expect more recent sets of accounts to disclose more extensive data, albeit not universally so. For example, Bunzl plc acquired Provend Group plc in March 1999. There is no reference to this particular acquisition by name, or to any other, although it is clear from Bunzl’s 1999 accounts that they made acquisitions in this period. The principal information given by Bunzl’s 1999 accounts is shown in Tables 6.6 and 6.7. The cost of acquisitions cannot be reconciled between the two sets of information, and no mention is made of consideration either in total or by type. Provend does not appear to be listed in Note 32 as a principal undertaking. From these accounts, it seems clear that Provend is not the only business acquired, but it cannot be determined, from the accounts at least, what these businesses were. The 1999 accounts of Scottish Media Group plc present a similar situation. Scottish Media acquired Primesight plc in March 1999 (of interest here because it involved loan notes); however, while the separate targets are identified in Note 23, the consideration received is not itemized between the different companies. Slough Estates plc’s 1998 acquisition of Bilton plc also shows consideration for all acquisitions combined. In Whitbread plc’s 2000 acquisition of Swallow Table 6.6 Bunzl plc, Annual Report and Accounts for the year ended 31 December 1999, Note 10
Bunzl plc: Beginning of year at cost less provisions Additions Disposals End of year at cost less provisions
Own shares £m
Investments in subsidiary undertakings £m
Total £m
1.1
242.9
244.0
2.2 – 3.3
44.7 (5.9) 281.7
46.9 (5.9) 285.0
The investments in subsidiary undertakings at 31 December 1999 are stated net of provisions of £30.4 m (1998: £30.4 m). Principal subsidiary undertakings are listed in Note 32.
10
This comment should not in any way be construed as a criticism of any company’s accounts or actions, but merely a comment on their usefulness to anyone interested in tracing details of completed deals.
Mix-and-match facilities and loan notes in acquisitions
151
Table 6.7 Bunzl plc, Annual Report and Accounts for the year ended 31 December 1999, Note 29 Book Provisional fair value adjustments value at acquisition Revaluation Consistency of Other £m accountancy £m £m policy £m Summary of the effect Tangible fixed assets Stocks Debtors Creditors Net bank overdrafts Provisions for liabilities and charges Taxation
of the acquisitions 3.8 (0.9) 12.1 21.4 (23.7) (7.7) – (0.7)
(1.2) 4.7
(1.6)
(0.2) (0.5) 0.2 (4.2)
(0.7)
(4.2)
(1.2)
Goodwill Consideration Satisfied by: Cash consideration Deferred consideration The net outflow of cash with respect to the acquisition of businesses Cash consideration Overdrafts of businesses acquired Net outflow of cash in respect of the acquisition of businesses
Fair value of assets acquired £m
2.9 11.9 20.9 (23.5) (7.7) (5.6)
(1.2) (2.3) 57.9 55.6 52.4 3.2 55.6 52.4 7.7 60.1
Group plc, there is no way to determine from Note 28 the split of consideration for new businesses acquired. The same is true for Wolseley plc’s 1999 acquisitions. Tracing the details for the acquisition of British Fittings Group plc (involving loan notes) is not possible from information given. The barest information is provided in Note 23 and in the financial review on pp. 26–29. Names of acquisitions are not generally provided. These cases are, however, fairly unusual— most companies disclose transactions in very much more detail. Confidentiality also remains an issue to consider. Some companies may not wish to disclose detailed information that may be of use to rivals. Additionally, there is the question of materiality. Some companies’ accounting figures are reported in millions
152
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
of pounds, and a deal may not be sufficiently significant to itemize separately, even though the same figures might be significant for a smaller company. In most cases, however, details of acquisitions are very clear, as in the case of Misys plc’s 2001 acquisition of DBS Data Management plc, as shown in Table 6.8. Misys had several acquisitions in that year and disclosed similar detail for all of them. Nothing is consistent, however, with respect to how information is provided in sets of accounts, as each company seems to develop its own methods. Table 6.9 relates to Abbey National plc’s acquisition of Cater Allen Holdings in 1997, Table 6.8 Misys plc, Annual Report and Accounts for the year ended 31 May 2002, Note 16 (showing DBS acquisition)a,b Book value
Tangible fixed assets Investments in associate undertakings and other investments Debtors Cash at bank and in hand Bank loans Creditors Provisions for liabilities and charges Deferred income
Accounting policy alignment
Fair value adjustments
Fair value
3.8 1.3
(0.2) –
(0.7) (0.2)
2.9 1.1
17.5 27.2 (0.5) (31.7) (4.0)
14.2 –
(0.8) –
1.5 (14.0)
– (5.6)
30.9 27.2 (0.5) (30.2) (23.6)
(0.9)
–
12.7
1.5
–
Goodwill capitalized
6.9 70.1
Fair value of consideration
77.0
Of which: Cash consideration and acquisition expenses Loan notes issued
(7.3)
(0.9)
59.1 17.9 77.0
a
DBS: The acquisition of DBS Management plc (DBS), a provider of IFA network solutions in the United Kingdom, was completed on 29 August 2001. For the period post 31 March 2001, being DBS’ previous year end date, to 29 August 2001, DBS reported turnover of £78.8 m and a loss after taxation of £3.3 m. For the year to 31 March 2001, DBS reported turnover of £191.5 m and a loss after taxation of £1.6 m. The accounting policy alignments primarily comprise the grossing up of amounts recoverable with respect to regulatory reviews in debtors and provisions of £12.5 m. The fair value adjustments primarily comprise the Directors’ valuation of provisions in respect of regulatory reviews of £4.3 m as well as the Directors’ valuation of fixed assets, debtors, liabilities, and onerous contracts. b All figures in £ millions
Mix-and-match facilities and loan notes in acquisitions
153
Table 6.9 Abbey National plc, Annual Report and Accounts for the year ended 31 December 1997, Note 23 23. Shares in Group undertakings
Subsidiary undertakings Banks Others
1997
1996
Cost and book value £m
Cost and book value £m
442 1,993 2,435
451 1,994 2,445
The movement in shares in Group undertakings was as follows: Company £m At 1 January 1997 Exchange adjustments Additions Disposals Amounts written off At 31 December 1997
2,445 (69) 363 (283) (21) 2,435
Abbey National plc entered into the following transactions: On 24 July, Abbey National plc acquired Cater Allen Holdings PLC. The amount paid in consideration for the issued share capital was £193m, of which £191m was payable in cash and £2m in the form of loan notes. In addition, acquisition costs of £2m were incurred. The company was subsequently sold to Abbey National Treasury Services plc. On 6 May, the investment in HMC Group PLC was transferred to First National Bank Plc. On 7 May, Abbey National plc purchased for cash the remaining 25% interest in Abbey National General Insurance Services Ltd for £4 m. Subscriptions for additional share capital in subsidiary undertakings during the year amounted to £166 m.
which has a different format from Misys. This is not atypical of the way in which multiple acquisitions are disclosed. In conclusion, the comments made previously about tracing details of deals in annual accounts may also be made here. However, it is clear that accounts produced more recently, by and large, tend to be more informative as a result of a change in the culture of corporate disclosure.
6.6.4
Creditors’ disclosures
The £2 m loan notes in the Abbey National extract, which form part of the consideration, are not separately traceable in the creditors’ notes. Given the volume and type of transactions generally for this group and the overall materiality of the loan note value, this omission is not unexpected. It is common to be unable to trace loan notes issued, as they are sometimes subsumed within a larger total
154
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
of similar items. Table 6.10 lists deals where it has not been possible to trace the loan notes issued to specific items within creditors’ notes. In Arlen plc’s 1998 accounts, with respect to loan notes issued for the acquisition of Plasmec plc, it seems that some of the £854 k issued may have been repaid before the end of the period in which they were accounted for, although this is not at all clear. This also seems true of the £3,686 k of loan notes disclosed in Note 15 as part of the consideration for BSS Group plc’s acquisition of PTS Group plc in late 1999. The only loan notes disclosed anywhere in the relevant BSS Group plc’s annual report and accounts (for 2000) are £3,593 k in Note 20 for creditors due within 1 year. Likewise, Hill & Smith Holdings plc’s 2000 acquisition of Ash & Lacy plc, at Note 24 in the acquiring company’s 2001 accounts, shows loan note consideration of £1,759 k, but Notes 15 and 16 show only £1,731 k. Again, as with issues previously discussed, the reason for difficulties in tracing individual loan note issues may be because of confidentiality and materiality. Following on from materiality, issues may be individually untraceable as they become merged in accounts with other like items, hence forming a small part of a larger total.
6.6.5
Reflection of offer documents in annual reports and accounts
Annual accounts report quantifiable events, such as the issue of shares, cash, and loan notes (as appropriate) as consideration, but do not generally report on how these reflect the take-up of particular options offered to target shareholders. This is the case even when deals are not particularly complex. In the case of Anglian Water plc’s 1997 acquisition of Hartlepool Water plc, Anglian Water’s offer was to acquire the entire ordinary share capital, or 7,320,000 ordinary shares, of Hartlepool Water at £2.66 (U.S. $4.36) per share, or a total of £19.47 m (U.S. Table 6.10 Deals in which loan note issues cannot be identified as specific creditors Year
Target
Acquiring company
1997 1998 1999 1999 1999 2000 2000 2000 2001 2001 2002 2003 2004
Cater Allen Holdings Doeflex plc Hicking Pentecost plc Just Rubber plc Milner Estates plc Harveys Furnishing plc Morrison plc Real Time Control plc CandB Publishing plc Meconic plc Prowting plc HP Bulmer Holdings plc Johnston Group plc
Abbey National plc British Vita plc Coats Viyella plc Scapa Group plc Delancey Estates plc Homestyle Group plc AWG plc NSB Retail Systems plc Chrysalis plc Johnson Matthey plc Westbury plc Scottish and Newcastle plc Ennstone plc
Mix-and-match facilities and loan notes in acquisitions
155
$31.94 m). Alternatively, loan notes were offered in lieu of cash. Note 29c of Anglian Water plc’s 1997 accounts shows a total consideration of £20.0 m, made up of £16.9 m in cash and £3.1 m in loan notes. Furthermore, Anglian acquired cash of £3.1 m as part of the target’s assets. This amount is disclosed separately as part of the same note, showing that the net cash cost to Anglian was £13.8 m. Some companies make no such disclosure or show such cash acquired differently. However, Anglian does disclose more information with respect to its 2000 takeover of Morrison plc (see Table 6.11). Anglo American plc, however, with respect to its 2000 acquisition of Tarmac plc, shows only the overall net cost of acquisition after disposing of the Tarmac businesses and assets, which it did not integrate into the group (see Table 6.12). In this case, nothing is disclosed at all about consideration, although the cost of acquisition is stated to be $U.S.1,913 m. (Anglo American produces annual reports in two currencies, pounds sterling and U.S. dollars, and the dollar version was used here.) Hillsdown Holdings plc’s 1996 acquisition of Hobson plc reveals in its 1996 accounts (at p. 13) an acquisition cost for Hobson as £124.4 m. This information is given in a review of activities, which follows the chief executive’s review. However, the comments on p. 13 seem to exclude the £12.7 m stated in Note 19 as net debt acquired in Hobson, which is an additional cost. (Conceivably, this latter sum represents loans already existing in the target.) It is not possible to trace this £12.7 m anywhere in creditors’ notes. Annual accounts, as mentioned in the preceding discussion, only report transactions that have happened. One cannot expect to find an analysis of details of offers that were not taken up, nor a breakdown of how individual Table 6.11
AWG plc, Annual Report and Accounts for the year ended 31 March 2001, Note 15
The offer to Morrison shareholders was for cash but also included partial share and loan alternatives. Under the partial share alternative, Morrison shareholders could elect to receive up to half of their consideration in the form of new Anglian Water Plc shares. Under the loan-note alternative, Morrison shareholders could elect to receive part or all of their consideration in the form of loan notes with a redemption date of 30 September 2007. The loan notes are unsecured and bear interest in arrears based on LIBID minus 0.5%.
Table 6.12 Anglo American plc’s Annual Report and Accounts for the year ended 31 December 2000, Note 28. On 1 March 2000, the Group acquired 100% of Tarmac plc the building materials and aggregates business, for U.S. $1,106 million, which is net of proceeds received from the sale of assets held for resale of U.S. $709 million and net cash on hand of U.S. $98 million.
156
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
elements of offers that were taken up were satisfied. These details may not be material, or the company may wish to keep them confidential. The fact that accounts then also report how an acquisition was accounted for adds an extra layer of complexity: For instance, cash acquired in the target company as part of the takeover may appear netted off against overall cost of acquisition, or target debt may be acquired, which is accounted for as a separate element. This is a mingling of the deal itself and what is acquired is a result of the deal. The fact that both are accounted for in the same set of accounts makes information difficult to unravel. In no way should one infer criticism of any company’s practices; rather such omissions and subsumptions represent an inherent feature of company accounts, which are not designed to provide this detail. To summarize, this section has revealed some of the complexities involved in determining the actual take-up of the different forms of consideration in acquisitions. In particular, it can be sometimes difficult—and in a few cases impossible—to establish how much cash and how many loan notes were taken up. The difficulty in obtaining the information on the split between cash and loan notes may explain why previous studies have either mistakenly taken loan notes for cash or for purposes of simplification treated loan notes as cash.
6.7 6.7.1
Loan notes—tax choices Introduction
Loan notes, particularly when redemption dates have a wide range or redemption is at the option of the stockholder, provide the opportunity for the stockholder to manage the tax liability that may arise as a result of the deal. This may prove very welcome to the target shareholders as, effectively, they may face a forced disposal of shares, which could give rise to liability to capital gains tax (see also Section 6.5.2).11 When a target shareholder’s shares are acquired, prima facie, this is a disposal for tax purposes. If the target shareholder receives the acquiring company’s shares in exchange for the target shares, then—assuming that the provisions of S.135 of the Taxation of Chargeable Gains Act (TCGA) 1992 apply—this transaction should involve no gain or loss: the new shares take the place of the old shares, taking on the same cost allowable per legislation for disposal purposes. If the target shareholder receives shares and cash, the cash consideration may generate a capital/chargeable gain arising with respect to the cash element, on the agreed date of disposal. Such cash will be taxable for the fiscal year in which the agreement is made. Cash may be received at a date later than the date of agreement, which would be when any deal is declared unconditional. It is the date of the agreement that determines when any tax liability may arise, not the 11
Companies do not pay capital gains tax. They do, however, pay corporation tax on any chargeable gains, so the effect may be the same.
Mix-and-match facilities and loan notes in acquisitions
157
date of the actual cash receipt. A problem may arise if the agreement is in one tax year and the cash received is in another, but given the lag in paying tax for individuals, the problem may be potential rather than actual. For individuals, tax due for the fiscal year ended 5 April 2006 is payable on or before 31 January 2007. The situation for corporate taxpayers may be different, as discussed in the following section. For tax purposes, loan notes are regarded as deferred cash and will not give rise to any tax liability until an agreed date of redemption. Hence the shareholder may be able to seek repayment of the loan notes to suit particular circumstances. If there is a predetermined series of redemption dates, the shareholder may plan other transactions to allow for this contingency. This is true whether the loan stockholder is an individual (natural person) or a corporate body (legal person). When target shareholders accept cash or loan notes, according to capital gains tax rules, there is a part disposal of the original shares, and the usual calculation for part disposals will apply. If loan notes are redeemed at different times, each redemption will be a part disposal. This adds a degree of complexity, but it is not unmanageable. The next section offers models to determine the choices individual and corporate loan stockholders may make for tax purposes when considering whether to accept loan notes or not.
6.7.2
Modeling tax choices
We assume in the following models that the taxpayer, whether individual or corporate, would seek to minimize tax payable, even though some taxpayers cannot obtain such a benefit. If they pay tax at the highest rates, before taking gains into account, then any tax on gains will be paid, regardless, at that highest rate. When a gain would increase a taxpayer’s income/profits to a level taxable at higher rates, then the taxpayer would likely seek to manage the timing of the gain. For highest rate taxpayers, loan notes would offer only a possibility of deferring tax, not reducing the overall amount payable (subject to tax rates and levels of income/profits remaining unchanged).
Individual Ic
⫽
F([(Income other than capital gains ⫹ Igain) ⫻ TRband], Idue date, other needs)
⫽ ⫽ ⫽ ⫽
individual choice individual’s capital gain, calculated per Table 6.13 date when tax on combined income would be payable tax rates applicable to bands of income
where Ic Igain Idue date TR
(continued)
158
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Corporate body Cc
⫽
F([{Profits ⫹ income other than chargeable gains ⫹ Cgain] ⫻ [TRband ⫹ Cdue date/s]}, other needs)
Cc Cgain
⫽ ⫽
TR
⫽
Cdue date/s Other needs
⫽ ⫽
corporate body choice corporate body’s chargeable gain, calculated per Table 6.14 tax rates applicable to levels of profits chargeable to corporation tax date/s when tax on combined profits would be payable any reason a company may have for making a decision, e.g., the need for cash, etc.
where
In calculating a capital/chargeable gain, we find significant differences between individuals and corporate bodies, typically companies. Indexation allowance is an uplift to cost, calculated by a defined formula by reference to the Retail Price Index, to allow for the effect of inflation on original cost. It has not been available to individuals since April 1998, but companies receive it up to the date of disposal. Individuals, instead, are compensated to some extent by receiving taper relief, which companies do not receive. Taper relief is a defined percentage deduction from a gain, determined by length of ownership after April 1998
Table 6.13 Model of individual’s tax choice £ Disposal proceeds Less: Incidental disposal costs Less: Allowable expenditure (usually base costs) Unindexed gain Less: Indexation allowance to April 1998 Gain before loss relief and taper relief Less: Current year allowable losses Less: Allowable losses b/f at beginning of tax year Less: Taper relief Less: Annual exemption for current tax yeara Taxable gains for current year a
X (X) X (X) X (X) X (X) X (X) X (X) X (X) X
This is deducted from total gains arising for the individual in any tax year, and it may be that gains that have already been realized have fully or partly utilized it. Hence it may not be available at all, or available to the extent unutilized by other gains, and may be particularly significant in an individual’s decision as to whether to accept loan notes or not.
Mix-and-match facilities and loan notes in acquisitions
159
Table 6.14 Model of corporate body’s tax choice £ Disposal proceeds Less: Incidental disposal costs Less: Allowable expenditure Unindexed gain Less: Indexation allowance to date of disposal Gain before loss relief Less: Current year allowable losses Less: Allowable losses b/f at beginning of tax year Taxable gains for current year
X (X) X (X) X (X) X (X) X (X) X
(the date indexation allowance ceased to be available for individuals) and whether the asset disposed of is a business or non-business asset (again, as defined). Moreover, individuals receive an annual exemption to reduce the overall gain, which companies do not receive. The following example shows the effect of these differences on the calculation of a gain with respect to the same purchase of shares, but with different ownership—an individual and a company, respectively (assuming no other gains or losses in the same period and no losses brought forward). Example: Shares in a listed company were purchased on 1 April 1983 for £25,000.12 The shares were sold on 31 August 2004 for consideration of £100,000 as a result of a takeover of the company by another. (To facilitate simplicity of calculation, assume that no incidental disposal costs are incurred, and that in the case of the individual, the shares were not a business asset.) Gain arising: The gain arising for this example is given in Table 6.15
For companies, the rate of tax is also connected to due date of payment, in a way that it is not for individuals. Companies paying the full rate of corporation tax (typically “large” companies with profits above the stipulated limit13) will pay corporation tax in four quarterly installments on the basis of their estimated tax liability for the accounting period.14 Companies paying small (19%) and 12
One must note that there are complex “matching rules” and pooling rules which may apply where both individuals and companies acquire shares over a period of time. For the sake of simplicity, as this is an illustrative example only, it should be assumed that the shares purchased on 1 April 1983 are the only shares owned. 13 £1,500,000 per annum, adjusted as necessary for the number of associated companies in a group (as defined) and for any reduction in accounting period to less than one year in length. 14 A “large” company is not liable to pay tax by instalments for an accounting period if it was not “large” (i.e., profits less than the stipulated limit) in the previous period and its taxable profits for the current accounting period do not exceed £10 million per annum (divided by the number of associated companies, as calculated at the end of the previous period).
160
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 6.15 Gain arising for example disposal Individuala
Company
£
£
Disposal proceeds Less: Allowable expenditure: Purchase price
100,000
100,000
(25,000)
(25,000)
Unindexed gain Less: Indexation allowance Acquisition expenditure: Individual: £25,000 ⫻ 0.929b [ ⫽ (162.6 – 84.28)/84.28] Company: £25,000 ⫻ 1.224 [ ⫽ (187.4 – 84.28)/84.28]
75,000
75,000
Gain Taper relief: £51,775 ⫻ 25% (including bonus year)
51,775 (12,944)
44,400 –
Less: Annual exemption
38,831 (8,200)
44,400 –
Chargeable gain
30,631
44,400
(23,225) (30,600)
a
Strictly, the calculation of the gain for an individual would show the “indexed cost” for shares in a case such as this. This means that the indexation allowance would be added to the purchase cost, and the total deducted from disposal proceeds. The effect is the same, however. b The indexation factors here are calculated following the permitted formula using changes in the Retail Price Index to adjust the original cost for inflationary changes. Strictly as a result of a loophole in relevant legislation, the indexation factor on disposal of shares is not restricted to three decimal places, as it is on disposals of other assets. However, this restriction is imposed here to make the calculation simpler.
marginal relief rates of corporation tax are not required to pay it by installments.15 Thus large companies will have to estimate their corporation tax for each period as the first two installments are due even before the period has ended. Despite these complications and differences, it is clear that both companies and individuals could legally reduce their tax liability on any gain arising if loan notes were offered in lieu of cash. Assuming the terms of the offer allowed target shareholders to have cash/loan notes to the extent they might choose, and to redeem them on demand (subject to all loan notes being redeemed within, say, 10 years), they might not each desire the same amount of consideration as loan notes. The individual could quite easily reduce the gain to below the level of the annual exemption by choosing to accept cash consideration now for one-fifth 15
Companies with profits (adjusted, as appropriate, per footnote 13) between £0 and £300,000 are taxed at 19 per cent, between £300,001 and £1,500,000 at 30% less marginal relief and over £1,500,000 at 30%.
Mix-and-match facilities and loan notes in acquisitions
161
Table 6.16 Individual’s calculation showing reduction of gain as a result of taking up one-fifth of consideration as cash now (and loan notes for the next four years) £ Disposal proceeds Less: Allowable expenditure: Purchase pricea
20,000
Unindexed gain Less: Indexation allowance Acquisition expenditure: Individual: £5,000 ⫻ 0.929 [ ⫽ (162.6–84.28)/84.28] Gain before taper relief Taper relief: £10,355 ⫻ 25% (including bonus year)
15,000
Less: Annual exemption (part remaining unused) Chargeable gain
(5,000)
(4,645) 10,355 (2,589) 7,766 (7,766) –
a
Allowable expenditure in a part disposal is calculated by multiplying the base cost by the fraction A/(A ⫹ B), where A ⫽ the value of the part disposed of (i.e., the disposal value), and B ⫽ the value of the part remaining in the taxpayer’s ownership.
of the shares and the loan notes as the remainder, with a view to redeeming them over the next four years. This is shown in Table 6.16. The company’s situation would be more difficult, as there is no annual exemption to eliminate any gain. To the extent it cannot be reduced by other capital losses, the gain will always be taxable. The company will therefore choose a combination of cash and/or loan notes, which at least does not take profits into a band taxable at higher rates. Conceivably a gain of £42,500, if the company’s profits are near the upper level of any band, could increase profits up to the next level, such that they might become taxable at higher rates. Certainly, it would seek to preclude any need to make payments on account for the current year and future years, if it were not doing so already.
6.8
Conclusion
This chapter has looked at acquisition deals that offer the target shareholders a choice of different forms of consideration and also frequently allow the target shareholders to mix and match the different forms of consideration. This chapter is based on a pilot study, the obvious shortcomings of which include limited data coverage. Nevertheless, we have uncovered some interesting patterns and stylized facts. The chapter also contributes to the literature in a major way by studying loan notes used as a means of payment in acquisitions. Although loan notes are frequently offered as part of the consideration, to date the literature has largely ignored this form of consideration.
162
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
First, there was an increase in the popularity of loan notes over the late 1990s and early years of the 21st century. Second, contrary to what theories on asymmetric information and findings from empirical studies on wealth gains suggest, target shareholders do not always choose (exclusively) cash instead of shares. Third, there are clear benefits derived from loan notes issues to both the target shareholders and bidders. More precisely, using the tax models developed, we show that, in certain circumstances, the tax position of target shareholders may make loan notes a more attractive choice than other types of consideration. Furthermore, bidders derive benefits from loan notes in the form of improved cash management and cheap financing of acquisitions.
Table of statutes Great Britain. Taxation of Chargeable Gains Act: Elizabeth II. Chapter 12 (1992) London: The Stationery Office.
References Eckbo, E., and Thorburn, K. (2000). Gains to Bidder Firms Revisited: Domestic and Foreign Acquisitions in Canada. Journal of Financial and Quantitative Analysis, 35:1–25. Franks, J., and Harris, R. (1989). Shareholder Wealth Effects of Corporate Takeovers: The UK Experience 1955–85. Journal of Financial Economics. 23:225–249. Franks, J., Harris R., and Mayer, C. (1988). Means of Payment in Takeovers: Results for the United Kingdom and the United States. In Corporate Takeovers: Causes and Consequences (A.J. Auerbach, ed.). Chicago: Chicago University Press. Goergen, M., and Renneboog, L. (2004). Shareholder Wealth Effects of European Domestic and Cross-Border Takeover Bids. European Financial Management, 10:9–45. Healy, P., Palepu, K., and Ruback, R. (1992). Does Corporate Performance Improve after Mergers? Journal of Financial Economics, 31:135–175. Jarrell, G., and Poulsen, A. (1989). The Returns to Acquiring Firms in Tender Offers: Evidence from Three Decades. Financial Management, 18:12–19. Jensen, M., and Ruback, R. (1983). The Market for Corporate Control: The Scientific Evidence. Journal of Financial Economics, 11:5–50. Loderer, M., and Martin, K. (1990). Corporate Acquisitions by Listed Firms: The Experience of a Comprehensive Sample. Financial Management, 19:17–33. Maquiera, C., Megginson, W., and Nail, L. (1998). Wealth Creation versus Wealth Redistributions in Pure Stock-for-Stock Mergers. Journal of Financial Economics, 48:3–33.
Mix-and-match facilities and loan notes in acquisitions
163
Mitchell, M., and Stafford, E. (2000). Managerial Decisions and Long-Term Stock Price Performance. Journal of Business, 73:287–329. Mulherin, J., and Boone, A. (2000). Comparing Acquisitions and Divestitures. Journal of Corporate Finance, 6:117–139. Schwert, G. W. (2000). Hostility in Takeovers: In the Eyes of the Beholder? Journal of Finance, 55:2599–2640. Sirrower, M. (1997). The Synergy Trap: How Companies Lose the Acquisition Game. New York: The Free Press. Sudarsanam, S. (1995). The Role of Defensive Strategies and Ownership Structure of Target Firms: Evidence from UK Hostile Takeover Bids. European Financial Management, 1:223–240. Sudarsanam, S. (2003). Creating Value from Mergers and Acquisitions—The Challenges. Harlow (England): FT Prentice Hall. Walker, M. (2000). Corporate Takeovers, Strategic Objectives and Acquiring-Firm Shareholder Wealth. Financial Management, 29:53–66.
References to reports and accounts Abbey National plc, Annual Report and Accounts for the year ended 31 December 1997. Alexon plc, Report and Accounts for the period ended 29 January 2000. Allied Leisure plc, Annual Report and Accounts for the year ended 30 June 2000. Anglian Water plc, Annual Report and Accounts for the year ended 31 March 1998. Anglo American plc, Annual Report and Accounts for the year ended 31 December 2000. Arlen plc, Annual Report and Accounts for the year ended 31 December 1998. AWG plc, Annual Report and Accounts for the year ended 31 March 2001. British Vita plc, Annual Report and Accounts for the year ended 31 December 1998. BSS Group plc, Annual Report and Accounts for the year ended 31 March 2000. Bunzl plc, Annual Report and Accounts for the year ended 31 December 1999. Burton Group plc, Annual Report and Accounts for the year ended 31 August 1996. Caledonia Investments plc, Annual Report and Accounts for the year ended 31 March 2000. Chrysalis Group plc, Annual Report and Accounts for the year ended 31 August 2001. Coats Viyella plc, Annual Report and Accounts for the year ended 31 December 1999. Delancey Estates plc, Annual Report and Accounts for the year ended 31 March 2000.
164
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Ennstone plc, Annual Report and Accounts for the year ended 31 December 2000. Ennstone plc, Annual Report and Accounts for the year ended 31 December 2004. Enterprise Inns plc, Annual Report and Accounts for the year ended 30 September 1998. Finelist Group plc, Annual Report and Accounts for the year ended 30 June 1998. Greene King plc, Report and Accounts for the period ended 29 April 2000. Greene King plc, Report and Accounts for the period ended 4 May 2002. Hanson plc, Annual Report and Accounts for the year ended 30 September 1995. Hanson plc, Annual Report and Accounts for the year ended 30 September 1996. Hill & Smith Holdings plc, Annual Report and Accounts for the year ended 30 September 2001. Hillsdown Holdings plc, Annual Report and Accounts for the year ended 31 December 1996. Homestyle Group plc, Report and Accounts for the period ended 3 March 2001. Invensys plc, Report and Accounts for the period ended 2 January 1993. Johnson Matthey plc, Annual Report and Accounts for the year ended 31 March 2002. Laporte plc, Report and Accounts for the period ended 31 March 1999. L. Gardner Group plc, Annual Report and Accounts for the year ended 31 August 1999. Marlborough Stirling plc, Annual Report and Accounts for the year ended 31 December 2001. Misys plc, Annual Report and Accounts for the year ended 31 May 2002. NSB Retail Systems plc, Annual Report and Accounts for the year ended 31 December 2000. Pearson plc, Annual Report and Accounts for the year ended 31 December 2001. Persimmon plc, Annual Report and Accounts for the year ended 31 December 2001. Prudential plc, Annual Report and Accounts for the year ended 31 December 1999. Redrow plc, Annual Report and Accounts for the year ended 31 December 2002. RMC Group plc, Annual Report and Accounts for the year ended 31 December 2000. Rolls–Royce plc, Annual Report and Accounts for the year ended 31 December 1999. Ryland Group plc, Annual Report and Accounts for the year ended 31 December 1999. Scottish Media Group plc, Annual Report and Accounts for the year ended 31 December 1999. Slough Estates plc, Annual Report and Accounts for the year ended 31 December 1998. Sportsworld Media Group plc, Annual Report and Accounts for the year ended 30 June 2000. The Berkeley Group plc, Annual Report and Accounts for the year ended 30 April 1991.
Mix-and-match facilities and loan notes in acquisitions
165
The Wolverhampton & Dudley Breweries plc, Report and Accounts for the period ended 30 September 2000. Westbury plc, Annual Report and Accounts for the year ended 28 February 2003. Whitbread plc, Report and Accounts for the period ended 4 March 2000. Wolseley plc, Annual Report and Accounts for the year ended 31 July 1999. WPP Group plc, Annual Report and Accounts for the year ended 31 December 2001.
This page intentionally left blank
Part Two Special Types of Mergers and Acquisitions
This page intentionally left blank
7 Mergers and acquisitions in IPO markets: evidence from Germany David B. Audretsch and Erik E. Lehmann
Abstract This chapter analyzes empirically the determinants of mergers in the high-technology and knowledge-based sector. Using a unique dataset of 285 listed companies, we analyze empirically whether and how takeovers in the technology sector could be explained by firm characteristics. Our results show that the likelihood of being taken over within three years after IPO is significantly affected by the percentage of shares held by the initial owners at IPO. Our results also show that, in contrast to earlier studies, firm characteristics such as age, the market-to-book value, or the market value of assets do not significantly affect the likelihood of being a takeover target.
7.1
Introduction
It is a widely observed phenomenon that many public firms are acquired soon after their initial public offering (IPO). This IPO sell-off phenomenon is often described as a double-exit strategy. The question is why do firms forgo a private takeover, then sell soon after the IPO, even though the cost of going public accounts for more than 10% of the funds raised (Ritter, 1987). This chapter empirically analyzes the determinants of being acquired within three years after an IPO. In particular, we test whether firm characteristics affect the likelihood of a firm’s being an acquisition target. Although small business have long been recognized as a major source of change within industries, relatively little systematic research has been devoted to the study of the acquisition of small business. One of the major findings in this research is that smaller firms have a higher likelihood of being takeover targets than do larger firms (Palepu, 1986; Mulherin and Boone, 2000). Among the chief purposes of merger and acquisition (M&A) studies has been the effort to understand how different acquisition motives influence takeover activities and firm performance (see Andrade, Mitchell, and Stafford, 2001, for an excellent survey; Palepu, 1986; Agrawal and Jaffe, 2002; Field and Karpoff, 2002). Another strand of the literature compares private and public targets, finding that bidders choose to acquire public targets rather than private targets when acquiring young firms to engage in inter-industry transactions (Shen and Reuer, 2005) or that mergers
170
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
with private targets show positive shareholder wealth benefits for bidders and high premiums for privately held targets (Ang and Kohers, 2001). While financial research on IPOs has largely focused on potential challenges to the efficiency of capital markets due to potential anomalies such as underpricing, hot issue markets, and the long-run underperformance of newly public companies (Ritter and Welch, 2002), emerging findings also suggest that IPOs have implications for M&A markets since newly public firms exhibit a higher propensity of being acquired (Pagano, Panetta, and Zingales, 1998; Field and Mulherin, 1999; Reuer and Shen, 2003; Dai, 2005). The dot-comera and, in particular, the enormous number of M&As observed in the high-technology markets, stimulated researchers to analyze acquisition activities in IPO markets (Field and Karpoff, 2002; Dai, 2005). In this chapter we follow this strand of research, analyzing whether and how firm characteristics are useful to predict acquisition targets in the IPO market. We use a unique, handcollected dataset of 285 German firms that conducted their IPOs from March 1997 to March 2001. We found that nearly 12% were acquired before June 2002. The remainder of the chapter is organized as follows. The next section summarizes some of the findings of the empirical research analyzing firm characteristics as a predictor of becoming an acquisition target. Section 7.3 describes the sample and data used in this chapter. Section 7.4 examines why some firms are acquired so soon, given the costs of going public and the effect of firm characteristics on the likelihood of being acquired. Section 7.5 summarizes.
7.2
Predicting takeover targets: a short review of the literature
We start this short review of the literature with a puzzle of the empirical findings. Although the results reported by earlier studies (Belkoi, 1978; Dietrich and Soerensen, 1984; Palepu, 1986) indicate that statistical models using publicly available information are impressively capable of predicting acquisition targets long before the announcement of takeovers, the stock market seems incapable of predicting acquisition targets with the same high degree of accuracy, even in the short time prior to the announcement of takeover bids. Following Palepu (1986: 3), most of these earlier studies report prediction accuracies ranging from 70 to 90%. If the claims of these studies are valid, it should be possible to earn abnormal rents using the prediction models. Palepu (1986) undertakes a critical examination of the methodology used by earlier studies, showing that some principal methodology flaws make their reported prediction accuracies unreliable. Using different estimation and sampling techniques, Palepu (1986) shows that firm size, industry, excess returns, and growth are significant predictors of becoming an acquisition target; he also shows that predicting targets based on his findings does not lead to abnormal rents on the stock markets. He concludes that his prediction results show that investing in the potential targets
Mergers and acquisitions in IPO markets: evidence from Germany
171
identified by the model does not yield significant excess returns and that the model’s ability to predict takeover targets is not superior to that of the stock market. Recently, Dai (2005) studied whether venture-capital backing conditions the post-IPO acquisition behavior. Among a sample of 638 technology firms, he found that nearly 23% are acquired within the first three years of going public. Controlling for firm and industry characteristics, he showed that venture capital–backed firms are more likely to be acquired within this three-year period than other firms. In interpreting his results, he posits that venture capitalists sell their stakes more aggressively than do controlling initial owners in non-venture capital–backed firms. Thus, taking the firm public will be not the end of mission for venture capitalists. Since venture capitalists cannot sell their whole stakes at the IPO owing to lock-up agreements with the underwriter, venture capitalists face the decision of how to unload their investments when the lock-up period expires. Thus, Dai (2005) found that the duration from IPO to the date of acquisition is much shorter for venture capital–backed firms than for non-backed firms. However, he could not find that firm characteristics significantly and robustly influence the likelihood of being acquired. Only firm age seems to reduce the likelihood of being acquired, meaning that elder firms are less likely to be acquired than younger firms. Brau, Francis, and Kohers (2005) analyze the firm owner’s choice between an IPO and a takeover by a public acquirer. Using a sample of more than 9.500 U.S. privately held firms, they address the IPO-versus-takeover issue by examining market-timing, industry, deal-specific, and fund-demand factors of the IPOversus-acquisition choice. Their results show that the concentration of the industry, the high-tech status of the private firm, the current cost of debt, the “hotness” of the IPO market relative to the takeover market, the percentage of insider ownership, and the size of the firm are all positively related to the probability that a firm will conduct an IPO. In contrast, private companies in high marketto-book industries, firms in financial service sectors, firms in highly leveraged industries, and companies pursuing deals involving greater liquidity for selling insiders show a stronger likelihood for takeovers. Lian and Wang (2006) examine why dual-tracking firms—private firms entertaining acquisition offers at the same time as preparing for initial public offerings (IPOs)—withdraw from their IPO registrations after spending considerable money and effort preparing for IPOs, only to be purchased by public acquirers. Thus, going public can be viewed as a signaling mechanism that discriminates high- from low-quality firms, but also as a focal point for potential acquirers since small-business firms are often difficult to locate as potential targets. They argue that, in dual tracking, filing for IPO registrations reduces the valuation uncertainty between withdrawn IPO private targets and their bidders. Their findings show that withdrawn-IPO private targets sell at a 47% acquisition premium relative to comparable pure private targets that never file for IPO registrations. Despite paying significantly higher prices for withdrawn-IPO private targets, acquirers of withdrawn-IPO private targets still earn a substantial average abnormal
172
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
announcement return of 2.5%, which is close to the 2.56% abnormal announcement return earned by acquirers of pure private targets. Their findings are in line with studies comparing private and public takeovers, showing that private takeovers exhibit relatively high premiums for privately held firms (Ang and Kohers, 2002; Shen and Reuer, 2005). In contrast, Field and Karpoff (2002) analyze takeover defenses of firms when they go public. They find that IPO managers are more likely to deploy defenses when their compensation is high, shareholdings are small, and oversight from nonmanagerial shareholders is weak. The presence of a takeover defense at the time of the IPO is negatively related to subsequent acquisition likelihood, yet has no impact on takeover premiums for firms that are acquired. Their results do not support arguments that takeover defenses facilitate the eventual sale of the IPO firm at high takeover premiums. In interpreting their results, the authors suggest that managers shift the cost of takeover protection onto nonmanagerial shareholders, thereby concluding that agency problems are important even for firms at the IPO stage.
7.3
Data and the sample
During the period from 1997 and 2001, about 341 firms went public in Germany. After excluding foreign issuers, banks, and holding companies, the final sample consists of 275 IPOs. For each IPO firm, we collected individual data from IPO prospectuses, along with publicly available information from on-line data sources including the Deutsche Boerse AG (www.deutsche-boerse.com). We record the IPO date, the proceeds to the company, the percentage of firm stock held by CEOs, the board of directors, venture capitalists, banks, firms, and the freefloat before and after IPO, as well as firm characteristics such as age, total assets, and number of employees. From 1997 until 2002 we observed about 65 delistings (see Audretsch and Lehmann, 2005). From these firms, 23 were acquired soon after their IPO. On average, it needs about 22 month before a firm is acquired or taken over on the stock market. This, however, is much longer than the lock-up agreement as mentioned by Dai (2005). Firm characteristics at the time of IPO are examined in Table 7.1 and allow a comparison of firms being acquired and those not being acquired. Column 3 in Table 7.1 predicts the results from a two-tailed test of mean comparisons. First, we find that firms being acquired are significantly older than those not being acquired. Firm age is an important variable characterizing small and young firms (Audretsch, Keilbach, and Lehmann, 2006). Newly founded firms will tend to exhibit higher uncertainty and higher degrees of information asymmetry than do established firms. These factors imply that it will be difficult to assess the value of younger firms and thus may increase the costs of the acquisition of young firms; moreover, potential investors may also overestimate the value of the targets. While the average age of the full sample is about 9 years, with a median of 8 years, most of the firms that are acquired after IPO are much older. The average age of the firms acquired is about 10.7 years, while the average age of the firm that is not acquired is about 6.6 years. Recently, while Dai
Mergers and acquisitions in IPO markets: evidence from Germany
173
Table 7.1 Characteristics of Firms issuing IPOs during the period 1997 to 2000a Variable
Takeover N ⫽ 23 Mean (Std. err.)
Non-takeover 261 Mean (Std. err.)
Ha: diff ⬎ ⬍ 0 t-value
Firm age Size (employees) Firm patents Market-book ratio Market value of assets at IPO Underpricing Ownership CEO (post) Ownership board of directors Ownership venture capitalists Ownership banks Ownership firms Freefloat
10,731 (0.6323) 345.95 (90.43) 165 (1.21) 0.19 (0.059) 51.4 Mio € (9,19 Mio) 45.18 (16.33) 23.043 (4.091) 6.95 (2.709)
6.63 (1.342) 282.39 (24.399) 3.48 (0.69) 0.33 (0.062) 63.8 Mio € (9,99 Mio) 54.42 (4.77) 23.77 (81.3939) 4.709 (0.675)
1.882** 0.7376 0.737 0.568 0.366
14.001 (4.329)
8.093 (0.944)
1.726**
2.383 (1.809) 6.18 (3.597) 29.01 (2.543)
0.817 (0.223) 8.079 (1.115) 38.384 (1.61)
1.684* 0.462 2.3021**
0.5509 0.150 0.953
a
This table presents descriptive statistics of the included variables. The fourth column contains the results from two-sample t-test statistics. While the underlying null hypothesis is that no significant differences exist between the group means, the alternative hypotheses test whether those differences are significantly different from zero. **,**,* Significant at the 1%, 5%, and 10% levels, respectively.
(2005) could not find significant differences in firm age across acquired IPO and IPO not acquired, the firms in our sample differ significantly. However, average firm age is nearly the same as that Dai (2005) reports for his sample of 638 IPOs. However, he finds significant differences across venture-backed firms and non– venture-backed firms (see also Audretsch and Lehmann, 2004, for German IPO). The average number of employees as a measure for firm size is about 345 for the acquired firms and 282 for the firms not being acquired. Although there is a difference of about 20%, this difference is not statistically significant on the 10% level. The level of asymmetric information also hinges upon the nature of the resources being acquired. Information about the value of undifferentiated physical assets can be conveyed more easily, so valuation problems will be lower for the acquisition of such resources (Shin and Reuer, 2005). In contrast, acquiring firms may be more interested in the intangible resources to obtain synergies, which, however, increases the likelihood of an ex ante misinterpretation of the value of intangible assets. We include the number of patents owned by a firm as a measure and value of the intangible assets. Although the average number of acquired firms is about 1.6 and thus lower than in an IPO not being acquired, the difference is not statistically significant on the 10% level. The mean market capitalization at the first day of IPO is 51.4 million € for firms being acquired and about 63.8 million € for firms not being acquired. Although the difference is about 10 million €, there are no statistically significant
174
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
differences between both groups of firms. However, compared to the mean market capitalization at the first day of 755 million U.S.$, as shown by Dai (2005), German IPO realize a significant smaller market capitalization. The market-to-book ratio is the ratio of the sum of market capitalization at the first day of IPO and the difference between total assets and book value of equity in the end of fiscal year right before IPO. On average, firms being acquired show a lower market-to-book ratio with 19% compared to 33% for firms not being acquired. The median over all firms is about 23%. As before, although large, this difference is not statistically significant. Compared to the sample of 638 IPO used by Dai (2005), the average market-to-book value is rather low. While the mean market-to-book ratio is about 71.9, which is about 3 times the ratio of the mean value of German IPO, the median value with 22.9 is quite similar to the median value of our sample. The large difference indicates that some firms in the United States are valued more highly for their intangible assets or growth opportunities compared to those in Germany. This is also reflected by the large differences in market capitalization as mentioned previously. The average first-day return or underpricing at IPO is about 50% for the whole sample. Dai (2005) reports a mean initial return of 67.1%, while Ljungqvist and Wilhelm (2003) report a mean initial return of 35.7%. Thus, the underpricing in this sample lies within the range of findings from U.S. IPOs. Although the mean initial return is higher in firms not being acquired, the difference is not statistically significant on the 10% level. Finally, we take a closer look on the governance structure of the IPO firms, as expressed by their major shareholders. Firms being acquired differ significantly in the percentage of shares held by venture capitalists. On average, 14% of the equity of firms being acquired is in the hand or under the control of venture capitalists, compared to only 8% of the firms not being acquired. Although banks held significant larger equity stakes in firms being acquired, the percentage of shares owned by banks is rather small (see Audretsch and Lehmann, 2005a, Lehmann, 2006). The largest shareholder in both groups is the CEO, who holds about 23% of the shares. In summary, firms being acquired differ significantly from firms not being acquired in their age and the percentage of shares owned by venture capitalists. However, they do not differ significantly in their size, as measured by the number of employees or in the market value of assets at the time of IPO. In the next section, we analyze how firm characteristics may influence the likelihood of being acquired after IPO.
7.4
Empirical results
In this section we use logistic regressions to test various firm characteristics that affect the likelihood of a firm’s being an acquisition target. Since our analysis is more exploratory, we did not test different hypotheses. Most of the included variables may affect the likelihood of being an acquisition target in two different
Mergers and acquisitions in IPO markets: evidence from Germany
175
ways, and we are not able to discriminate among the positive and negative effect with this dataset. Let us take the freefloat of shares as an example. If a firm’s shares are widespread, a potential acquirer could buy the shares from the stock market. Otherwise, the well known moral hazard problem, as described by Holmström (1982), increases the costs of the acquisition. Thus, the percentage of shares that are widespread could increase or decrease the likelihood of being a takeover target. The same holds for large equity shareholders, such as venture capitalists, managers, other firms or banks, which, as blockholders, can extract part of the private benefits of control from the buyer through direct bargaining. Thus, the existence of blockholders positively affects the likelihood of being an acquisition target. However, the opposite also holds that large blockholders are less reluctant to sell their equity stakes, thus decreasing the likelihood of being taken over (Field and Karpoff, 2002). To examine whether the likelihood of being acquired within the next three years after IPO is influenced by firm characteristics, we use logistic regression and estimate the following regressions: Prob(being acquired) ⫽ f[lnage, lnsize, patents, ln(asset value), marketto-book value, underpricing, ownership CEO, ownership venture capital, ownership banks, ownership firms, freefloat) ⫹ error term] The dependent variable is the likelihood of being acquired within three years of IPO. The independent variables are taken from Table 7.1. Most of the variables like size, age, market-to-book-value, underpricing, or the market value of assets are similar to other studies (Palepu, 1986; Shen and Reuer, 2005; Dai, 2005), thus allowing a detailed comparison of the results. In addition to most of the empirical studies, we include variables characterizing the governance of a firm by its largest shareholders. In particular, we estimate the preceding regression in three different ways. The first regression (I) is without industry and IPO dummies. Then we include industry dummies (II), and finally we added IPO dummies (III). To avoid problems of multicollinearity the variable indicating the percentage of ownership by the CEO is captured by the intercept. The results of the regression analysis are presented in Table 7.2. In the first regression (column I), age and size are firm characteristics that affect the likelihood of a firm’s being an acquisition target. Thus, the likelihood of being acquired increases with firm size, as measured by the natural log of employees, and decreases with firm age. As in Palepu (1986), the likelihood of being an acquisition target is affected by firm characteristics. In contrast, we could not find significant results for the market value of assets, the underpricing, or the market-to-book value of the firm. If we include industry dummies (column II) and IPO dummies (column III), the significant impact of firm age on the likelihood of being acquired within three years after IPO disappears. However, firm
176
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 7.2 Likelihood of being a takeover targeta Variable
(I)
(II)
(III)
LnAge LnSize Market–Book Ratio LNmarketvalue Underpricing Patents Ownership venture capitalists Ownership banks Ownership firms Freefloat Industry dummies IPO Dummies Intercept Pseudo R2 Log pseudo LL
⫺0.335 (1.93)* 0.389 (1.81)* ⫺0.239 (0.63) 0.143 (0.42) ⫺0.004 (1.439) ⫺0.015 (0.38) 0.004 (0.32)
⫺0.293 (1.43) 0.6229 (2.57)** ⫺0.042 (0.18) 0.1304 (0.31) ⫺0.004 (1.50) 0.025 (0.91) ⫺0.002 (0.12)
⫺0.298 (1.54) 0.645 (2.48)** ⫺0.127 (0.48) 0.321 (0.87) ⫺0.003 (1.18) 0.005 (0.16) ⫺0.001 (0.04)
0.0885 (2.02)** ⫺0.027 (1.52) ⫺0.047 (2.16)** No No ⫺4.194 (0.82) 0.1385 48.612
0.095 (2.42)** ⫺0.042 (1.84)* ⫺0.009 (2.36)** Insignificant No ⫺5.958 (0.81) 0.1847 44.1848
0.098 (2.47)*** ⫺0.042 (1.91)* ⫺0.067 (2.78)*** Insignificant Insignificant ⫺8.997 (1.42) 0.1888 42.367
a
This table presents logistic regression analysis. z-Values are in parentheses. ***,**,* Significant at the 1%, 5%, and 10% levels, respectively. Column (I) shows the results without industry and IPO dummies. Column (II) shows the results with industry dummies, and column (III) shows the results with industry and IPO dummies.
size still remains significant. Thus, larger firms show a higher likelihood of being taken over than smaller firms. However, neither the market-to-book value, the market value of assets at the time of IPO, the underpricing, nor the number of patents significantly affect the likelihood of being a takeover target. In contrast, Dai (2005) reports a significant and negative impact of the first-day return (underpricing) and the market value of the assets at the time of IPO on the likelihood of being a takeover target. If we take the number of patents as a proxy variable for the innovation capacity of a firm, our results contradict findings from Lehto and Lehtoranta (2006) for Finland. They report that in some industries M&As are used to transfer technology among firms. Next, we include the distribution of ownership shares into the regression estimations. Some empirical evidence supports the role of venture capitalists in selling their shares after the IPO (Dai, 2005). If an acquisition provides a better return than a direct selling in the open market, venture capitalists will prefer the former. Dai (2005) assumes that venture capitalists’ desire for liquidity and capital gains will lead to their relatively more aggressive exit behavior after they take the firm public. Thus, by selling their stakes to an acquirer, venture capitalists are able to gain immediate liquidity if the acquirers are willing to pay cash directly.
Mergers and acquisitions in IPO markets: evidence from Germany
177
However, in contrast to Dai (2005), we could not find a significant effect of venture capitalists on the likelihood of being taken over after the IPO. While Dai (2005) uses a dummy variable indicating whether a firm is a venture capital–backed firm or not, we included the percentage of equity shares hold by venture capitalists. If we use this dummy variable instead of the percentage of ownership held by venture capitalists, we receive the same results as Dai (2005). Interestingly, the variables indicating the different blockholders significantly influence the likelihood of being taken over after the IPO. The impact of banks as blockholders significantly increases the likelihood of being a takeover target; but when other firms are major shareholders the likelihood of being taken over decreases. Also, a high freefloat of the shares decreases this likelihood. Like that of venture capitalists, the major interest of banks is to unload their investments to receive liquidity and capital gains. In contrast, other firms may receive additional benefits from their investment via spillovers and other resources, and thus may be less reluctant to sell their shares to acquirers. Table 7.2 also shows that a high freefloat of the shares decreases the likelihood of being a takeover target. Finally, we could not find a significant impact of the industry dummies or the year of IPO on the likelihood of being taken over.
7.5
Conclusion
Many newly public firms are acquired within a short period after their IPO. Among our sample of 285 high-technology and knowledge-intensive firms that went public from 1997 through 2000, about 12% of the firms were acquired within three years of their IPO. This study adds to previous research about the determinants of newly public firms’ acquisition by providing evidence that the identity of the controlling initial owners—and thus the corporate governance structure of firms—affects the likelihood of acquisitions after IPO. Initial owners, such as venture capitalists, managers, banks, or corporations, behave differently after the firm goes public owing to their different objective functions. Although venture capitalists may desire liquidity and returns more than control of the firm, we could not find a significant impact of the percentage of equity held by venture capitalists on the likelihood of being an acquisition target within three years after IPO. However, we find that banks, as initial owners of IPO firms, significantly increase the likelihood of being taken over after IPO. Corporations may favor control of the firm more than the immediate exit, and thus may be less reluctant to sell their stakes to an acquiring firm. The lack of significant influence of firm characteristics on the likelihood of acquisitions, as found in earlier studies, may suggest that the motive for acquisitions of newly public firms may differ from that behind other acquisitions. There may be several special factors driving the acquisitions of newly public firms. One factor that conditions the likelihood of newly public firms, as shown
178
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
in the regression analysis, is the percentage of equity held by the different initial owners at time of IPO.
References Agrawal, A., and Jaffe, J. F. (2002). Do Takeovers Targets Under-Perform? Evidence from Operating and Stock Returns. Working Paper. University of Alabama. Andrade, G., Mitchell, M., and Stafford, E. (2001). New Evidence and Perspectives on Mergers. Journal of Economic Perspectives, 15:103–120. Ang, J., and Kohers, N. (2001). The Take-Over Market for Privately Held Companies: The US Experience. Cambridge Journal of Economics, 25:723–748. Audretsch, D. B., and Lehmann, E. E. (2004). Debt or Equity: The Role of Venture Capital in Financing High-Tech Firms in Germany. Schmalenbach Business Review, 56:340–357. Audretsch, D. B., and Lehmann, E. E. (2005). The Effects of Experience, Ownership, and Knowledge on IPO Survival: Empirical Evidence from Germany. Review of Accounting and Finance, 4:13–37. Audretsch, D. B., Keilbach, M., and Lehmann, E. E. (2006). Entrepreneurship and Economic Growth. Oxford University Press: Oxford. Belkoui, A. (1978). Financial Ratios as Predictors of Canadian Takeovers. Journal of Business Finance and Accounting, 5(1):93–107. Brau, J. C., Francis, B. C., and Kohers, N. (2003). The choice of IPO versus Takeover. Empirical Evidence. Journal of Business, 76:583–612. Dai, N. (2005). The Double Exit Puzzle: Venture Capitalists and Acquisitions Following IPOs. Working Paper. University of Kansas. Dietrich, J. K., and Sorenson, E. (1984). An Application of Logit Analysis to Prediction of Merger Targets. Journal of Business Research, 12:393–402. Field, L. C., and Karpoff, J. M. (2002). Takeover Defenses at IPO Firms. Journal of Finance, 62:1857–1889. Field, L. C., and Mulerhin, H. J. (1999). Newly Public Firms as Acquisition Targets: A Comparison with Established Target Firms. Working Paper, Penn State University. Holmström, B. (1982). Moral hazard in Teams. Bell Journal of Economics, 13(2):324–340. Lehmann, E. E. (2006). Does Venture Capital Syndication Spur Employment Growth and Shareholder Value? Evidence from German IPO Data. Small Business Economics, 26:455–464. Lehto, E., and Lehtoranta, O. (2006). How Do Innovations Affect Mergers and Acquisitions—Evidence from Finland. Journal of Industry, Competition and Trade, 6:5–25. Lian, Q., and Wang, Q. (2006). The Dual Tracking Puzzle: When IPO Plans Turn into Mergers. Working Paper. University of Alabama.
Mergers and acquisitions in IPO markets: evidence from Germany
179
Ljungqvist, A., and Wilhelm, W. J. (2003). IPO Pricing in the Dot-com Bubble. Journal of Finance, 58:723–752. Martynova, M., and Renneboog, L. (2006). Mergers and Acquisitions in Europe. Working Paper ECGI No. 114/2006. Mulherin, J. H., and Bonne, A. L. (2000). Comparing Acquisitions and Divestitures. Journal of Corporate Finance, 6:117–139. Pagano, M., Panetta, F., and Zingales, L. (1998). Why Do Companies Go Public? An Empirical Analysis. Journal of Finance, 53:27–64. Palepu, K. G. (1996). Predicting Takeover Targets. A Methodological and Empirical Analysis. Journal of Accounting and Economics, 8:3–35. Renneboog, L., and Szilagyi, P. G. (2006). How Do Mergers and Acquisitions Affect Bondholders in Europe? Evidence on the Impact and Spillover of Governance and Legal Standards. Working Paper ECGI No. 125/2006. Reuer, J. J., and Shen, J.-C. (2003). The Extended Merger and Acquisition Process: Understanding the Role of IPOs in Corporate Strategy. European Management Journal, 21:192–198. Ritter, J. (1987). The Costs of Going Public. Journal of Financial Economics, 19(2):269–281. Ritter, J. R., and Welch, I. (2002). A Review of IPO Activity, Pricing, and Allocations. Journal of Finance, 62:1795–1828. Shen, J.-C., and Reuer, J. J. (2005). Adverse Selection in Acquisitions of Small Manufacturing Firms: A Comparison of Private and Public Targets. Small Business Economics, 24:393–407.
This page intentionally left blank
8 Reverse mergers in the United Kingdom: listed targets and private acquirers1 Peter Roosenboom and Willem Schramade
Abstract Reverse mergers provide an alternative way of going public. In reverse mergers a private company de facto acquires a public target company whereas, de iure, it is the public company that acquires the private company. This way the private company, the acquirer in economic terms, can obtain a stock market listing for its shares via the “backdoor” without the costs and time it takes to conduct an IPO. Target firms in reverse mergers are typically unprofitable shell companies. We first study abnormal returns to reverse-merger announcements in the United Kingdom, as well as their determinants. Subsequently, we measure the long-run stock price and operating performance of reverse-merger firms and compare them with IPO firms. By definition, stock price reactions to reverse mergers can be measured only for the target firms. We find target returns to be significantly positive. Moreover, abnormal returns are higher when target firms are in bad financial condition. This is consistent with the existence of large, previously unused tax shields. In addition, abnormal returns increase as a function of the size of the private acquirer relative to the public target. When bidders are relatively large, their chances of successful completion of the takeover are deemed higher. The method of payment does not affect abnormal returns. We document that reverse-merger firms and matched IPO firms display similar stock and operating performance. This is inconsistent with the conventional wisdom that reverse mergers are mainly conducted by poorquality private firms that want to escape regulatory scrutiny. Thus, reverse mergers do not seem to deserve their bad name.
8.1
Introduction
In the 1990s, reverse mergers took a bad name in the United States: While good companies benefited from the hot equity markets by doing a regular Initial 1
Corresponding author is Willem Schramade: Erasmus Shool of Economics, PO Box 1738, 3000 DR Rotterdam, The Netherlands. Phone: ⫹31 10 408 1507, Fax: ⫹31 10 408 9165. Email:
[email protected]. We thank Osman Batal and Jasmeet Kamal Kanwal for excellent data assistance.
182
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Public Offering (IPO), many dubious firms made the transition from private to public by acquiring a listed shell company, often with disastrous results for investors.2 Reverse mergers can also be successful, however, as examples such as Berkshire Hathaway and Turner Broadcasting have shown. In this chapter, we consider reverse mergers in the United Kingdom, where the phenomenon is even more widespread than in the United States, accounting for about 19% of firms that go public.3 Prominent U.K. deals include the reverse mergers of football club Leicester City and PGA European Tour Courses, which exploits golf courses in Europe. A reverse merger is the event in which an unlisted company takes over a listed company. This can in fact be regarded as an alternative way of going public. Reverse mergers offer firms several potential advantages. By acquiring a listed firm, the private firm avoids the IPO process and its costs. Whereas an IPO requires a careful procedure and intense regulatory scrutiny, a reverse merger is faster and offers greater flexibility. In addition, firms have to pay investment banks high fees to assist them in conducting the IPO.4 Moreover, an IPO can be very difficult in poor stock market conditions. The advantages of reverse mergers do not apply equally to all firms. Owing to the large fixed component in underwriting fees, small firms suffer more heavily from the high fees of a top-tier investment banking oligopoly. Low-quality firms may also have increased incentives to avoid an IPO and conduct a reverse merger, as they stand to lose more from the intense scrutiny of the IPO process. Although cold equity markets are inconvenient to virtually any firm, large well-known firms can probably still go to the market. But access can be more difficult for small and low-quality firms, giving them all the more reason to resort to a reverse merger. Reverse mergers have received limited academic attention. The only published paper on reverse mergers we know of is Gleason, Rosenthal, and Wiggins (2005), who investigate U.S. reverse mergers during the 1990s. They find highly positive announcement returns for shareholders in the shell firms but poor subsequent performance. Since their sample is from the 1990s, it includes the abovementioned dubious firms that gave reverse mergers a bad name. This bad name is provided with theoretical support by Arellano-Ostoa and Brusco (2002), who build a model for the choice between a reverse merger and an IPO. They find a separating equilibrium in which high-quality firms choose the IPO and low-quality firms prefer a reverse merger. In this chapter, we investigate the wealth effects and long-run performance of a sample of 80 U.K. reverse mergers over the period 1990 to 2001. We extend the analysis in Gleason, Rosenthal, and Wiggins (2005) by comparing reverse mergers with a matched sample of IPOs. Furthermore, we look at reverse mergers 2
3
4
James Altucher, Mergers that go forward in reverse. Financial Times USA Edition, June 27, 2006, p.10. The London Stock Exchange had 1239 new admissions to trading during the period 1990–2001. Reverse mergers accounted for 233 new admissions (18.8%). Source: Our own calculations. Torstila (2003) reports that fees average 6% of proceeds in the United Kingdom.
Reverse mergers in the United Kingdom: listed targets and private acquirers
183
in the United Kingdom instead of the United States. The United Kingdom is interesting since it has the same high level of investor protection as the United States, but reverse mergers appear to be a more important mechanism in the United Kingdom than in the U.S. As in the United States (Gleason, Rosenthal, and Wiggins, 2005), we find that reverse takeovers typically concern unprofitable public companies that are taken over by healthier private companies. Stock price reactions to reverse-merger announcements are significantly positive. However, the average stock price reaction of 3.6% during the day before and the day of the announcement of the news of the reverse merger is much lower than the average stock price reaction of 15.6% in the United States reported by Gleason, Rosenthal, and Wiggins (2005). In contrast to Gleason, Rosenthal, and Wiggins (2005), we find that distressed targets have higher announcement returns, consistent with the exploitation of valuable tax shields. Moreover, stock price reactions are higher when the acquirer is larger relative to the target, since this increases the likelihood of success of the newly combined firm. In contrast to ordinary takeovers, announcement returns to reverse mergers are not affected by the method of payment. Conventional wisdom suggests that poor-quality private firms that want to escape regulatory scrutiny are likely to go public via a reverse merger. Highquality companies are likely to opt for an IPO. However, our results do not support this conjecture. We find that reverse mergers do not display worse long-term stock price performance or operating performance compared to a sample of matched IPO firms. The chapter proceeds as follows. Section 8.2 two explains how a reverse merger transaction works. Section 8.3 discusses data and methodology. Section 8.4 presents our empirical results. Section 8.5 offers conclusions.
8.2
Reverse-merger mechanics
A reverse merger is the event in which an unlisted company takes over a listed company. The term reverse merger derives from the fact that the buyer and seller roles reverse, depending on whether one looks at it from an economic or a legal perspective: the de facto acquirer is the de iure target. That is, while the private firm is the buyer in economic terms, the legal acquirer is in fact the listed (shell) firm, since this is the legal entity with the stock exchange listing that will subsequently be used. However, the private firm brings in the cash, management team, operational assets, and usually the name as well. Therefore, the private firm is the acquirer in economic terms. Throughout this chapter, we will use the economic definitions of acquirer and target. The main attraction of a reverse merger is the avoidance of the IPO process and its costs. The procedures surrounding the IPO process are more rigorous and time-consuming than those in a reverse merger. An IPO also attracts more intense regulatory scrutiny. As a result, a reverse merger is faster and offers greater flexibility than an IPO. Whereas an IPO takes at least 6 months to complete, a reverse
184
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
merger can be conducted within 3 months. In addition, attracting a top-tier investment bank to assist in conducting the IPO can be difficult, especially for small companies.5 And even if they can do the IPO, firms face high fees. Fees to investment banks for IPO assistance are on average 4% of proceeds in Europe, 6% in the United Kingdom, and 7% in North America (Torstila, 2003). Sometimes, an IPO is simply not an option owing to unfavorable equity market conditions. This also makes IPOs more uncertain events than reverse mergers: a planned IPO may need to be cancelled if market conditions suddenly take a bad turn, whereas a reverse merger is almost certain to be completed once both parties have agreed to it. Reverse mergers also have drawbacks. First, the listed firm may have hidden liabilities, for example, in the form of pending lawsuits or unreliable reporting. Second, it might be hard to identify and find all shareholders in the listed company. As a result, it may not be possible to acquire all shares. Third, the absence of a high-profile investment bank also implies that the firm lacks the certification effect of being backed by a third party with a high reputation at stake (Beatty and Ritter, 1986). In fact, since reverse mergers have a bad name, going public by this back door can damage the firm’s reputation. Finally, while a reverse merger yields the firm a listing, it does not raise capital as in an IPO. To do so, firms will have to access the capital markets at a later stage. Motives for doing a reverse merger thus seem to depend on equity market conditions, capital needs, investment banking competition, and regulatory requirements. This should also have consequences for the type of firms that choose a reverse merger. If avoiding regulatory scrutiny is the main motive, reverse mergers should notably attract low-quality firms (consistent with the model by ArellanoOstoa and Brusco, 2002), while cold-equity markets and the top-tier investment banking oligopoly should especially encourage small firms to do a reverse merger. Furthermore, since reverse mergers do not raise capital, they are more attractive to firms that do not need additional capital immediately and to those that expect to have little difficulty in attracting capital at a later stage (in spite of possibly incurring negative reputation effects of doing the reverse merger in the first place). If private firms decide to go public through a reverse merger, the first step is to identify a suitable target company with a listing. This is either a company that has become inactive, or a so-called shell company—a listed company with insignificant assets and operations that was formed with the intention of reducing taxes or obtaining financing. The second step is to decide on the method of acquisition. The firm can buy a major block of existing shares; alternatively, it can be agreed that a substantial number of shares be newly issued and transferred to the acquirer; or the firms can swap shares. In the United Kingdom the reverse merger is conditional on the firms’ shareholders’ giving consent in a general meeting. When shareholder approval is given, trading in the securities
5
James Altucher, The delicate art of a good reverse merger. Financial Times USA Edition, January 18, 2005, p. 10.
Reverse mergers in the United Kingdom: listed targets and private acquirers
185
of the listed company will be cancelled. Under rule 18, the firm then has to make an admission document available to the public for at least one month. Furthermore, the firm has to submit admission documents and forms to the Exchange before the securities can be admitted again.
8.3
Hypotheses
In this section we formulate four hypotheses regarding the abnormal stock price reactions to reverse-merger announcements. There is widespread empirical evidence that target firm shareholders experience positive shareholder wealth effects in takeovers (Bruner, 2004). This arises from the control premiums that bidders pay (e.g., Jensen and Ruback, 1983) and the competition among bidders (Bradley et al., 1988). Gleason, Rosenthal, and Wiggins (2005) find that positive shareholder wealth effects are even stronger for reverse mergers in the United States. We expect this also to be the case for reverse mergers in the United Kingdom. H1: Reverse-takeover announcements result in positive announcement returns for the public target company’s shareholders However, abnormal returns are likely to differ across firms. For example, the method of payment may matter. In the Myers and Majluf (1984) model, stock issues signal overvaluation of the issuer, since issuing stock is cheapest when the firm is overvalued. The same problem occurs when the acquirer pays for the acquisition with its own shares. Stock payments should thus result in lower abnormal returns for bidder shareholders. Many authors (e.g., Andrade, Mitchell, and Stafford, 2001; Travlos, 1987) indeed document that bidder returns in ordinary takeovers are higher if the cash component is larger. Since stock payments turn target shareholders into bidder shareholders, they might also be negatively affected. H2: Stock payments result in lower announcement returns than mixed payments Target firm profitability is likely to be an important determinant of announcement returns surrounding the news of a reverse merger. Loss-making firms can have valuable tax shields, which may result in large valuation gaps between current shareholders and prospective buyers. The buying firm is then more likely to be willing to pay a large premium. Moreover, the shareholders in distressed firms are pleased to see that their uncertainty regarding the recuperation of their investment is reduced owing to the finalization of the takeover. Therefore, we expect loss-making firms to have higher announcement returns. H3: Prior losses by the public target are positively related to announcement returns As Jarrell and Poulsen (1989) show, the relative size of the target is an important determinant of bidder returns, since small investments hardly affect shareholder value. However, for target shareholders (for which we can measure returns) it is about the whole firm, and the impact of the size of their firm relative to the acquirer is less obvious. In fact, Servaes (1991) finds no relationship between the relative size of the target and target announcement returns. However, in the case of reverse mergers, target firms are typically vulnerable, and chances of success of the new combination might be perceived to be higher when the bidder is large in comparison to the target.
186
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
H4: Announcement returns increase in the relative size of the private bidder A large body of literature has documented that IPO firms perform worse than nonIPO firms using various measurement techniques (e.g., Ritter, 1991; Loughran and Ritter, 1995). Gleason, Rosenthal, and Wiggins (2005) examine the long-term operating performance of reverse mergers in the United States. They find that operating performance does not improve significantly after the reverse merger and that only 46% of the sample survive two years. They conclude that reverse mergers are risky and fail to generate long-term wealth for shareholders. However, they do not provide a comparison between the performance of reverse-merger firms and IPO firms. Arellano-Ostoa and Brusco (2002) build a 3-period model in which a firm can choose between a reverse merger and an IPO, while facing uncertainty about its investment prospects and financing needs. After the firm learns its probability of having a positive NPV project, it will choose between the reverse merger (with possibly a subsequent equity offering) and the IPO as possibilities for financing the project. In the final stage, all uncertainty regarding the project vanishes, and the firm can immediately do the project (in case it had chosen the IPO) or do it somewhat later (at lower value) by conducting a follow-on equity offering (if it had chosen the reverse merger). Thus, the higher the probability of a positive NPV project being undertaken, the higher the value of doing the IPO and the lower the value of the flexibility offered by the reverse merger. The model yields a separating equilibrium in which high-quality firms prefer to do an IPO, while low-quality firms choose to do the reverse merger. Choosing an IPO is thus a signal of quality. Therefore, we expect reverse-merger firms to perform worse than IPO firms, both in terms of both stock returns and operating results. H5: Reverse-merger firms have worse long-run stock performance than IPO firms H6: Reverse merger firms experience worse long-run operating performance than IPO firms
8.4 8.4.1
Data and methodology Sample description
We identify reverse mergers by searching the Financial Times. Unfortunately, the Financial Times uses the very broad definition of a reverse merger applied by the United Kingdom Listing Authority. According to AIM listing rule 13, a reverse takeover is an acquisition in which the listed company fundamentally changes its business, board, or voting control. In this definition, a reverse is not necessarily an acquisition of a listed company by a private company. We therefore check Datastream to establish whether the target firm actually had a listing on the London Stock Exchange (LSE) prior to the event and whether the acquirer was previously unlisted. Moreover, we consult the “AIM Statistics: cancellations of admission” provided by the LSE, to crosscheck whether the firms found have cancelled or suspended their shares. This yields an initial sample of 233 completed reverse takeovers during the period January 1990 to December 2001. For these deals we collect prospectuses from Thomson Research and deal information from Thomson Financial Securities Data Corporation (SDC). The resulting final
Reverse mergers in the United Kingdom: listed targets and private acquirers
187
sample consists of 80 U.K. reverse takeovers for which prospectuses and deal information could be obtained. More than 80% of these are from the second part of the sample period 1996–2001. There are 47 deals that involve a public target listed on the Official List and 33 reverse mergers that involve a public target listed on the AIM. Among the acquiring private firms, the service industry dominates, while the industry distribution is more equal among the public targets. Table 8.1 shows the descriptive statistics of our sample. The two merging companies are of similar size. The average [median] total assets in the year preceding the reverse merger equals £19.3 million [£4.1 million] for the public company and £18.1 million [£6.6 million] for the private company. The public company experiences poor performance in the year before the reverse merger transaction. The average [median] return on assets equals ⫺36.3% [⫺2.4%] and the average [median] return on equity equals ⫺58.4% [⫺0.1%]. This shows that most public targets are loss-reporting firms. These loss-reporting firms may offer valuable tax shields for profitable private acquirers. The average [median] public company holds 35.2% [16.4%] of its assets in cash. This reflects that many of the public companies are cash-rich companies that have either recently sold their assets or have raised money in an IPO but have failed to find an interesting investment opportunity for the money.
8.4.2
Event study methodology
A standard event study procedure is used to assess whether stock prices of the listed target change around the announcement of a reverse merger. The market model is used to calculate daily abnormal returns. The market model assumes
Table 8.1 Descriptive statistics Mean Total assets (£ million); public company Total assets (£ million); private company Return on assets (%); public company Return on equity (%); public company Cash to total assets (%); public company
Median
Maximum
Minimum
Std. dev.
19.31
4.11
370.90
0.01
51.09
18.04
6.60
282.07
0.02
38.74
⫺36.31
⫺2.44
45.75
⫺920.53
128.45
⫺58.37
0.12
541.45
⫺1379.35
263.45
35.23
16.42
100.00
0.00
36.45
Note: The descriptive statistics for 80 U.K. reverse mergers completed during January 1990 and December 2001. All data relate to the year prior to the reverse merger transaction. Return on assets expresses net income as a percentage of total assets. Return on equity expresses net income as a percentage of the book value of equity. Cash to total assets measures cash balances reported on the balance sheet as a percentage of total assets.
188
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
that the return on an individual stock is linearly related to the market return. The model can be expressed as Rj , t ⫽ αj ⫹ β j Rm, t ⫹ εj , t
(8.1)
where R j,t is firm j’s return on day t, Rm,t is the return on the market index, and αj and βj are firm j’s market model parameters. The market model is estimated over a 100 trading day pre-event period beginning 111 trading days before the event date and ending 11 days before the event date. Given that our sample firms are small relative to other publicly listed companies, we employ the FTSE small-cap index as the market index. The abnormal return for firm j over day t (ARj,t) is computed as: ARj , t ⫽ Rj , t ⫺ (aj ⫹ bj Rm, t )
(8.2)
where aj and bj are the ordinary least-squares estimates of firm j’s market model parameters αj and βj. Abnormal returns are computed during the event period from 10 trading days before the event date to the event date itself. We cannot include post-announcement days in our event window because the LSE typically suspends the shares of the listed target company the day after or on the day of the announcement of the reverse merger. Abnormal returns are then averaged across firms to generate the average abnormal return (AARt). Cumulative average abnormal returns (CAARt ,t ) are calculated by summing the average abnormal returns over the event window [t1, t2] around the event date, which is labeled day 0. Parametric tests of statistical significance are based on standardized abnormal returns (see Brown and Warner, 1985). Standardized abnormal returns (SARjt) for firm j on day t are determined by dividing the abnormal returns (ARjt) by the standard deviation of firm j’s abnormal returns [σ(ARj)] during the 100 trading day estimation period. Standardized abnormal returns are then averaged across firms to form the average standardized abnormal return (ASARt). Cumulative average standardized abnormal returns (CASARt ,t ) are calculated by summing the average standardized abnormal returns for the event window [t1, t2]. The following t-statistic is calculated: 1
2
1
t⫽
CASARt
1 ,t 2
σ(ASAR)* T
2
(8.3)
where (ASAR) denotes the standard deviation of average standardized abnormal returns during the estimation period of 100 trading days and T is the number of days in the event window.
Reverse mergers in the United Kingdom: listed targets and private acquirers
8.4.3
189
Cross-sectional regression model
We estimate a cross-sectional regression model using OLS to test hypotheses H1 to H4. We use the cumulative abnormal return from day ⫺1 to day 0 (CARj,⫺1,0) as our dependent variable. This cumulative abnormal return is computed by adding the abnormal return for firm j over these two event days. The regression model reads as follows: CARj ,⫺1,0 ⫽ β0 ⫹ β1PAYMENTj ⫹ β2ROEj ⫹ β3RSIZEj ⫹ εj
(8.4)
where PAYMENT is a dummy variable that takes on the value 1 if the deal is paid for in stock, ROE is the return on equity of the listed target company in the year before the reverse merger, and RSIZE is the ratio of the total assets of the private acquirer to the total assets of the public target company in the year before the reverse merger.
8.4.4
Measuring long-term performance
In order to test hypotheses H5 and H6, we measure long-term performance, and we match each reverse-merger firm to a comparable IPO firm. The matching procedure works as follows. We identify IPO firms that went public in the period from one year before up to one year after the year in which the reverse merger was completed. Hence, a reverse merger that became effective in March 1998 is matched to IPO firms that went public between March 1997 and March 1999. Next, we require the IPO firm to be comparable in size to the reverse-merger firm. Size is measured by the market value on the first trading day. The matching IPO firm is not allowed to deviate more than 25% from the size of the reverse merger firm. As a final step we match the IPO firm to the reverse-merger firm based on industry. We start matching on a four-digit SIC code. If this does not result in a matching IPO firm, we switch to a lower level digit. We succeed to industry match on a three- or four-digit industry level in 15% of the cases. In 20% [65%] of the cases, the IPO firm is in the same two[one-] digit level industry as the reverse merger firm. We compute buy-and-hold returns over a one-, three-, and five-year period after admission to the LSE for both the reverse merger firm and the matching IPO firm. If the firm is delisted before the end of the one-, three-, and five-year holding period, we compute buy-and-hold returns until the delisting date. We then compare the mean and median stock price performance of reverse merger firms and matching IPO firms. We conduct a simple t-test for difference in means and a Wilcoxon signed-rank test for difference in medians. In addition, we investigate the operating performance for the reverse mergers and matching IPO firms. Operating performance is measured as earnings before interest, taxes, and depreciation and amortization (EBITDA) expressed as a percentage of total assets. We measure operating performance on an
190
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
annual basis for up to five years after the reverse merger or IPO or until the year of delisting, whichever is earlier.
8.5
Empirical results
We conduct an event study to test our first hypothesis. Table 8.2 shows the event study results. We find that the abnormal returns surrounding the announcement of the reverse merger are significantly positive. During the event window from day ⫺10 to day 0 (the day that the reverse merger is announced), the cumulative abnormal return for the shareholders of the listed target company equals 4.1%. The cumulative abnormal return from day ⫺1 to 0 is equal to 3.6%. This suggests that the reverse merger is value-creating for the shareholders of the listed target company. We attempt to explain the cumulative abnormal returns from day ⫺1 to day 0 by estimating regression equation (8.4). Table 8.3 shows the results. Model (1) regresses the announcement return on a dummy variable that takes on the value 1 if the deal is financed entirely with stock (PAYMENT). None of the deals is paid fully in cash. There are 23 deals that are financed by a combination of cash and stock, while the remaining 57 deals are financed using stock only. We do not find a significant relationship between the method of payment and the announcement return. This finding is at odds with the hypothesized negative relation. Model (2) includes the return on equity (ROE) of the public target company as an additional explanatory variable. We find that the announcement return increases in the return on equity of the public company in the year prior to the Table 8.2 Abnormal returns around the announcement of a reverse merger Event day
AAR
t-Statistic
Event window
CAAR
t-Statistic
⫺10 ⫺9 ⫺8 ⫺7 ⫺6 ⫺5 ⫺4 ⫺3 ⫺2 ⫺1 0
⫺0.03% 0.52% ⫺0.32% ⫺0.28% 0.12% ⫺0.13% ⫺0.16% 0.37% 0.48% 1.01% 2.56%
⫺0.10 1.50 ⫺0.89 ⫺0.88 0.51 ⫺0.31 ⫺0.37 0.91 0.95 2.04b 2.23b
(⫺10,0) (⫺1,0)
4.14% 3.57%
2.15b 2.67a
Note: The event study results for different intervals surrounding the announcement of the reverse merger. t-Statistics are computed using equation (3). a,b,cSignificance at the 1%, 5%, and 10% level, respectively (two-tailed test).
Reverse mergers in the United Kingdom: listed targets and private acquirers
191
Table 8.3 Cross-sectional regressions of announcement returns on various explanatory variables
PAYMENT ROE RSIZE Intercept 2
R adjusted F-test
Model (1)
Model (2)
Model (3)
⫺0.008 (⫺0.275)
⫺0.005 (⫺0.172) ⫺0.014 (⫺2.800)a
0.034 (2.000)b
0.027 (1.588)
0.013 ⫺0.002 0.002 0.034
0.001 0.078
0.087 3.443b
(0.464) (⫺0.286) (2.000)b (2.125)b
0.201 5.799a
Note: The cross-sectional regression results using the cumulative abnormal return (CAR) from day ⫺1 to day 0 as our dependent variable. PAYMENT is a dummy variable that takes on the value 1 if the deal is paid for in stock. ROE is the return on equity of the listed target company in the year before the reverse merger. RSIZE is the ratio of the total assets of the private acquirer to the total assets of the public target company in the year before the reverse merger. a,b,c Significance at the 1%, 5%, and 10% level, respectively (two-tailed test).
reverse merger. This is consistent with hypothesis H3. We argue that lower performance indicates the presence of valuable tax shields for the private acquirer. The shareholders of the target company seem to capture part of these valuable tax shields in the premium that they receive in the reverse-merger transaction. Model (3) also includes the relative size of the private acquirer to the public company (RSIZE). We find a positive relation between relative size and the announcement return. If the private acquirer is larger than the public target, it indicates that the combined firm may be more successful. Shareholders of the public company capture part of that gain in the form of a higher premium. When we include the relative size to the regression, the significant inverse relation between the return on equity and the announcement return disappear. Panel A of Table 8.4 shows long-term buy-and-hold returns both for the reverse-merger firms and for the matched IPO firms. Returns are measured until the end of the holding period of one, three, or five years or until the time of delisting, whichever is earlier. However, at the time of this study (June 2004), the deals conducted in 2000/2001 did not yet have data for the full five years. This explains why the number of observations reduces as the holding period increases. We find that the median stock price performance of reverse-merger firms is lower than the performance of matched IPO firms during each of the three holding periods we look at. However, these differences are not statistically significant, neither for averages nor for medians. Panel B of Table 8.4 shows the operating performance up to five years after the reverse merger or IPO. The number of observation declines over time because of limited data availability and mortality. There are 24 reverse-merger firms and 18 matched IPO firms that have been taken over or that have liquidated within five years after the transaction or before June 2004, whichever is earlier. We document that reverse-merger firms and matched IPO firms display similar average (median) return on assets. No evidence suggests
192
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 8.4 Long-run stock price and operating performance after reverse mergers Panel A: Average (median) buy-and-hold returns Holding N period (in months) 12 36 60
Reverse mergers
N
Matched IPO firm
t-Test for difference (Wilcoxon z-test for difference)
80 28.52% (⫺10.65%) 80 29.05% (4.97%) 0.021 (1.227) 77 25.32% (⫺7.41%) 77 15.61% (⫺6.85%) 0.561 (0.625) 66 23.22% (⫺21.72%) 64 28.86% (⫺13.99%) 0.267 (1.050) Panel B: Average (median) return on assets
Year
N
Reverse mergers
N
Matched IPO firm
t-Test for difference (Wilcoxon z-test for difference)
1 2 3 4 5
67 66 62 40 25
4.27% 1.11% 3.37% 5.62% 4.57%
77 73 70 50 36
0.04% 6.27% 3.92% 9.31% 5.70%
0.393 0.582 0.075 0.509 0.105
(2.97%) (5.33%) (4.95%) (5.61%) (9.29%)
(5.82%) (7.48%) (7.71%) (7.00%) (10.58%)
(1.249) (1.617) (1.662)c (1.360) (0.213)
Note: The second column of Panel A shows the number of reverse merger firms and the third column shows the average (median) buy-and-hold returns. The number of matched IPO firms is shown in the fourth column, while the fifth column shows their average (median) buy-and-hold returns. The last column shows the test statistics for differences in mean (median) between the buy-and-hold returns of the reverse merger firms and matched IPO firms. Panel B has a structure similar to that of Panel A but relates to the return on assets. Return on assets is measured as the earnings before interest, taxes, depreciation and amortization (EBITDA) divided by total assets. a,b,cSignificance at the 1%, 5%, and 10% level respectively (two-tailed test).
that reverse-merger firms perform worse than IPO firms in the long run. This is inconsistent with our hypotheses H5 and H6.
8.6
Conclusion
We have investigated reverse mergers in the United Kingdom during the years 1990–2001. Reverse mergers are an alternative way of taking a company public. In reverse mergers a private company de facto acquires a public target company whereas, de iure, it is the public company that acquires the private company. This way the private company, the acquirer in economic terms, can obtain a stock market listing for its shares via the backdoor without the costs and time it takes to conduct an IPO. We find that reverse mergers are value-creating events for the shareholders of the public company, i.e., the target in economic terms. The reverse merger exhibits
Reverse mergers in the United Kingdom: listed targets and private acquirers
193
a positive stock price reaction of 3.6%. In addition, we find that the announcement return is higher if the target company experiences poor performance in the year prior to the reverse merger and if the private company is large relative to the public company. That is, abnormal returns are higher when the target firm is more in need of a takeover and when the takeover is more likely to succeed. We document that both the long-run stock returns and the operating performance of reverse-merger firms are similar to those of their matched IPO counterparts. Thus, reverse-merger firms that go public via the backdoor do not seem to be inferior to IPO firms that go public via the front door. This finding is inconsistent with the predictions of the model of Arellano-Ostoa and Brusco (2002) and challenges conventional wisdom that regards reverse-merger firms as poorquality firms trying to avoid regulatory scrutiny at the time of going public. The extensive academic IPO literature may lead one to think that IPOs are the only way of going public. However, reverse mergers provide a feasible alternative of going public and warrant more attention from academic researchers.
References Andrade, G., Mitchell, M., and Stafford, E. M. (2001). New Evidence and Perspectives on Mergers. Journal of Economic Perspectives, 15(2):103–120. Arellano-Ostoa, A., and Brusco, S. (2002). Understanding Reverse Mergers: A First Approach. Working Paper. Universidad Carlos III De Madrid. Beatty, R. P., and Ritter, J. R. (1986). Investment Banking, Reputation, and the Underpricing of Initial Public Offerings. Journal of Financial Economics, 15:213–232. Bradley, M., Desai, A., and Kim, E. H. (1988). Synergistic Gains from Corporate Acquisitions and Their Division between the Stockholders of Target and Acquiring Firms. Journal of Financial Economics, 21:3–40. Brown, S., and Warner, J. (1985). Using Daily Stock Returns: The Case of Event Studies. Journal of Financial Economics, 14:3–31. Bruner, R. F. (2004). Does M&A Pay? In Applied Mergers and Acquisitions (Bruner, R. F., ed.). John Wiley & Sons, pp. 1056–1097. Gleason, K. C., Rosenthal, L., and Wiggins, R. A. (2005). Backing into Being Public: An Exploratory Analysis of Reverse Takeovers. Journal of Corporate Finance, 12(1):54–79. Jarrell, G. A., and Poulsen, A. B. (1989). The Returns to Acquiring Firms in Tender Offers: Evidence from Three Decades. Financial Management, 18(3):12–19. Jensen, M. C., and Ruback, R. (1983). The Market for Corporate Control: The Scientific Evidence. Journal of Financial Economics, 11(1–4):5–50. Loughran, T., and Ritter, J. R. (1995). The New Issues Puzzle. Journal of Finance, 50(1):23–51.
194
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Myers, S. C., and Majluf, N. S. (1984). Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have. Journal of Financial Economics, 13:87–221. Ritter, J. R. (1991). The Long-Run Performance of Initial Public Offerings. Journal of Finance, 46(1):3–27. Servaes, H. (1991). Tobin’s q and the Gains from Takeovers. Journal of Finance, 46(1):409–419. Torstila, S. (2003). The Clustering of IPO Gross Spreads: International Evidence. Journal of Financial and Quantitative Analysis, 38(3):673–694. Travlos, N. G. (1987). Takeover Bids, Methods of Payment, and Bidding Firms’ Stock Returns. Journal of Finance, 42(4):943–963. White, H. L. (1980). A Heteroscedastic-Consistent Covariance Matrix Estimator and a Direct Test of Heteroscedasticity. Econometrica, 48:817–838.
9 The profile of venture capital exits in Canada1
Douglas Cumming and Sofia Johan
Abstract This chapter introduces a new dataset that depicts a comprehensive profile of 518 venture capital exits in Canada for the years 1991–2004. The complete class of exits is considered: initial public offerings (IPOs, or new listings on a stock exchange for sale to the general public), acquisitions (in which the investor and entrepreneur sell to a larger company), secondary sales (in which the investor sells to another company or another investor, but the entrepreneur does not sell), buybacks (in which the entrepreneur repurchases the interest of the investor), and write-offs (liquidations). The data show patterns of exit vary depending on the exit year, the characteristics of the venture capital investor (private limited partnership, corporate, and government), the characteristics of the investee firm (industry and stage of development at first investment), and the characteristics of the transaction (capital requirements, syndication and security design).
9.1
Introduction
Venture capitalists (VCs) typically invest in early stage high-tech entrepreneurial firms for the purpose of achieving capital gains upon an exit transaction. Early-stage high-tech firms generally do not have the positive cash flow to pay interest on debt and/or dividends on equity investments; hence, venture capital (VC) returns are derived from capital gains upon exit transactions. VCs typically invest for 2–7 years prior to exit events. VCs are widely regarded as having the expertise to carry out the due diligence upon screening potential investments, and adding value to their investee firms through sitting on boards of directors and providing financial, strategic, marketing, and managerial advice, as well as facilitating a network of contacts for investee firms with suppliers, accountants, lawyers, and investment banks (Sahlman, 1990). There are five primary types of VC exits: initial public offerings (IPOs, or new listings on a stock exchange for sale to the general public), acquisitions (in which the investor and entrepreneur sell to a larger company), secondary sales (in which 1
Venture capital data for this chapter were generously provided by Macdonald and Associates, Limited (Toronto).
196
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
the investor sells to another company or another investor, but the entrepreneur does not sell), buybacks (in which the entrepreneur repurchases the interest of the investor), and write-offs (liquidations). Prior work is consistent with the view that the best investee firms are exited by means of IPOs or acquisitions, while secondary sales and buybacks are less desirable forms of exit (Cumming and MacIntosh, 2001; 2003a,b). The intuition and related research is discussed in Section 2. This chapter builds on the literature on VC exits by providing a description of the exit profile of Canadian VCs. A comprehensive dataset for all exited VC-backed firms in Canada is introduced for the first time in this paper. The data comprise 518 VC exits over the period 1991–2004. The rich details in the data provide new insights into the ways VCs exit their investments, and how exit patterns depend on market conditions in the exit year, the characteristics of the VC investor (private limited partnership, corporate, and government), the characteristics of the investee firms (industry and stage of development at first investment), and the characteristics of the transaction (capital requirements, syndication, and security design). This chapter is organized as follows. Section 9.2 discusses factors that give rise to VC-backed IPOs versus acquisitions, secondary sales, buybacks, and writeoffs. Section 9.3 describes the Canadian VC market and prior academic research related to VC in Canada. Section 9.4 introduces the exit data and describes the profile of exited firms in Canada. Concluding remarks and policy implications are discussed in the last section.
9.2
Venture capital exits: theory and prior evidence
Research in the area of VC has well established the proposition that IPOs and acquisitions are the best exit outcomes for VC-backed firms (Cumming and MacIntosh, 2001; 2003a,b). Briefly stated, the main intuition is as follows. The more cash received by the VC and the VC-backed firm from an external investor, the better the exit. The best exit outcome, therefore, is one that provides the VC with the highest return on investment and the investee firm with the most capital. IPOs and acquisitions by external investors bring the most new capital to the investee firm and the highest returns to both the VC and the entrepreneur as they both divest their interests to the external investor. These exit transactions are also deemed to be the most preferred by new investors. New investors are willing to pay a premium for the “clean” break from the VC and entrepreneur and start afresh with promoting their own agenda for the firm. Secondary sales only involve the current VC’s divesting its interest to the external investor, with the entrepreneur retaining his interest in the firm. The extent to which the external investor will pay a premium for the investment, and thus reduce the amount to be paid to the VC, is diminished as the purchase price will reflect the “baggage” that comes along with the entrepreneur still maintaining his interest in the firm. As secondary-sale exits are often indicative of a divergence of exit objectives
The profile of venture capital exits in Canada
197
between the VC and entrepreneur, the new investor will take into account foreseeable conflicts with the entrepreneur with regard to the running of the firm in the purchase price. Buybacks bring in no new capital to the firm; rather, the entrepreneur merely repurchases the shares from the VC. Hence, firms with the better growth potential are not likely to be exited by a buyback. Write-offs are, of course, the least desirable form of exit. Prior academic work has identified a number of potential factors that could affect VC-backed IPOs versus acquisitions. A detailed survey of these factors is provided by Cumming and MacIntosh (2003b) (see also Zingales, 1995; Yosha, 1995; Black and Gilson, 1998; and Gompers and Lerner, 1999). In brief, however, these factors include the following: (1) Market factors such as the cyclicality of valuations in IPO markets, the transaction costs of effecting a sale, the liquidity of exit to the seller, the liquidity of the investment to the buyer, and transaction synergies; (2) new owners’ perceptions, such as the new owners’ ability to value the firm, resolve information asymmetry, and align interests with that of existing firm managers (insiders); and (3) firm factors such as firm size, growth of the firm and the demand for funds, the ongoing costs of operating as a public versus a private firm, insider ownership, and managerial incentives post-exit by VC and founding entrepreneur, as well as the reporting requirements and public profile of a firm that goes public. These factors are briefly reviewed in this section. Market and regulatory factors that could affect VC-backed IPOs versus acquisitions are factors not within the control of either VC or entrepreneur. IPO markets are not only subject to swings (Barry, Muscarella, and Peavy III, and Vetsuypens, 1990; Megginson and Weiss, 1991; Lerner, 1994; Brav and Gompers, 1997; Gompers and Lerner, 1999; for VC-backed IPOs; more generally on IPOs see also Ritter, 1984, and Brav and Gompers, 2003), but also suffer the related issue of the regulation of these markets as well as regulation of the firms listed on these markets. The data (described in the succeeding) are from Canada. Canada comprises 10 provinces and 3 northern territories. Each province’s regulatory system is based on the common law system, with the exception of Quebec. To offer securities for sale throughout Canada, however, regulatory standards of each of the provincial securities commissions must be met by firms; therefore, regulatory issues are not expected to generate regional differences in a VC’s or entrepreneur’s propensity to divest through an IPO. Regional differences may nevertheless exist owing to economic conditions in different provinces. These issues are considered in the analysis of the data. A number of different entrepreneurial firm characteristics can affect IPOs versus acquisitions; examples include the size of a firm’s assets (minimum listing requirements for IPOs include minimum asset size), and the primary industry in which it operates. Generally, these characteristics affect the ability of the new owners to assess the firm’s financial viability, value the firm’s technology, and monitor the firm. Firm characteristics determine not only the cost of IPOs versus acquisitions, but also the amount of proceeds from such divestment. Younger high-tech firms without a track record have more pronounced information problems; therefore, all else being equal, such firms face greater costs associated with going public
198
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
than with being acquired. Firms in high-tech industries have higher market/book ratios and greater growth options and therefore may be more likely to need to go public (Gompers and Lerner, 1999). Industry characteristics can also be related to transaction synergies, which could, however, favor acquisitions relative to IPOs for certain high-tech firms. (We noted previously that market conditions obviously also play a significant role.) All else being equal, the cost of an IPO is lower, and the proceeds higher, when a firm’s information asymmetries attributable to its own characteristics are less pronounced. In addition, because IPOs and acquisitions may generate different proceeds in the sale of the firm, a firm’s growth potential and capital needs may also be relevant for the choice between an IPO and acquisition. A VC wears many hats during his involvement in the investee firm. He is initially a financial intermediary between the limited partners of the fund and the investee firm. Upon obtaining interest in the investee firm, the VC will play the role of advisor and monitor to add value to the firm (including strategic, financial, and marketing advice; the VC will also facilitate a network of contacts with legal and accounting advisors, investment banks, suppliers; see Sahlman, 1990; Manigart, et al., 1996; Gompers and Lerner, 1999; Lockett and Wright, 2001). Finally, the VC will, by its participation in the firm, certify the quality of the entrepreneurial firm to public shareholders in the event of an IPO, and the acquiring firm in the event of an acquisition (Barry, Muscarella, and Peavy, III and Vetsuypens, 1990; Megginson and Weiss, 1991; Lerner, 1994; Gompers and Lerner, 1999), so that information asymmetries between the firm and the new owners are lower than they otherwise would have been. As a result of these roles played by the VC, proceeds from the divestment of the firm may be higher as external investors are more willing to pay the premium. The certification role of a VC differs from that of an IPO and acquisition because the new owners in an IPO may range from unsophisticated retail investors to sophisticated institutional investors, while the new owners in an acquisition will be more likely to be a corporation. We may conjecture (based on citations listed in notes 1–3) that a corporation acquiring the entrepreneurial firm is better able to value the firm, resolve information problems, and monitor the firm relative to a group of disparate, relatively unsophisticated shareholders in the case of an IPO. In an IPO, greater information asymmetry (all else being equal) is associated with more underpricing; therefore, the VC’s certifying role is more pertinent for IPOs than for acquisitions. Information asymmetries, however, do matter to a significant degree in sale transactions as they are inversely correlated with the sale price. All else being equal, the certification provided by the VC lowers the indirect costs of an IPO (underpricing) and increases the probability of an IPO relative to an acquisition (Barry Muscarella, and Peavy, III and Vetsuypens, 1990; Megginson and Weiss, 1991; Lerner, 1994; Gompers and Lerner, 1999). Both the value-added and certification roles of a VC, of course, depend on the characteristics of the VC; therefore, in our empirical analysis we consider the type of VC fund (private independent limited partnership, corporate versus government funds). Prior research is consistent with the proposition that VCs
The profile of venture capital exits in Canada
199
structured as limited partnerships provide greater value-added and quality certification relative to other types of VCs (Gompers and Lerner, 1999). Government funds in Canada, for reasons summarized in Section 9.3, are expected to provide a much less pronounced certification effect. Recent work has also examined the role of VC control rights in IPOs versus acquisitions. Financial contracts allocate the decision-making process for future events and contingencies to avoid any potential conflicts (Hart, 2001). In the context of VC contracts, these control rights include the conventional shareholder rights dealing with the entitlements that come with the ownership of the shares of the firm, such as redemption rights, anti-dilution rights, and drag-along rights, as well as more stringent and distinctive rights including, for example, the right to replace the CEO and other veto rights to facilitate the monitoring role played by the VC. Although these control rights are structured to manage any potential conflicts, not all conflicts are avoidable. For example, even when an acquisition is financially superior to an IPO, an entrepreneur may nevertheless prefer an IPO owing to the private benefits he may enjoy, say, being the CEO of a publicly listed firm (Black and Gilson, 1998). The existence of entrepreneurial private benefits therefore gives rise to a prediction that strong VC control should be associated with a greater probability of an acquisition (which is typically 3–4 years after the initial investment) since strong control rights enable VCs to minimize entrepreneurial private benefits. Cumming (2006b) finds evidence in support of this theory that strong VC control rights are associated with a greater probability of an acquisition than an IPO. In sum, the central theme in this section was to point out a potentially pivotal role for VCs in an entrepreneurial firm’s IPO or acquisition outcome. This section also discussed the importance of market and institutional conditions, as well as VC and entrepreneurial firm specific characteristics. Section 9.4 discusses evidence on VC exits in Canada in relation to this prior theoretical and empirical evidence. Section 9.3 first describes the VC setting in Canada to facilitate an interpretation of the evidence in relation to prior theoretical and empirical work.
9.3
Venture capital in Canada
To provide a perspective on VC in Canada, we begin by presenting aggregate Canadian VC industry statistics collected by Macdonald & Associates, Ltd. (Toronto) for the Canadian Venture Capital Association (CVCA) (for the years for which data are available up to 2004). The data are presented in Figures 9.1–9.3 and described in this section. Macdonald (1992) and Cumming and MacIntosh (2006, 2007) previously discussed some of the data. As explained in this section, the most notable characteristic of VC in Canada is the presence of tax-subsidized funds. As illustrated in Figure 9.1, there are six types of VC funds in Canada (see also Macdonald, 1992; MacIntosh, 1994; Amit, Brander, and Zott, 1998): private independent, corporate, government, institutional direct (formerly known
200
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Can $ (billions of 2004 dollars)
25 20 15 10 5 0 92
93
94
95
96
97
98
99
00
01
02
03
04
Year Corporate
Government
Labour sponsored
Institutional direct/Foreign
Private independent
Figure 9.1 Venture capital under management by investor type in Canada, 1992–2004.
as hybrid funds), foreign, and Labour-sponsored venture capital corporations (LSVCCs). Canadian private independent funds (VC funds, unless otherwise stipulated) are similar to their U.S. counterparts (described by Gompers and Lerner, 1999; Cumming and Johan, 2006b; Sahlman, 1990), and are organized as limited partnerships. They tend, however, toward a lower degree of specialization in investment activity than their U.S. counterparts (MacIntosh, 1994; 1998). Canadian corporate VCs are analogous to U.S. corporate VCs (as described by Gompers and Lerner, 1999), but tend to finance a somewhat more heterogeneous group of entrepreneurial firms (Cumming, 2005a,b; 2006a). Government funds are managed by independent professional VC managers and also finance a wide variety of different entrepreneurial firms. The hybrid category consists of public pension funds that invest money directly in entrepreneurial firms, rather than investing through the medium of a private fund. As of 2004, all institutional actors that make direct investments are reported as “institutional direct” funds, although this category largely corresponds with funds previously reported as hybrid funds (i.e., most are public pension funds). Foreign funds are mostly U.S.-based VC funds operating in Canada. Figure 9.1 indicates the amount of funds under administration at each fund type from 1992 to 2004. In the past decade, it is clear that most of the growth in the industry can be attributed to the LSVCCs. The capital under administration by VC funds—the LSVCC’s main competitor—remained largely static between 1992 and 1999. While Figure 9.3 indicates substantial increases in the capital administered by VC funds in 2000 and 2001, this increase is illusory. Each fund
The profile of venture capital exits in Canada
201
type in Figure 9.1 is a net of five factors; capital contributions, returns of capital and/or profits to investors, changes in capital under administration resulting from revaluations of fund portfolios, changes resulting from realizations on the sale of investments, and the composition of the respondents to the CVCA’s annual information-gathering survey. According to the data gathered for the CVCA, much of the apparent increase in private funding in the year 2000 is attributable to the VCs’ realized capital gains, as well as the fundraising from a large VC fund introduced to the database in that year. Of the total $3.4 billion increase in private capital reported in 2000, only $1.6 billion can be attributed to new commitments to VC funds. Cumming and MacIntosh (2006) argue that the LSVCCs have attracted the most capital owing to generous tax incentives provided to these funds, not to the returns (see also Figure 9.3). This government-sponsored tax program for LSVCCs has crowded out private VC fund investment in Canada (Cumming and MacIntosh, 2006). Figure 9.2 presents data from the CVCA indicating capital under management, capital available for investment, and new funds for the 1988–2004 period. The capital available for investment reflects the extent to which contributions to these funds have outstripped the funds’ capacity to invest these contributions. On the basis of the CVCA data, it can be seen from Figure 9.2 that, historically, there has been a large “overhang” of uninvested capital in Canada.
Can $ (billions of 2004 dollars)
25,000
20,000
15,000
10,000
5000
0 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 Year 03 04 New capital raised Capital for investment Capital under management
Figure 9.2 Capital for investment in Canada, 1988–2004.
202
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Cumming and MacIntosh (2006) have argued that this overhang is attributable to the overly generous tax subsidies provided to LSVCCs. As discussed in Cumming and MacIntosh (2006, 2007) and shown in Figure 9.3, the increase in LSVCC capital is surprising in light of the low returns realized by these types of funds. It seems reasonably clear that the capital flows to LSVCCs are best explained by strong tax incentives that are not shared by investors in other types of funds. Figure 9.3 presents the performance of LSVCCs over the past 10 years.2 Figure 9.3 clearly indicates that an index of LSVCC performance has greatly lagged comparable indices.3 This is consistent with related evidence documenting inferior LSVCC performance relative to the U.S. VC investments (Cumming and MacIntosh, 2003a,b), and the inferior performance of LSVCC investments relative to other types of fund investments, especially other Canadian VC investments (Halpern, 1997). It is natural to expect that the tax-subsidized LSVCCs will affect the pattern of IPOs, acquisitions, secondary sales, buybacks, and write-offs among all types of funds in Canada, especially VC funds. The next section introduces a new dataset that provides for the first time direct insight into this issue.
700
Percentage return
600 500 400
The Peng (2001) data stops at 1999. The Venture Economics Post Venture Capital Index (PVCI) indicates an index value of 361.36 as at 06/28/2002 and a value of 557.77 as at 02/28/2005. The Peng (2001) Index is based on Venture Economics data, but there are some differences in the index computation methods.
300 200 100 0 −100
Mar 93
Sep 94
Mar 96
Sep 97
Mar 98
Sep 00
Mar 02
Sep 03
Dec 04
Date Globe LSIF Peer Index
Globe Canadian Small Cap Peer Index
TSE 300 Composite Index/TSX Total Return Index US VC index (Peng 2001, Figure 7)
30-Day Treasury Bill Index
Figure 9.3 Selected indexes, 1992–2005.
2 3
Canadian data sources for Figure 9.3: www.globefunds.com, www.morningstar.ca. The U.S. VC Index value from Peng (2001) is not available for 2000 and 2001. Peng’s data are from Venture Economics. Venture Economics has posted on their web page (www.ventureeconomics. com) a value of their own index for the date 06/28/2002 (only) of 361.36 that is based over a similar horizon used by Peng. The authors owe thanks to Peng for directing us to the Venture Economics citation for a recent comparable value for the U.S. index. It is noteworthy that Peng’s index calculations are more economically and statistically rigorous than that posted by Venture Economics.
The profile of venture capital exits in Canada
9.4
203
The data: exited Canadian venture capital investment, 1991–2004
In this section we introduce a dataset from 518 exited VC-backed firms in Canada over the 1991–2004 period. The data are first graphically presented in Figures 9.4–9.14. Statistical comparison tests are also presented in Table 9.1. The data are from Macdonald and Associates, Ltd., Toronto. Additional data on IPOs from the Toronto Stock Exchange and firm prospectuses were matched with the Macdonald and Associates data. We cannot be certain that the data are 100% comprehensive for the non-IPO exits (although Macdonald and Associates believe their data are nearly comprehensive), but we do believe the IPO data are comprehensive. Figure 9.4 depicts the percentage of exits by exit type. There are 32 IPOs, 197 acquisitions, 66 secondary sales, 116 buybacks, and 107 write-offs. In total 518 firms exited between 1991 and 2004. The number of VC-backed IPOs relative to other exit types is extremely low in Canada, and lower still in other countries. VCs achieved 34% of their exits as IPOs in Australasia in 1989–2001 (Cumming and Macintosh, 2006) and 17% of their exits as IPOs in Europe in 1995–2005 (Cumming, 2006b). A related study showed similar evidence that VCs achieved 25% of their exits as IPOs in the 1990–2001 period in Europe (Schwienbacher, 2002). VCs achieved 35% of their exits as IPOs in the U.S. in the 1991–2004 period (Cumming and Johan, 2006a). The poor performance of VCs in achieving IPO exits is most likely attributable to the dominant presence of LSVCCs in Canada.4 The profile of exits over time is depicted in Figure 9.5. Buybacks were the most frequently used type of exit in 2004, followed by acquisitions, and then an equal number of secondary sales and write-offs. There were no VC-backed IPOs in IPO 6%
Write-off 21%
Acquisition 38% Buyback 22% Sec sale 13%
Figure 9.4 Percentage of exits by exit type. 4
The low number of IPOs is not related to listing standards; IPO listing standards in Canada are much lower than those for NASDAQ and the NYSE. See Cumming and Johan (2006a).
204
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
60
Number of exits
50
40
30
20
10
0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Year IPO Buyback
Acquisition
Secondary sale
Write-off
Figure 9.5 Number of venture capital–backed exits by year, 1991–1994.
Canada in 2003 or 2004. VC-backed IPOs in Canada reached their peak in 2000, consistent with the peak of the Internet bubble. Figure 9.6 categorizes the percentage of exit outcomes for each of the different types of funds. Importantly, note that the data in Figure 9.6 are presented for the VC(s) that invested first-round (i.e., they served as the deal originators). Corporate VCs and VCs structured as private independent limited partnerships have had much greater success at picking firms that eventually achieved an IPO than have government VCs, hybrid VCs, and LSVCCs. Government VCs, hybrid VCs, and LSVCCs have had a much higher proportion of their exits as buybacks. LSVCCs have had the greatest proportion of both buybacks and secondary sales, which are inferior exit outcomes (see Section 2). This is consistent with prior work indicating poor governance associated with LSVCCs (Cumming and MacIntosh, 2006; 2007). Table 9.1 also shows that LSVCCs are statistically most likely to exit by secondary sales, followed by buybacks, write-offs, IPOs and acquisitions. Corporate VCs and VCs structured as private independent limited partnerships, by contrast, are statistically more likely to exit with IPOs and acquisitions than with secondary sales, buybacks, and write-offs. Figure 9.7 and Table 9.1 show the percentage of exit outcomes for firms in the life sciences industry (medical, environmental, and biotechnology), other types of high-tech (computers, electronics, Internet) and non–high-tech (such as manufacturing) industries. Life science firms were more likely to go public than other types
The profile of venture capital exits in Canada
205
60.00 50.00
(%)
40.00 30.00 20.00 10.00
Write-off Buyback Sec sale Acquisition IPO Unknown
Private independent
Labour sponsored
Hybrid
Government
Corporate
0.00
Figure 9.6 Percentage of exit outcomes for each type of investor.
60.00 50.00
(%)
40.00 30.00 20.00 10.00 Life sciences 0.00
Other high tech IPO
Acquisition
Secondary sale
Non high tech Buyback
Write-off
Figure 9.7 Percentage of exit outcomes for each industry.
206
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 9.1 The profile of venture Variable
Mean values for each type of exit IPOs
Acquisitions Secondary Buyback sales
Write-off
IPO vs IPO vs Acquisition Secondary sale
2.000
1.893
1.515
1.397
1.720
0.308
1.379
Amount invested $5,519 (thousands $Can 2004)
$2,256
$1,730
$1,640
$1,443
1.810*
2.061**
Deal size all $8,400 syndicated investors (thousands $Can 2004)
$4,275
$3,227
$2,418
$2,996
1.797*
1.961**
Life science firm dummy variable
0.406
0.157
0.076
0.078
0.131
3.315***
3.962***
Other high tech dummy variable
0.406
0.721
0.288
0.310
0.523
⫺3.529*** 1.172
Seed dummy variable
0.219
0.426
0.273
0.353
0.449
⫺2.226**
⫺0.575
Early stage dummy variable
0.125
0.218
0.106
0.129
0.168
⫺1.212
0.279
Expansion dummy variable
0.438
0.294
0.242
0.293
0.308
1.617
1.965**
Buyout dummy variable
0.031
0.030
0.076
0.078
0.037
0.024
⫺0.862
Turnaround dummy variable
0.063
0.005
0.242
0.112
0.037
2.650***
⫺2.157**
Other/unknown stage dummy variable
0.125
0.025
0.061
0.034
0.000
2.690***
1.092
Common Equity or Warrants
0.688
0.393
0.465
0.289
0.343
3.114***
2.075**
Number of syndicated investors
Preferred equity
0.000
0.077
0.043
0.051
0.090
⫺1.624
⫺1.189
Convertible preferred equity
0.063
0.099
0.015
0.072
0.121
⫺0.652
1.276
Debt
0.000
0.099
0.210
0.182
0.117
⫺1.860*
⫺2.795***
Convertible debt
0.031
0.114
0.086
0.080
0.092
⫺1.432
⫺1.007
Common and preferred and/or debt
0.094
0.045
0.124
0.163
0.110
1.145
⫺0.438
Other/unknown securities
0.125
0.172
0.058
0.163
0.127
⫺0.661
1.146
The profile of venture capital exits in Canada
207
capital exits in Canada 1991–2004 Comparisons of means and proportions test statistics IPO vs Buyback
IPO vs Write-off
Acquisition Acquisition vs vs Secondary Buyback sale
Acquisition vs Write-off
Secondary sale vs Buyback
Secondary sale vs Write-off
Buyback vs Write-off
1.743*
0.797
2.591***
3.746***
1.186
0.806
⫺1.280
⫺2.187**
2.178**
2.277**
0.894
1.643
1.960*
0.174
0.524
0.638
2.640***
2.318**
0.709
3.072***
1.569
0.565
0.150
⫺ 0.794
4.627***
3.455***
1.669*
2.042**
0.622
⫺0.045
⫺1.126
⫺1.306
1.021
⫺1.163
6.247***
7.082***
3.450***
⫺0.317
⫺3.036*** ⫺3.228***
⫺1.441
⫺2.333** 2.217**
1.273
⫺0.373
⫺1.119
⫺2.313**
⫺1.450
⫺0.065
⫺0.588
2.011**
1.956**
1.041
⫺0.463
⫺1.130
⫺0.818
1.545
1.354
0.813
0.025
⫺0.254
⫺ 0.736
⫺0.936
⫺0.249
⫺0.925
⫺0.163
⫺1.591
⫺1.885*
⫺0.323
⫺ 0.045
1.104
1.280
⫺0.823
0.613
⫺6.787*** ⫺4.423***
⫺2.115**
2.310**
4.097***
2.100**
2.005**
3.711***
⫺1.362
⫺0.465
1.662*
0.827
2.576**
1.938*
4.127***
3.465***
⫺1.020
1.870*
0.870
2.390**
1.598
⫺0.869
⫺1.301
⫺1.759*
0.948
0.892
⫺0.396
⫺0.239
⫺1.162
⫺1.148
⫺0.192
⫺0.938
2.194**
0.793
⫺0.602
⫺1.676*
⫺2.484**
⫺1.236
⫺2.609*** ⫺2.033** ⫺2.340**
⫺2.120**
⫺0.503
0.450
1.637
1.350
⫺0.964
⫺1.127
0.641
0.956
0.590
0.132
⫺0.142
⫺0.318
⫺0.977
⫺0.261
⫺2.239**
⫺3.537***
⫺2.140**
⫺0.717
0.276
1.150
⫺0.523
⫺ 0.026
2.283**
0.208
1.035
⫺2.054**
⫺1.458
0.762
(continued)
208
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 9.1 Variable
Mean values for each type of exit IPOs
Acquisitions Secondary Buyback sales
Write-off
IPO vs IPO vs Acquisition Secondary sale
Corporate VC
0.203
0.134
0.005
0.054
0.125
1.026
3.610***
Government VC
0.016
0.078
0.101
0.102
0.113
⫺1.292
⫺1.523
Hybrid VC
0.000
0.055
0.008
0.048
0.055
⫺1.366
⫺0.494
LSVCC
0.305
0.182
0.720
0.598
0.358
1.611
⫺3.906***
Private Independent limited Partnership VC
0.476
0.392
0.106
0.159
0.208
0.900
4.103***
Other/ unknown VC
0.000
0.157
0.061
0.040
0.142
⫺2.414**
⫺1.422
British Columbia 0.125 investee firm
0.122
0.061
0.026
0.140
0.051
1.092
Alberta investee firm
0.031
0.051
0.000
0.026
0.009
⫺ 0.479
1.444
Saskatchewan investee firm
0.000
0.000
0.000
0.000
0.000
N/a
N/a
Manitoba investee firm
0.031
0.015
0.076
0.026
0.000
0.642
⫺0.862
Ontario investee firm
0.344
0.492
0.076
0.190
0.262
⫺1.562
3.366***
Quebec investee firm
0.438
0.289
0.788
0.724
0.589
1.681*
⫺3.469***
Newfoundland investee firm
0.000
0.000
0.000
0.000
0.000
N/a
N/a
New Brunswick investee firm
0.031
0.000
0.000
0.000
0.000
2.487**
1.444
Nova Scotia investee firm
0.000
0.030
0.000
0.009
0.000
⫺1.000
N/a
Prince Edward Island investee firm
0.000
0.000
0.000
0.000
0.000
N/a
N/a
Northern Territories investee firm
0.000
0.000
0.000
0.000
0.000
N/a
N/a
British Columbia 0.125 investor
0.217
0.058
0.065
0.197
⫺1.203
1.146
Alberta investor
0.016
0.055
0.025
0.040
0.011
⫺0.945
⫺0.304
Saskatchewan investor
0.000
0.054
0.013
0.046
0.126
⫺1.341
⫺0.638
The profile of venture capital exits in Canada
209
(continued) Comparisons of means and proportions test statistics IPO vs Buyback
IPO vs Write-off
Acquisition Acquisition vs vs Secondary Buyback sale
Acquisition vs Write-off
Secondary sale vs Buyback
Secondary sale vs Write-off
Buyback vs Write-off
2.674***
1.115
3.006***
2.261**
0.244
⫺1.689*
⫺2.831*** ⫺1.874*
⫺1.563
⫺1.686*
⫺0.578
⫺0.715
⫺1.014
⫺0.017
⫺0.250
⫺0.274
⫺1.257
⫺1.349
1.652*
0.304
0.036
⫺1.451
⫺1.597
⫺0.236
⫺2.941*** ⫺0.551
⫺8.119*** ⫺7.517***
⫺3.401***
1.647*
4.625***
3.586***
3.793***
3.004***
4.310***
4.328***
3.273***
⫺0.986
⫺1.733*
⫺0.946
⫺1.155
⫺2.261*** 2.004***
3.146***
0.351
0.615
⫺1.661*
⫺2.666***
2.339**
⫺0.219
1.396
2.921***
⫺0.457
1.172
⫺1.626
⫺3.131***
0.166
0.913
1.866*
1.066
1.847*
⫺1.317
⫺0.788
0.928
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
0.166
1.835*
⫺2.478**
⫺0.663
1.283
1.579
2.889***
1.675*
1.854*
0.907
6.012***
5.329***
3.904***
⫺2.078**
⫺3.023*** ⫺1.288
⫺3.035*** ⫺1.511
⫺7.116*** ⫺7.467***
⫺5.101***
0.951
2.695***
2.130**
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
1.910*
1.835
N/a
N/a
N/a
N/a
N/a
N/a
⫺0.527
N/a
1.434
1.262
1.823*
⫺0.756
N/a
0.963
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
1.117
⫺0.926
2.933***
3.541***
0.424
⫺0.190
⫺2.524**
⫺2.935***
⫺0.660
0.199
0.971
0.588
1.857*
⫺0.513
0.703
1.333
⫺1.235
⫺2.115** 1.416
0.298
⫺2.237**
⫺1.196
⫺2.632*** ⫺2.152**
(continued)
210
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 9.1 Variable
Mean values for each type of exit IPOs
Acquisitions Secondary Buyback sales
Write-off
IPO vs IPO vs Acquisition Secondary sale
Manitoba investor
0.042
0.045
0.096
0.048
0.017
⫺0.093
⫺0.941
Ontario investor
0.365
0.591
0.144
0.218
0.322
⫺2.387**
2.490**
Quebec investor
0.422
0.444
0.773
0.736
0.623
⫺0.232
⫺3.434***
Newfoundland investor
0.000
0.000
0.000
0.000
0.000
N/a
N/a
New Brunswick investor
0.000
0.000
0.000
0.000
0.000
N/a
N/a
Nova Scotia investor
0.000
0.089
0.038
0.051
0.133
⫺1.758*
⫺1.115
Prince Edward Island investor
0.000
0.000
0.000
0.000
0.000
N/a
N/a
Foreign investor
0.031
0.166
0.033
0.012
0.059
⫺1.991**
⫺0.041
Investee firm and investor in same province
0.969
0.853
0.970
0.966
0.916
1.803*
⫺0.026
Year of First investment
1996.844 1997.914
1997.258
1997.750
1998.122 ⫺2.438**
Year of exit
1999.313 2001.629
2001.621
2002.224
2001.290 ⫺7.351*** ⫺5.782***
Duration in years from investment to exit
2.469
4.364
4.474
3.168
3.716
⫺0.787
⫺3.363*** ⫺4.267***
This table summarizes the characteristics of exited VC investments in Canada from 1991 to 1994. in total, 518 firms exited. The data are summarized for the characteristics of the investee and the size, stage of development, security choice) are provided for the first round investment only and not dummy variables) are provided for each exit outcome versus the others. N/a not applicable. *, **, *** Statistically significant at the 10%, 5% and 1% levels, respectively.
The profile of venture capital exits in Canada
211
(continued) Comparisons of means and proportions test statistics IPO vs Buyback
IPO vs Write-off
Acquisition Acquisition vs vs Secondary Buyback sale
Acquisition vs Write-off
Secondary sale vs Buyback
Secondary sale vs Write-off
Buyback vs Write-off
⫺ 0.159
0.835
⫺1.527
⫺ 0.124
1.291
1.249
2.396**
1.318
1.701*
0.445
6.287***
6.401***
4.466***
⫺1.217
⫺2.619*** ⫺1.764*
⫺3.345*** ⫺2.023** ⫺4.632*** ⫺5.026***
⫺2.988***
0.545
2.048**
1.812*
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
N/a
⫺1.306
⫺2.182** 1.358
1.231
⫺1.206
⫺0.410
⫺2.064**
⫺2.139**
N/a
N/a
N/a
N/a
N/a
N/a
N/a
0.755
⫺0.612
2.755***
4.211***
2.664***
0.966
⫺0.768
⫺1.903*
0.090
1.015
⫺2.549**
⫺3.136***
⫺1.589
0.152
1.409
1.580
⫺1.831*
⫺2.769*** 1.696*
0.476
⫺0.707
⫺1.095
⫺2.095**
⫺0.997
⫺8.264*** ⫺5.893*** 0.026
⫺2.272**
1.424
⫺1.681*
0.969
3.268***
⫺4.759*** ⫺1.824*
⫺2.430**
2.118**
⫺0.279
3.360***
3.994***
N/a
⫺1.894*
There were 32 IPOs, 197 acquisitions, 66 secondary sales, 116 buybacks, and 107 write-offs; and, investors. Transaction characteristics (number of syndicated investors, amounts invested, deal subsequent financing rounds. Comparisons of means tests (comparisons of proportions tests for
212
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
of high-tech and non–high-tech firms. Non–high-tech firms are more likely to be exited as buybacks, which is expected, as such firms typically have positive cash flows that may enable the entrepreneur to repurchase the interests of the VCs. The percentage of exits by the stage of first investment is depicted in Figure 9.8 and Table 9.1. Write-offs are most likely to have been seed-stage investments at first investment, and least likely to have been buyouts and turnaround investments. IPOs are more likely to have been involved in later stage investments such as buyouts. Secondary sales and buybacks were most often turnaround investments. Acquisitions were more often early-stage and seed investments than later stage investments. The percentage of exit outcomes by type of security used at the initial investment round is depicted in Figure 9.9. For most IPOs, the security used at the initial investment is unknown; however, for those that are known, common equity was used for investments that went public. Securities for which the investor has downside protection are more likely to be exited as acquisitions than IPOs. These findings are consistent with prior evidence (Cumming, 2006b) that strong VC control rights are associated with acquisitions and weak VC control rights are associated with IPOs. Table 9.1 is consistent in that it shows that IPOs are also less likely than acquisitions when the security used has downside protection for
50.00 45.00 40.00 35.00
(%)
30.00 25.00 20.00 15.00 10.00 Acquisition Buyback Secondary sale IPO
5.00 0.00 Seed
Early Expansion
Buyout
Write-off Turnaround Unknown
Figure 9.8 Percentage of exit outcomes for each stage of first investment.
The profile of venture capital exits in Canada
213
50.00 45.00 40.00 35.00
(%)
30.00 25.00 20.00 15.00 10.00 Acquisition Write-off Buyback Secondary sale IPO
5.00
Unknown
Debt
Mixed common with preferred and/or debt
Convertible debt
Convertible preferred
Preferred
Common and/or warrants
0.00
Figure 9.9 Percentage of exit outcomes for each type of security.
the investor (although these differences between IPOs and acquisitions are significant and at the 10% level for debt only). Moreover, the data indicate buyback exits are more likely to have involved some debt, which is also expected, as debt does not afford ownership to the VC and the entrepreneurial firm repays the principal upon expiration of the debt investment. Figure 9.10 presents the exit data for transactions for which the fund investor and investee firm were and were not resident in the same province at the time of first investment. Exits are more likely to be acquisitions when the investee and investor were not in the same province, and more likely to be IPOs for investments in the same province. Note that secondary sales and write-offs are more likely for investments in the same province. Figure 9.6 indicates that LSVCCs are more likely to experience buybacks and secondary sales. LSVCCs are also bound by statute to invest in the same province; hence, it not surprising that buybacks and secondary sales are observed where the investor and investee are resident in the same province. Figure 9.11 shows the exit outcomes for the actual province in which the entrepreneurial firm is based. Canada comprises 10 provinces: British Columbia (BC),
214
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
70.00 60.00
(%)
50.00 40.00 30.00 20.00 10.00 Investor and Investee not in same Province Investor and Investee in same Province
0.00 IPO
Acquisition
Secondary sale
Buyback
Write-off
Figure 9.10 Percentage of exits for intra- and inter-provincial investment.
100.00 90.00 80.00 70.00
(%)
60.00 50.00 40.00 30.00 20.00 10.00 Acquisition Write-off Buyback Secondary sale IPO
0.00 BC
AB
SK
MB
ON
OC
NF
NB
NS
PEI
NT
Figure 9.11 Percentage of exit outcomes for each province in which entrepreneurial firm is based.
Alberta (AB), Saskatchewan (SK), Manitoba (MB), Ontario (ON), Quebec (QC), Newfoundland (NF), New Brunswick (NB), Nova Scotia (NS), and Prince Edward Island (PEI). In addition, there are the Northern Territories (NT). Very few exits are from firms based in the Maritime Provinces (NF, BB, NS, and PEI). There was only one exit from NB (which was an IPO). The majority of activity is from BC, ON, and QC. VC-backed firms in BC, ON, and QC have had a greater number of IPOs
The profile of venture capital exits in Canada
215
and acquisitions than firms in the other provinces (Table 9.1), although the IPO and acquisition exit outcomes as a fraction of total exits within the province is high in AB, SK, NS, and NB (Figures 9.11 and 9.12). Figure 9.12 presents complementary evidence to Figures 9.10 and 9.11 by showing the exit outcomes in relation to the location of the investor fund. We may expect regional differences that are in part associated with different skill levels among VCs in different provinces, as well as proximity to greater levels of economic activity. The data indicate IPOs are more likely to have been financed by VCs based in ON. Write-offs are more likely to have resulted among VCs in SK and NS. Foreign investors are more likely to have exited by acquisitions. Figure 9.13 depicts the exit outcomes by the duration of investment (time in years from first investment to exit) and the number of syndicated VC investors at the time of first investment. IPOs and acquisitions have a slightly higher average number of syndicated VC investors relative to secondary sales and buybacks. This is in part consistent with the idea that syndication facilitates screening and valueadded. Cumming (2005c) also shows LSVCCs are less likely to syndicate, and this is consistent with the greater proportion of LSVCC exits as secondary sales buybacks (Figure 9.6). Write-offs also have a slightly higher average number of syndicated investors for the first investment round. The shortest time to exit is experienced for IPOs, followed by write-offs and acquisitions. Buybacks and secondary sales have the longest investment duration. Cumming and MacIntosh (2001; 2003a,b) hypothesized that longer investment duration is appropriate for investments in which information asymmetry is most pronounced. This would mean that investment duration would normally expect to be the longest for IPOs. The data, however, suggest that VCs rush to IPO to
80.00 70.00 60.00
(%)
50.00 40.00 30.00 20.00 10.00 Acquisition Write-off Buyback Secondary sale
0.00 BC
AB
SK
MB
ON
OC
NF
NB
IPO NS
PET
NT Foreign
Figure 9.12 Percentage of exit outcomes for each investor provincial location.
216
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
take advantage of stronger market conditions (see also Cumming and Johan, 2006a). Figure 9.14 presents the size of the average VC investment and deal size (including syndicated VC investments) for the first-round investment for each
5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 IPO Acquisition Secondary sale
Buyback
Write-off
Investment duration Number of syndicated investors
Figure 9.13 Average number of investment years from first investment to exit and first-round syndicated investors.
$9,00,000.00 $8,00,000.00 $7,00,000.00 $6,00,000.00 $5,00,000.00 $4,00,000.00 $3,00,000.00 $2,00,000.00 $1,00,000.00 $0.00 IPO
Acquisition Secondary sale
Buyback
Write-off
Average deal size Average amounts invested
Figure 9.14 Average initial amouts invested (real 2004 $Can) and average initial total deal size (including syndicated $ invested).
The profile of venture capital exits in Canada
217
exit type. IPOs and acquisitions offered much larger initial investments. Secondary sales, buybacks, and write-offs were much smaller investments, suggesting possibly riskier investments resulting in worse outcomes. These results are statistically significant, as indicated in Table 9.1.
9.5
Conclusion
This chapter presented evidence on VC-backed IPOs, acquisitions, secondary sales, buybacks, and write-offs in Canada over the period 1991–2004. Among other things, the data provide insight as to when investments are more likely to result in acquisitions versus IPOs. As in prior work, the data here are consistent with the view that the best investee firms are exited by means of IPOs or acquisitions, while secondary sales and buybacks are less desirable forms of exit (Cumming and MacIntosh, 2001, 2003a,b). The data show that VC governance affects the capacity for firms to achieve successful outcomes. The data further indicate VC-backed IPOs are more likely than acquisitions for common equity investments, investments in which the investee firm and VC investor are resident in the same province, and larger investments. The data also indicate a strong influence of government sponsored LSVCCs on VC exit outcomes in Canada. As of 2004, LSVCCs comprise approximately 50% of capital under management in Canada, and are an inferior governmentsponsored organizational structure that crowds out private VC funds. VCs in Canada have had less success in achieving IPOs than VCs in Australasia, Europe, and the U.S. The data are consistent with the view that the dominant presence of LSVCCs in Canada has given rise to a high proportion of less successful exits by means of buybacks and secondary sales, and fewer IPOs and acquisition exits. Consistent with prior work, our data suggest that the policy objective associated with LSVCCs in terms of creating sustainable firms that will enhance entrepreneurial firm outcomes does not appear to have been successful.
References Amit, R., Brander, J., and Zott, C. (1998). Why Do Venture Capital Firms Exist? Theory and Canadian Evidence. Journal of Business Venturing, 13:441–466. Barry, C. B., Muscarella, C. J., and Peavy III, J. W., and Vetsuypens, M. R. (1990). The Role of Venture Capitalists in the Creation of Public Companies: Evidence from the Going Public Process. Journal of Financial Economics, 27:447–471. Black, B. S., and Gilson, R. J. (1998). Venture Capital and the Structure of Capital Markets: Banks Versus Stock Markets. Journal of Financial Economics, 47:243–277. Brav, A., and Gompers, P. (1997). Myth or reality? The long-run underperformance of initial public offerings: evidence from venture and nonventure capital-backed companies. Journal of Finance, 52:1701–1821.
218
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Brav, A., and Gompers, P. A. (2003). The Role of Lockups in Initial Public Offerings. Review of Financial Studies, 16:1–29. Cumming, D. J. (2005a). Agency Costs, Institutions, Learning and Taxation in Venture Capital Contracting. Journal of Business Venturing, 20:573–622. Cumming, D. J. (2005b). Capital Structure in Venture Finance. Journal of Corporate Finance, 11:550–585. Cumming, D. J. (2005c). Adverse Selection and Capital Structure: Evidence from Venture Capital. Entrepreneurship Theory and Practice, 30:155–184. Cumming, D. J. (2006a). The Determinants of Venture Capital Portfolio Size: Empirical Evidence. Journal of Business, 79:1083–1126. Cumming, D. (2006b). Contracts and Exits in Venture Capital Finance. Working Paper. AFA 2003 washington DC meetings. Available on SSRN: http://ssrn.com/abstract-302695. Cumming, D. and Johan, S. A. (2006a). Do companies go public too early in Canada. Working Paper. TILEC Working Paper, Tilburg University, Center for Business Law. Cumming, D., and Johan, S. A. (2006b). Is It the Law or the Lawyers? Investment Covenants around the World. European Financial Management, 12:535–574. Cumming, D. J., and MacIntosh, J. G. (2001). Venture Capital Investment Duration in Canada and the United States. Journal of Multinational Financial Management, 11:445–463. Cumming, D. J., and MacIntosh, J. G. (2003a). A Cross-Country Comparison of Full and Partial Venture Capital Exits. Journal of Banking and Finance, 27:511–548. Cumming, D. J., and MacIntosh, J. G. (2003b). Venture Capital Exits in Canada and the United States. University of Toronto Law Journal, 53:101–200. Cumming, D. J., and MacIntosh, J. G. (2006). Crowding out Private Equity: Canadian Evidence. Journal of Business Venturing, 21:569–609. Cumming, D. J., and MacIntosh, J. G. (2007). Mutual Funds That Invest In Private Equity? An Analysis of Labour Sponsored Investment Funds. Cambridge Journal of Economics (forthcoming). Gompers, P. A. and Lerner, J. (1999). The Venture Capital Cycle. Cambridge: MIT Press. Halpern, P., ed. (1997). Financing Growth in Canada. University of Calgary Press. Lerner, J. (1994). Venture Capitalists and the Decision to Go Public. Journal of Financial Economics, 35:293–316. Macdonald, M. (1992), Venture Capital in Canada: A Guide and Sources. Toronto: Canadian Venture Capital Association. MacIntosh, J. G. (1994). Legal and Institutional Barriers to Financing Innovative Enterprise in Canada. Monograph prepared for the Government and Competitiveness Project, School of Policy Studies, Queen’s University, Kingston, Discussion paper, 94–10.
The profile of venture capital exits in Canada
219
Manigart, S., Sapienza, H., and Vermier, W. (1996). Venture capital Governance and Value Added in Four Countries. Journal of Business venturing II, 439–469. Megginson, W., and Weiss, K. (1991). Venture Capital Certification in Initial Public Offerings. Journal of Finance, 46:879–893. Manigart, S., H. Sapienza, and W. Vermier, 1996. Venture capital Governance and Value Added in Four Countries. Journal of Business venturing II, 439–469. Peng, L. (2001). Building a Venture Capital Index. Working Paper. Yale Center for International Finance. Ritter, J. R. (1984). The hot issue market of 1980. Journal of Business, 32:215–240. Sahlman, W. A. (1990). The Structure and Governance of Venture Capital Organizations. Journal of Financial Economics, 27:473–521. Schwienbacher, A. (2002). Venture Capital Exits in Europe and the United States. Working Paper. University of Amsterdam. Yosha, O. (1995). Information Disclosure Costs and the Choice of Financing Source. Journal of Financial Intermediation, 4:3–20. Zingales, L. (1995). Insider Ownership and the Decision to Go Public. Review of Economic Studies, 62:425–448.
This page intentionally left blank
Part Three Valuation and Irrationality in Takeover Decision Making
This page intentionally left blank
10 Executive compensation and managerial overconfidence: impact on risk taking and shareholder value in corporate acquisitions Sudi Sudarsanam and Jian Huang
Abstract One solution to this risk-related agency conflict is to structure the managers’ compensation in such a way that it provides risk incentives. Executive stock options offer a device to achieve this aim. We argue that an executive compensation contract is a heterogeneous mix of components with different risk incentives, some risk-inducing and some reinforcing managerial risk aversion; thus, we need a good understanding of the impact of these components on risk taking before modeling their impact on performance. We also consider the behavioral agency model that posits managers’ behavioral biases as influencing their risk preferences and focus on one particular psychological trait, i.e., overconfidence. Both monetary risk incentives and overconfidence can be complementary and lead to excessive risk taking. Avoiding such excessive risk taking or, conversely, minimizing inadequate risk taking by risk-averse managers, requires a new monitoring role for corporate governance. We present an integrated model in which risk taking and consequent firm performance are subject to the interacting influences of executive compensation structure, behavioral bias, and corporate governance.
10.1
Introduction
The traditional agency model assumes that corporate managers, the agents of shareholders, are rational utility maximizers and, given the appropriate incentives, work in a way that aligns their interests with those of their principals. An executive compensation contract is such an incentive device. One source of misalignment between managers and shareholders is their divergent risk preferences. The traditional agency model assumes that for a variety of reasons, e.g., undiversified human capital invested in their employer firm, managers may be risk-averse and under-invest in high-risk, albeit value-creating, projects. Such passing up of potentially positive NPV projects imposes an opportunity cost on shareholders
224
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
who are risk-neutral. Varying the compensation contract to include elements such as stock options that carry high risk incentives for managers is a possible solution to the risk misalignment problem. Conditional upon such risk incentives, managers will avoid underinvestment, thereby creating more value for shareholders. The behavioral agency view deviates from the assumption of rational managers and conceptualizes them as psychologically prone to various behavioral biases, e.g., loss aversion, overconfidence, or hubris. These biases influence managerial risk-taking behavior and may result in corporate investment and financing decisions characterized by excessive risk or inadequate risk in comparison to an optimal risk level that aligns managers’ and shareholders’ interests. In recent years many researchers have studied behavioral biases in the context of corporate acquisitions (Heaton, 2002; Malmendier and Tate, 2005c), although Roll (1986) was the first to identify hubris as a possible determinant of overpayment and value destruction in acquisitions. An important implication of the behavioral biases such as overconfidence and hubris is that managers may choose excessively risky investments, again causing a misalignment between themselves and their shareholders. Juxtaposing the traditional and behavioral agency models raises some interesting issues. One issue is how monetary incentives embedded in compensation contracts and behavioral biases interact. Under some circumstances, risk incentives through compensation contracts may be redundant and, under others, may reinforce behavioral biases. Are the effects of monetary risk incentives and nonmonetary psychological proclivities substitutive or complimentary? In contrast to most of the studies in behavioral finance that postulate the adverse shareholder value effects of managerial biases (Roll, 1986; Heaton, 2002; Malmendier and Tate, 2005b; Gervais, Heaton, and Odean, 2005) posit that overconfident managers are naturally more likely to take risks than are less confident, risk-averse and rational managers, and therefore fewer pay incentives are needed to motivate them to undertake risky investments. Managerial overconfidence can, therefore, provide an alternative remedy for the risk-related agency problem, i.e., underinvestment. Combining the two agency schools, the risk-related agency problem can be considered to encompass both underinvestment in risky projects caused by managerial risk aversion and overinvestment in risky projects induced by managerial risk-seeking. How do shareholders ensure that this mix of managerial risk incentives leads to neither too much risk taking nor too little? Corporate governance mechanisms through their monitoring role can steer managers toward optimal-risk investment and avoid firm-value destruction from either risk deficit or risk excess on the part of their managers (Wright, Ferris, Sarin, and Awasthi, 1996; Hayward and Hambrick, 1997; Jensen and Murphy, 2004; Malmendier and Tate, 2005a; Paredes, 2005). This chapter aims to provide an integrated framework to explore the relationships among executive compensation, CEO behavioral bias, corporate governance, risk-taking, and firm performance. Even though the framework could be applied to all corporate finance and investment decisions, we focus our analysis
Executive compensation and managerial overconfidence
225
on corporate acquisitions because they are very often relatively large and externally observable; they are also discretionary long-term investments that can substantially alter the risk profile of acquirers and thereby exacerbate the potential risk-related conflict of interests between managers and shareholders. The remainder of the chapter proceeds as follows. Section 10.2 discusses risk alignment of shareholder and managerial interests under the traditional agency model. Section 10.3 reviews the behavioral perspective on risk taking. Section 10.4 explores the joint impact of executive compensation and CEO behavioral biases. Section 10.5 examines the joint impact in the context of corporate acquisitions and empirical results of relevant and recent studies of such impact are presented in Section 10.6. Section 10.7 identifies unresolved issues in theory and empirical research. Section 8 concludes with suggestions for future research.
10.2
Alignment of shareholder and managerial interests
10.2.1
Traditional agency perspective and solutions
The conflict between shareholders and corporate management arising from the separation of ownership and control in a publicly held corporation has been well recognized in the modern literature since Berle and Means (1932)—indeed, since the writings of Adam Smith (Smith, 1776:700). Managers as agents of shareholders may make investment and financing decisions that serve their own interests to the detriment of those of shareholders. Since the seminal work of Jensen and Meckling (1976), the literature has focused on how managerial ownership and compensation contracting can help to align the interests of the managers with those of the shareholders (Baker, Jensen, and Murphy, 1988).
Managerial stock ownership and corporate governance An executive compensation package or equity ownership that enhances managers’ wealth in line with increase in corporate performance and firm’s stock value has generally been considered a solution to the agency problem (Baker, Jensen, and Murphy, 1988). This requires that managers be rewarded in the form of stockrelated compensation so that managers have an incentive to take decisions that will enhance stock value. An alternative, though not mutually exclusive, device is an effective corporate governance system that monitors and regulates managerial decisions to ensure their conformity to shareholder interests. The corporate governance mechanism is a complex web of mutually interacting control mechanisms (Sudarsanam, 2000). Along with executive compensation, other corporate governance mechanisms can, through their monitoring role, help to ensure that managers act in the best interests of shareholders in making corporate investment and financing decisions (Wright, Ferris, Sarin, and Awasthi, 1996; Hermalin and Weisbach 2003). Jensen and Murphy (2004) emphasize the importance of robust corporate governance in negotiating compensation contracts, monitoring the conduct of executives, and ensuring that compensation
226
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
contracts do not provide them with skewed incentives for making investment decisions. External block (share) holders have strong economic incentives to undertake effective monitoring because they are able to capture a large fraction of the wealth gains from the corporate value enhancement (Shleifer and Vishny, 1986). They can force acquirer managers to examine carefully their acquisition strategies and reduce the scope for suboptimal risk avoidance or risk seeking. Another corporate control mechanism is the composition of the board of directors. Corporate boards in the United States and several other countries, like the United Kingdom, generally comprise both executive directors and nonexecutive directors. Nonexecutive or independent directors have the responsibility to provide independent monitoring and control of executive managers on behalf of external shareholders (Fama and Jensen, 1983). Therefore, the ability of the board to act as a guardian of stockholder welfare is a function of board composition. Furthermore, the vigilance of board, as suggested by Hermalin (2005) can co-vary with executive compensation. Neither executive compensation nor corporate governance systems are free of deficiencies, and we review some evidence that points to their failure in aligning shareholder and manager interests. We now discuss the different components of executive compensation contracts, present the recent trends in the compensation of U.S. chief executives, and highlight the trends in these components when benchmarked against the stock market performance as a measure of shareholder value gains. We examine the empirical evidence concerning the link between compensation and shareholder value performance. We emphasize that different compensation components provide different risk incentives for managers, with some likely to accentuate risk aversion while others encourage risk taking. We then discuss the evidence concerning the effectiveness of corporate governance mechanisms in aligning shareholder and managerial interests.
10.2.2
Executive compensation and its components
Because managers control access to corporate information, with the attendant information asymmetry, shareholders have limited abilities to monitor whether the risk level of a project chosen by managers is optimal, so that it leads to shareholder advantage. It is thus in the interests of shareholders to design appropriate corporate control or incentive mechanisms to drive managers to select valueenhancing risky projects. A compensation contract that links a portion of compensation to firm performance is one of the key corporate control devices. While the overall compensation package may be intended to motivate managers to increase shareholder value, the risk incentives are not uniform across the components of the package. We therefore need to disaggregate the package in order to understand how it influences managerial risk-taking behavior. Mainly, compensation contracts assume three types: ●
fixed compensation (i.e., any contractually guaranteed pay), such as basic salary, fees paid to nonexecutive directors, pension contributions, and related benefits;
Executive compensation and managerial overconfidence
●
●
227
short-term incentive plan, such as annual bonus, which is generally tied to yearly accounting performance; long-term incentive plan (LTIP), including LTIP cash or share awards and share options. Long-term incentive plans are typically tied to multiyear performance, either accounting-based or stock market–based.
Fixed compensation and annual bonus are largely in the form of cash (cash compensation). LTIP shares and stock options are equity-based compensation that aligns the interest between shareholders and managers by turning managers into owners, whereas cash compensation makes no such link. United Kingdom directors typically hold company stocks, which are accumulated through realized LTIP shares or exercised options in the past, or are acquired through the market (Conyon and Murphy, 2000). These shareholdings relate managerial wealth to firm stock price. Figure 10.1 reports the change in the total value and composition of CEO compensation of S&P 500 firms from 1992 to 2004.1 The stock option is the instrument that gives the recipient the right to buy a share of stock at a prespecified exercise price for a prespecified period. Stock is the restricted stock grant that is the major component of long-term incentive plans (LTIPs). Cash consists of salary and bonus, while others are other annual compensation categorized neither as salary nor bonus. Total compensation increased steadily from 1995 to 1999. Equity-based compensation, including stock and stock options, represents about 50% of the total compensation received by CEOs from 1996 onward. CEO total compensation soared in 1999 and 2000, largely owing to the dramatic increase in
Compensation Value ($000s)
16,000 14,000 12,000 10,000 8000 6000 4000 2000 0 92
93
94
95 96
97
98
99
00
01
02
03
04
Year Option
Cash
Stock
Other
Figure 10.1 CEO compensation and composition in the United States, 1992–2004. 1
The value of CEO compensation in the figure is adjusted for inflation and reported in 2004 dollar.
228
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
2000
10,000
1000 0
0 92
93
94
95
96
97
Stock option grant ($000s)
(a)
98 99 Year
00
01
02
03
04
2000
12,000 10,000 8000 6000 4000 2000 0
1500 1000 500 0 92
93
94
95
96
97
(b)
98
99
00
01
02
03
04
Year
1500
1500 1000
1000
500
500 0
0 92
93
94
95
96
97
(c)
Cash compensation ($000s)
S&P 500
2000
2000
98
99
00
01
02
03
04
Year 3500 3000 2500 2000 1500 1000 500 0
2000 1500 1000 500 0 92
93
94
95
96
97
98
99
00
Year
Figure 10.2 CEO compensation and S&P 500 index.
01
02
03
04
S&P 500
Stock grant ($000s)
2500
(d)
S&P 500
20,000
S&P 500
Total compensation ($000s)
stock option grants. Compared to the level in 2001, the value of CEO total compensation diminished by more than 30% in 2002, again largely owing to the shrinkage in stock option grant. Total CEO pay was fairly stable from 2002 to 2004 as the increase in stock grant offset the further decline in stock options grant. Figure 10.2 shows the change in CEO compensation value of S&P 500 firms and the evolution of S&P 500 index level from 1992 to 2004. Total CEO
Executive compensation and managerial overconfidence
229
compensation value and the value of its components are shown as columns, and the trend in the S&P index level is represented by the lines in the figures. As indicated in the top left figure, the trend in CEO compensation is highly associated with that in the contemporary S&P 500 index level. The breakdown of CEO compensation reveals that the alignment is trending away from stock options grants. The value of restricted stock grant, however, experienced a different trend during the period from 1999 to 2000. In 1999 the proportion of restricted stock grant slumps by more than 70% while the S&P 500 peaks at 1469.25 at the end of 1999. Apart from this anomalous year, the trend in value of stock grants has been one of steady increase independent of the stock market performance. The increase in total compensation and the dominance of the stock-based components during the 1990s is similar to that reported in several U.S. studies (Bryan, Hwang, and Lilien, 2000; Hall and Murphy, 2002; Bebchuck and Fried, 2004; Jensen and Murphy, 2004). These studies also note the relatively greater share of stock options within the stock-based compensation. The value of cash compensation which comprises salary and bonus increases steadily regardless the performance of S&P 500 index. Increase in cash compensation is, however, less steep than the corresponding increase in stock grant value; but both track shareholder value changes less faithfully than stock options. These trends suggest that some executive compensation components are more closely aligned to shareholder value than others.2
10.2.3
Link between executive pay and shareholder value
Pay for performance or pay without performance? According to a survey prepared by Mercer Human Resource Consulting for The Wall Street Journal, during 2005, pay for CEOs at the 10 businesses with the highest shareholder returns grew by 51.3% to more than $10.2 million. The 10 CEOs whose shareholders took the biggest hit last year suffered a 72.5% drop in compensation to less than $1.6 million.3 Despite the seemingly positive relation between CEO pay and performance, the pay-for-performance equation (typically between executive pay and shareholder return) remains controversial. In its report, Pay for Failure: The Compensation Committees Responsible, The Corporate Library found 11 companies that paid out $865 million to CEOs over the past five years, even though these companies lost a total of $640 billion in shareholder value. Those on the list are AT&T, BellSouth, Hewlett-Packard, Home Depot, Lucent Technologies, Merck, Pfizer, Safeway, Time Warner, Verizon
2
In the U.K. also during the 1990s stock-based compensation became popular. The median CEO option holdings increased from 0.09% in 1991 to 0.11% in 1997. In 1997 stock option grants averaged only 10% of the total compensation and LTIP shares about 9%. In contrast to the U.S. CEO compensation scene, stock-based compensation was still only a small part of CEO compensation in the U.K. (Conyon and Murphy, 2000). 3 Knowledge@Wharton, ‘CEO Pay: A Window into Corporate Governance’, May 17, 2006.
230
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Communications, and Wal-Mart.4 Apparently, extravagant compensation for CEOs has caused outrage among the public and provoked an outcry from investors (Kirkland and Burke, 2006). So far, academic studies on the relation between pay and performance, especially the relation between equity-based pay and firm performance, have not reached consistent conclusions. Several U.S. studies show that compensation is not strongly related to performance. CEOs received windfall payments from events, e.g., favorable settlement of lawsuits, although these suits had little to do with the incumbent CEO’s tenure (Bebchuk and Fried, 2004). Similarly, managers were rewarded for the favorable impact of sector-wide developments, e.g., oil price increases or exchange rate changes (Bertrand and Mullainathan, 2001). Grinstein and Hribar (2004) report that acquiring-company CEOs received deal completion bonuses, even when the acquirer shareholders lost money following it. The higher the bonus, the greater was the loss suffered by the shareholders. Even when their firms became distressed some CEOs received a bonus or a “golden parachute” worth several millions of dollars (Bebchuk and Fried, 2004). Thus the link between cash compensation and performance seems rather weak. On the other hand, Murphy (1986) documents a positive relation between salary and bonus and stock returns and between stock-based compensation and stock returns during 1975–1984. As for the effect of equity-based compensation, again the empirical evidence on a positive link to shareholder value performance is mixed. McConnell and Servaes (1990), Morck, Shleifer, and Vishny (1988), and Hermalin and Weisbach (1991) suggest a positive relation between managerial stock ownership and firm value up to a certain level and then a negative relation. Mehran (1995) argues that firm value and return on assets are both positively related to the percentage of equity held by managers and the percentage of executive compensation that is equity-based. Core, Daniel, and Naveen (1999) report that CEOs in firms with greater agency problems receive greater compensation and that these firms also perform relatively badly. Ittner, Lambert, and Larcker (2003) examine the impact of less-than-expected option grants on subsequent performance in both accounting and stock return terms in the new economy firms during 1999-2000 and find it negative. Habib and Ljungqvist (2005??) find that boards gave CEOs too many options with diminishing marginal benefits to shareholders. DeFusco, Johnson, and Zorn (1991) report that shareholder value gains as well as return on assets declined following the adoption of executive stock option plans. If the evidence on the pay-for-performance link is at best tenuous, what about the other control devices?
10.2.4
How effective is board and large-shareholder monitoring?
The empirical findings about the impact of nonexecutive directors on managerial firm performance are also inconclusive. Many studies identify a positive effect of 4
Knowledge@Wharton, ‘CEO Pay: A Window into Corporate Governance’, May 17, 2006.
Executive compensation and managerial overconfidence
231
board independence (Hayward and Hambrick, 1997; Franks, Mayer, and Renneboog, 2001; Sudarsanam and Mahate, 2006) on corporate performance in the context of mergers and restructuring. Other studies including Yermack (1996) and Weir, Laing, and Mcknight (2002), report insignificant effects, whereas Weir and Laing (2000) find negative effect. Bhagat and Black (2002) report that lowprofitability firms increase the independence of their boards of directors but find no evidence that this strategy works. Empirical evidence on the effectiveness of external block holder monitoring is also mixed. Some existing studies report that they have a positive impact on corporate performance (McConnell and Servaes, 1990; Tufano, 1996). However, other studies find them ineffective (Sudarsanam, Holl, and Salami, 1996; Wright, Ferris, Sarin, and Awasthi, 1996; Weir, Laing, and Mcknight, 2002). Faccio and Lasfer (2000) observe that the U.K. pension funds are not effective monitors and do not add value to their investee companies. The inconclusiveness of the results concerning the impact of executive compensation and corporate monitoring on firm performance may arise from several factors, such as data and methodological limitations. However, it may also be due to model misspecification that ignores some intermediate variables between compensation and monitoring on the one hand and performance on the other. How these devices affect managerial risk incentives and risk-taking behavior which, in turn, influence firm performance, may be a missing link. The previously cited studies do not explicitly recognize an important source of misalignment between shareholders and managers—i.e., in their risk preferences.
10.2.5
Shareholder and manager misalignment in risk taking
Why risk preferences may differ between them Shareholders are considered risk-neutral since they can hold their wealth in well diversified portfolios and thereby diversify away firm-specific risk. On the contrary, managers, whose human capital is invested in their own firm, hold undiversified portfolios. Additionally, when their money capital is invested in their company’s stock, the degree of nondiversification is intensified. The undiversified portfolio exposes managers to a high level of both systematic and firm-specific risk, inducing managers to be risk-averse. As a consequence, the risk-averse manager may behave opportunistically and pass up risky, but value-enhancing investment opportunities, to the detriment of shareholder value (Smith and Stulz, 1985; Guay, 1999, Parrino et al. 2005).
Fine-tuning compensation contracts for different risk preferences Different compensation contracts have different effects on aligning the risk preferences of managers and shareholders. This phenomenon has important consequences for the risk-taking behavior of managers and their performance.
232
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Cash compensation Fixed compensation protects managers against fluctuation in corporate performance or shareholder value. However, it provides no incentive for managers to increase firm risk (Larcker, 1992). It can even make managers avoid risk, which increases the probability of financial distress or bankruptcy (Lambert and Larcker, 1991). Annual bonus often ties managers’ remuneration to yearly accounting numbers. Narayanan (1996) demonstrates theoretically that managers select projects yielding short-term profits and that cash-only incentive contracts result in underinvestment in long-term projects that carry greater risks. Coles, Daniel, and Naveen (2006) provide evidence that cash compensation discourages managerial risk taking. They find that it is negatively related to R&D investment, encourages firm diversification, and induces firms to reduce leverage.
Long-term incentive plans (LTIPs) The remedy to the short-horizon problem induced by cash compensation is to provide managers with long-term compensation plans such as LTIP (cash or share) awards and stock options. Such plans lengthen managers’ decision horizons to several years, since the compensation is deferred. Thus managers have incentives to invest in projects whose payoff is long-term and hence likely to be more risky. In the United States, LTIP stocks, called restricted stocks, are much less common than stock options, although after the stock market collapse of 1999, they have become more popular (see Figure 10.1). These stocks become vested; i.e., ownership is transferred to directors, upon the achievement of agreed performance targets by the managers.5 The value of LTIP shares, hence managerial wealth, varies according to the underlying stock price. LTIP shares provide an incentive for managers to improve shareholder value performance. However, the linear payoff means managerial wealth declines with a fall in stock price and exposes managers to too much risk.6 This circumstance may increase managerial risk aversion (Smith and Stulz, 1985; Bryan, Hwang, and Lilien, 2000; Ryan and Wiggins, 2002).
5
In the U.K., LTIPs are normally awards or grants of shares that become vested only upon attainment of certain performance objectives over a period of time, generally 3 years. There are three conditions for the LTIP share awards before they can be transferred to directors. First condition relates to the performance objective, either accounting-based or stock market-based. The objective has to be achieved at the end of a specified period of time, i.e. vesting period. Vesting period therefore is the second condition. The third condition is the leaving constraint. Directors lose unvested LTIP shares if they leave voluntarily or involuntarily during the vesting period. These conditions make LTIP shares rather contingent in nature (Conyon and Murphy, 2000). 6 Bebchuk and Fried (2004, p170–1) point out that with restricted stocks managers can still enjoy windfall gains e.g. from the general stock market boom unrelated to the firm’s performance. By the same token the restricted stock also exposes managers to ‘windfall losses’.
Executive compensation and managerial overconfidence
233
LTIPs can also be awarded in cash. LTIP cash awards are more like annual bonus but with a vesting period of more than one year. Once a director achieves the performance target at the end of the vesting period, he or she is awarded the cash. Unlike LTIP shares, which have a linear payoff structure, the value of LTIP cash does not vary with the value of the company’s stock price. Since excessive risk taking can jeopardize the LTIP cash awards, managers may avoid high-risk decisions.7
Stock options A stock option given to managers creates long-term incentives to improve the stock price. It gives the right to purchase stock at a prespecified exercise price for a prespecified term conditional upon the achievement of certain predetermined performance benchmarks. With a stock option, above the exercise price, a $ increase (decrease) in firm stock value increases (decreases) managerial wealth by a $, but below the exercise price, a fall in the firm stock value does not reduce managerial wealth. Thus stock options have an asymmetric, convex payoff structure. Such a payoff structure, insuring managers against losses but giving them substantial upside gains, can motivate managers to invest in high, even excessively high, risk projects (Smith and Stulz, 1985; Core, Daniel, and Naveen, 2003). Datta, Iskandar-Datta, and Raman (2001) find that stock options encourage corporate executives to undertake riskier acquisitions than do other compensation forms.8 Several other recent studies also demonstrate a link between stock options and increased risk taking, and these are discussed in more detail in succeeding sections (Coles, Daniel, and Naveen, 2006; Williams and Rao, 2006).
Managerial shareholdings As with LTIP shares, managerial shareholding has a linear payoff function whose value is exposed to the variability of firm stock performance, thereby increasing managerial risk aversion (Smith and Stulz, 1985; Core, Daniel, and Naveen, 2003). Ryan and Wiggins (2002) find that CEO shareholdings significantly negatively affect firm R&D investment. Tufano (1996) finds that, in gold-mining firms, management teams with a higher level of managerial shareholdings conduct more corporate hedging to reduce gold price risk.
7 8
LTIP cash awards are rare (Stathopoulos et al, 2004). Some analysts have criticised that stock options turned directors into speculators and induced them to indulge in excessively risky projects to push up the stock prices during the dotcom bubble of the late 1990s (Elson, 2003). This line of argument has also been recognized by the literature on stock option repricing. For instance, Rogers (2005) states that deep out-of-money options can drive CEOs to engage in excessive risk taking to push up stock price at any (company) cost. Repricing the exercise price of such options can lessen the CEOs’ incentive toward inordinate gambles.
234
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Are risk incentives through compensation necessary? Nonetheless, other studies offer different views on stock-option compensation. Meulbroek (2001) argues that, as shareholders set constraints on a stock option award— including a vesting period—undiversified risk-averse managers value the stock option they receive from the firm less than its cost to the shareholders, largely because managers cannot trade or exercise the stock option freely.9 Hall and Murphy (2002) offer a similar argument concerning the divergence between the cost and the value of the stock option provision in executive compensation. They claim that the stock option is a particularly expensive way to award compensation. Ross (2004) argues that no incentives contract exists that will make all expected utility maximizers less risk-averse or more risk-prone. Their impact is likely to vary across individual decision makers. Carpenter (2000) argues that the convex payoff of stock options does not necessarily encourage managerial risk taking. She demonstrates through a theoretical model that the effects of options on different managers depend on their utility function. If the utility function is hyperbolic absolute risk-averse or constant relative risk-averse, granting them more stock options that they cannot hedge increases the volatility of their personal portfolios. To offset this, they tend to reduce the volatility of the underlying asset, perhaps through unrelated diversification or switching to low-risk businesses.
10.2.6
Excessive risk taking—the dark side of stock options
The flipside of risk aversion is excessive risk taking. Within the history of corporate America, corporate scandals or fraud can, to certain extent, be attributed to excessive risk taking.10 Skeel (2006) observes, from his analysis of corporate scandals, that excessive and sometimes fraudulent risk taking, as well as competition and the increasing size and complexity of organization, has been at the heart of every corporate breakdown in the early 21st century, just as it was in the previous century. CEO compensation, especially equity-based compensation, which is positively related to the size and complexity of firm, plays an influential role in motivating excessive risk taking. The suboptimal CEO compensation contract, i.e., risk-taking incentives beyond the optimal level, may cause the CEO to undertake tremendous risk, which results in great loss of shareholder value, if not the collapse of the firm.
Measuring risk incentives—Delta and Vega Early empirical studies in executive compensation utilize the value of different compensation components, such as bonus or LTIP stocks or stock options, as a 9
Bebchuk and Fried (2004, ch. 14) highlight a number of ways managers in the U.S. can get around the restrictions on trading in stocks or on hedging their portfolio exposure to their employer firm. Thus in practice the constraints that force manager into holding undiversified portfolios may be weaker than in theory. 10 Knowledge@Wharton, ‘Corporate Fraud on Trail: What Have We Learned’, March 30, 2006.
Executive compensation and managerial overconfidence
235
measure of incentives.11 According to Core and Guay (2002) and Coles, Daniel, and Naveen, (2006), it is a weak proxy, whereas Delta and Vega provide more direct measures of managerial incentives. Delta is defined as the change in the dollar value of the manager’s wealth for a 1% change in stock price, while Vega is the change in dollar value of the manager’s wealth for a 1% change in the annualized standard deviation of stock returns. While Vega reflects the managerial wealth gains from an increase in volatility of the firm’s returns, Delta reflects such gains from a change in firm value. Thus Vega captures risk incentives better than Delta. Figure 10.3 indicates the trend in the average dollar value (in constant 2004 dollar terms) of Delta and Vega for the S&P 500 U.S. firms during 1992–2004. The dollar value of Delta and Vega experience similar trends before 1998, even though the total value of Delta is generally 10 times more than that of Vega. Mean Delta value peaks in 1999 at around U.S. $7 millions ($m hereafter). It then decreases dramatically to the level of just above $2m in 2004. On the contrary, Vega increases steadily from $0.1 m level in 1992 to $0.5m in 2004.12 This phenomenon suggests that managerial risk-taking incentives in the United States embedded in executive compensation have increased quite substantially over the 8000
600
Value ($000s)
7000
500
6000 400
5000
300
4000 3000
200
2000 100
1000
0
0 92 93 94 95 96 97 98 99 00 01 02 03 04 Year Delta
Vega
Figure 10.3 Mean Delta and Vega value, 1992–2004. 11
For example, Datta Iskandar-Datta, and Raman (2001) use the sum of the value of new stock options granted to the top five executives as a percentage of total compensation paid to them as a measure of incentives. Mehran (1995) uses equity-based compensation as a proportion of total compensation as an incentive measure. 12 Core and Guay (1999) reports the mean Delta value of $ 0.56 m for the sample period 1992 to 1996. In Coles Daniel, and Naveen (2006), using 2001 compensation data, the mean Delta is $0.584 m while the mean Vega is $0.063 m. According to our calculation, the average Delta and Vega of the S&P 500 firms are $3.07 m and $0.37 m respectively. The reason that our estimated Delta and Vega are much larger is simply because S&P 500 firms are the largest firms which can be 10 times bigger than the firms in midcap and smallcap S&P indices.
236
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
last decade and remain high. It is arguable whether such high incentives caused U.S. managers to engage in suboptimally risky investments and acquisition.
10.3
Managers’ behavioral biases and risk taking
CEO behavioral biases, however, can also lead to corporate investment distortion, including excessive risk taking (Malmendier and Tate, 2005). Rovenpor (1993) examines the role that four CEO personal characteristics, including preference for organizational growth, belief in synergy, need for power, and self-confidence play in encouraging companies to engage in mergers and acquisitions. Rovenpor finds that these four CEO characteristics are highly and positively related to the level of mergers and acquisitions activities. Hayward and Hambrick’s (1997) study is the first to introduce observable proxies to measure the psychological construct termed CEO hubris. They put forward three variables—recent organizational success, recent media praise for CEOs, and the CEO’s self-importance—as proxies for CEO hubris. CEO’s self-importance is captured buy his/her relatively high pay. Based on managerial optimism, Heaton (2002) provides a theoretical model of CEO overoptimism and corporate investment. The major concern of that paper is that optimistic managers overvalue their own corporate projects while believing that the capital market undervalues their corporate shares. According to their biased belief, they may decline positive net present value (NPV) projects that must be financed externally, but undertake projects with negative NPV if they have enough free cash flow. These managers do not view themselves as wasting the free cash flow but as serving shareholder interests. This amounts to “honest mismanagement” of corporate resources (Paredes, 2005).
10.3.1
Behavioral perspective on risk taking—the behavioral agency model
A CEO does not live by compensation alone “Forget the salary packages, look at the size of their egos”13 heads a recent article in the Financial Times (Francesco Guerrera, 26th August 2006), suggesting that CEOs’ oversized egos, verging on narcissism, may induce their grandiose and often high-risk corporate ventures, thereby having an influence independent of the monetary incentives for risk taking. Recent papers on behavioral agency models argue that managerial risk taking is not a mere deviation from the traditional agency assumption of rational risk aversion (Wiseman and Gomez-Meijia, 1998; Wright, Ferris, Sarin, and Awasthi, 2001; Barberis and Thaler, 2002). Rational 13
In a new study, Chatterjee and Hambrick (2006) argue that “corporate chieftains with super sized egos favor grandiose and higher-risk strategies…acquisitions, large-scale product launches and aggressive international expansion…they can hit big but they can also miss big.”.
Executive compensation and managerial overconfidence
237
risk aversion may be a restrictive and unrealistic assumption about human behavior. Instead, managers may be “irrational” and, under psychological influences, exhibit different risk attitudes in different situations.14 Therefore, incentive alignment mechanisms, designed on the assumption that managers are rational and risk-averse, are unlikely to have much effect (in the desired direction) on irrational and risk-seeking managers. “These managers think that they are maximizing firm value, even if in reality, they are not. Since they think that they are already doing the right thing, stock options or debt are unlikely to change their behavior” (Barberis and Thaler, 2002).15 Such managerial attitudes may arise from behavioral biases such as overconfidence, overoptimism, and hubris. Existing literature suggests that these three related managerial biases may induce excessive risk taking. Researchers have found that managers, particularly senior managers, are prone to display overconfidence (March and Shapira, 1987; Goel and Thakor, 2005). March and Shapira (1987) find managers view risk as controllable and modifiable, and believe themselves clearly able to distinguish between gambling (i.e., where the chances of win or loss are uncontrollable) and risk taking (i.e., where uncertainty can be reduced by skill or information).16 Goel and Thakor (2005) find that managers are likely to demonstrate overconfidence when they are competing for leadership, and overconfident managers have a higher probability of being chosen as the leader than are otherwise identical managers.
Impact on managers’ risk-taking behavior Overoptimistic individuals underestimate the likelihood of hazards affecting them personally, and entertain the belief that “the future will be great, especially for me.” Overoptimistic managers may willingly expose themselves to a substantial degrees of risk because they misjudge the odds or rely on overly optimistic forecasts (Kahneman and Lovallo, 1993). Heaton (2002) and Malmendier and Tate (2005c) find that overoptimistic managers often show an upward bias in their
14
Such deviation from the normative expected utility maximisation (EU) model is the subject of prospect theory propounded by Kahneman and Tversky (1979). According to this model, decision makers in evaluating risky choices derive utility not from the final wealth position resulting from the gamble but from the potential losses and gains attendant upon the gamble and may attach differential utilities to losses and gains. For example, loss aversion, a psychological trait may lead to greater disutility from loss than the utility attached to gain of a similar amount. Some of the individual investment decisions and corporate investment and financing decisions may be analyzed and explained using prospect theory (Shefrin, 2005; Barberis and Thaler, 2002). However, in this chapter we do not explicitly consider this framework. 15 While compensation-based risk incentives aim to alter managerial motives and risk preferences, the behavioral model aims to explain how managerial biases influence their risk preferences. Another distinction is that the former fine-tunes extrinsic motivation whereas the latter seeks to exploit intrinsic motivation (Bénabou and Tirole, 2003). 16 This can be illustrated by the words of the president of a successful high technology company: “In starting my company I didn’t gamble; I was confident we were going to succeed” (March and Shapira, 1987, p.1410).
238
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
cash flow forecasts for investment projects. This causes the managers to overvalue their firms’ investment opportunities and to undertake projects that actually have negative cash flows.
Benefits of overconfidence Self-confidence, which in the extreme may yield to overconfidence, is not devoid of virtue. It is associated with self-esteem, willingness to take risks, decisiveness, and commitment. Diffident leaders are unlikely to inspire others to follow them. Confidence may signal competence and determination to the competitors. Selfconfident and charismatic leaders can instill self-confidence in their subordinates, encouraging them to be more innovative and take more risks than they otherwise might (Bénabou and Tirole, 2003; Gervais and Goldstein, 2004; Gervais, Heaton, and Odean, 2005; Goel and Thakor, 2005; Paredes, 2005). Jack Welch, the former CEO of GE, notes the thin line between self-confidence and hubris. After describing GE’s disastrous acquisition of Kidder Peabody, Welch (2001) says: “There is only a razor’s edge between self-confidence and hubris. This time hubris won and taught me a lesson I’d never forget.” This comment suggests that behavioral biases can provide the tipping point between optimal and excessive risk taking by top managers. In Section 10.3 we examined the compensation-based risk incentives; in the next section we consider the alternative behavioral perspective on risk taking. We discuss how monetary incentives and nonmonetary or psychological impulses interact to drive managers in their risky investment decisions in section 10.4.
10.4
Joint impact of executive compensation and overconfidence
The extant studies do not provide a framework that explains the relationships among executive compensation, risk-taking, and firm performance in an integrated manner. Studies reviewed in Section 10.2.3 link executive compensation or its components directly to firm performance. Other studies, reviewed in Section 10.2.5, examine the relation between executive compensation and risktaking decisions. Each of these approaches is, however, inadequate as risk taking is an intermediate variable that determines performance and is in turn influenced by incentives. Any study that links incentives directly to performance fails to account for the risk-taking decision (e.g., the risk profile of an acquisition) in the middle. Any study that links incentives to risk taking alone does not explain whether such risk taking is optimal in terms of performance outcome. Figure 10.4 illustrates a framework that links compensation to investment decision and the investment decision, in turn, to performance. Incentives affect the riskiness of the investment decision.
Executive compensation and managerial overconfidence
239
Executive Compensation
Investment Decisions
Firm Performance
Figure 10.4 Relations between executive compensation, risk, and firm performance
10.4.1
Understanding risk-taking behavior and its impact on corporate performance
In this chapter, we conjecture that the inconclusive results from extant empirical studies may be due to the lack of a comprehensive and integrated framework that takes into account the firm and managers’ characteristics. As previously discussed, overconfidence of CEO could be a genuine alternative solution to the traditional agency problem of managerial risk aversion. It could align a manager’s risky decisions with the interests of shareholders, increase firm value, and reduce the need for option-based compensation while still motivating an optimal level of managerial risk taking (Gervais, Heaton, and Odean, 2005). This argument also suggests that shareholders need to take into account the CEO’s confidence level, when the CEO compensation is designed. Compensating highly confident managers with stock awards and stock options, as if they were unbiased, however, has two serious drawbacks for shareholders. First, shareholder wealth may decrease because the managers are paid too much compared to an optimal contract that recognizes the bias toward overconfidence and adjusts the monetary risk incentives accordingly. In other words, monetary risk incentive may be redundant in the presence of overconfidence and, if awarded, represents an opportunity cost to shareholders. Second, driven by steep payoff convexity of, say, stock options, managers may ratchet up the riskiness of their investment projects beyond the level induced by their overconfidence. While the former cost arises from ignoring the substitutive effect of overconfidence the latter cost is due to ignoring its complementary effect. Neither cost is in the shareholders’ best interests. Both types of cost may be termed the agency cost of management’s behavioral bias. Gervais, Heaton, and Odean (2005) argue for flatter and less convex compensation elements to minimize this cost.
10.4.2
New role for corporate governance
According to Malmendier and Tate (2005a), standard incentives such as stockand option-based compensation are unlikely to mitigate the detrimental effects of managerial overconfidence. On the contrary, CEO confidence could be the product of a large executive compensation package. According to the tournament theory of executive compensation (Main, O’Reilly, and Wade 1993), relatively high CEO pay is a recognition that (s)he is a winner and thus may validate
240
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
the CEO’s self-image and feed his or her vanity, overconfidence, or hubris. A large executive compensation package gives positive feedback to a CEO and signals that the CEO is successful (Paredes, 2005). If so compensation and behavioral bias have a complementary relationship. Malmendier and Tate (2005b) call for more vigilant corporate control measures to alleviate the negative effect of CEO overconfidence. This calls for an enhanced role for corporate governance to reduce the agency costs of management bias. In addition to the traditional functions of the corporate governance mechanism, corporate governance should also account for human psychology by assessing the impact of behavioral bias, tailoring the compensation contract accordingly, moderating excessive or inadequate risk taking, and steering managers toward optimal risk taking. Paredes (2005) proposes strong corporate governance features as a way of controlling overconfident CEOs.17
10.4.3
A model of risk taking under compensation incentives, behavioral biases, and corporate governance
Summarizing our discussions regarding managerial risk drivers, we formulate the following conceptual model of managerial risk taking: Risk taking ⫽ f (Executive compensation, Managerial overconfidence, Corporate governance)
(10.1)
where Risk taking is the measure of riskiness of the investment or financing option; Executive compensation is the bundle of various compensation components and managerial ordinary share holdings; Corporate governance may include characteristics such as external block shareholding, proportion of nonexecutive directors, CEO and chairman-of-the-board (COB) duality, and the presence of a remuneration committee of the board. Managerial overconfidence captures the behavioral bias.
10.5
Joint impact of executive compensation, overconfidence, and governance on corporate acquisitions
Corporate acquisitions are major, externally observable, and discretionary longterm investments that can dramatically alter the risk profile of the acquiring firm and thereby exacerbate the potential risk-related conflict of interests between managers and shareholders. Dow and Raposo (2005) argue that the way executive compensation responds to changes in strategy can lead management to choose change that is more radical than the shareholders would optimally prefer, citing 17
Among the proposed features are a chief naysayer who will be the devil’s advocate against, say, the CEO’s investment strategy, tougher board fiduciary obligations and greater shareholder voice. Paredes (2005) also calls for greater self-education and self-awareness of CEOs i.e. metacognition.
Executive compensation and managerial overconfidence
241
mergers as “a clear example of dramatic change.” Within the genre of acquisitions, high-tech deals are likely to be even more dramatic. Diversifying acquisitions are thought to be driven by managerial preference for risk reduction (Amihud and Lev, 1981). On the other hand, acquisitions of targets rich in intangible assets, such as patents or R&D, obviously ratchet up the risk faced by the acquirers. Extensive evidence shows that acquirer shareholders do not gain from acquisitions in the short term and experience value losses in the longer term (see Weston, Mitchell, and Mulherin, 2004; Sudarsanam, 2003, for summary reviews). The last decade experienced the fifth merger wave in the United States (Sudarsanam, 2003). This wave also witnessed the greatest value destruction from acquisitions so far. Against an abnormal value loss of $4bn (over a 3-day event window) during 1980 to 1990 and an abnormal gain of $24bn from 1991 to 1997, the abnormal loss to acquirers during 1998–2001 amounted to a colossal $240bn, and the combined value loss to listed targets and acquirers was $134bn in the same period (Moeller, Schlingemann, and Stulz, 2005). This staggering loss points to the high-risk nature of acquisitions. Whether such value loss is due to skewed risk incentives that managerial compensation contracts provide is an interesting question to resolve empirically. We therefore consider acquisitions an appropriate corporate decision context in which to explore the relationships among compensation, investment risk profile, and shareholder value gains.
10.5.1
Does executive compensation drive acquisitions?
Furthermore, along with the previously noted explosion in CEO compensation especially stock option-based compensation, it is natural for us to speculate that the coexistence of the merger wave and explosion in option-based compensation in recent years is more than a simple coincidence. Figure 10.5 shows the trends of CEO equity-based compensation value and the mean transaction value of corporate acquisitions whose acquirers are S&P 500 firms from 1992 to 2004.18 The mean transaction value is in $ millions while the compensation value is in $ thousands. Table 10.1 reports the Spearman correlations between mean transaction value and incentives measures. All the correlations are significant. As Figure 10.5a indicates, the trend in mean transaction value is highly associated with that in the mean value of stock options grant estimated using the Black-Scholes model19, even if they are not perfectly matched. As indicated in Table 10.1, the correlation between transaction value and stock option value is 0.92. The mean value of restricted stock grant, as indicated in Figure 10.5b, however, does not follow the trend in the mean transaction value after 1997. Figure 10.5c demonstrates that Delta, the change in the dollar value of the CEO’s wealth for a 1% change in stock price, has a lead-lag effect upon merger wave. But Vega is highly correlated with the merger trend up to 2000 after which 18 19
All value is reported in 2004 U.S. dollar. We follow Guay (1999)’s methodology to estimate Delta and Vega of executive compensation package.
0 92 93 94 95 96 97 98 99 00 01 02 03 04 Year Transaction value
2500
1500
2000 1500
1000
1000
500
500
0
0 92 93 94 95 96 97 98 99 00 01 02 03 04 Year
Stock option
Transaction value
Restricted stock 600 500 400 300 200 100 0
1500
6000
1000
4000
500
2000 0 92 93 94 95 96 97 98 99 00 01 02 03 04 Year Transaction value
Delta
Transaction value ($ Million)
8000 Delta ($000s)
Transaction value ($ Million)
2000 2000
0
Restricted stock ($000s)
500
2000
1500 1000 500 0
92 93 94 95 96 97 98 99 00 01 02 03 04 Year Transaction value
Figure 10.5 Equity-based CEO compensation and mean M&A transaction value of S&P firms ($m0).
Vega
Vega ($000s)
1000
Transaction value ($ Million)
1500
Stock option value ($000s)
Transaction value ($ Million)
12,000 10,000 8000 6000 4000 2000 0
2000
[AU11]
Executive compensation and managerial overconfidence
243
Table 10.1 Correlation between mean M&A transaction value and incentives measures during 1992–2004 in the United States Compensation incentive
Mean transaction value
Stock Option Restricted Stock Vega Delta
0.92a (0.00) 0.52c (0.07) 0.83a (0.00) 0.63b (0.02)
a, b, c
Significance at 1%, 5%, and 10% levels, respectively (p-value in parentheses)
mean transaction value declines while Vega remains at a similar level. The correlation of mean acquisition value with stock option value (and Vega) is much stronger than with restricted stock value (and Delta). These figures lend credence to the view that the 1990s merger wave is at least partially caused by the use of stock options and driven by the high risk incentive they provided. Higher Vega seems to motivate more mega M&A deals.
10.5.2
Review of prior research
Empirical results concerning the incentive effect of executive compensation upon corporate acquisition performance are mixed. Datta, Iskandar-Datta, and Raman (2001) provide direct evidence that, for acquiring firms, providing stock option incentives to top executives can have a large positive impact on shareholder wealth. In the long run, managerial incentives can be effective in shaping long-term corporate investment policies and encourage managers to make decisions in the interests of shareholders. Using a sample of U.S. mergers, Williams and Rao (2006) document that relative risk, proxied by the ratio of post- to pre-merger stock return variance is positively related to the risk incentive effect of CEO stock option compensation. They also find that firm size has a moderating effect on such risk incentive effects.
10.5.3
Impact of acquisition on executive compensation
Many other studies examine the impact of corporate acquisitions on executive compensation. Various studies provide empirical evidence that executive pay rewards good M&A performance. Lambert and Larcker (1987) suggest that, on average, only executives selecting acquisitions that increase shareholder wealth experience a statistically significant increase in compensation and wealth. Khorana and Zenner (1998) reach similar results that good acquisitions increase compensation, while bad acquisitions have an insignificant effect on total compensation. Coakley and Iliopoulou (2006) provide U.K. evidence to support the managerial power hypothesis. Their empirical results indicate that less independent and larger boards award CEOs significantly higher bonuses and salary following M&A completion.
244
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Nevertheless, as mergers and acquisitions lead to increase in firm size, which is positively related to the size of CEO compensation, corporate acquisitions are normally followed by an increase in CEO compensation, regardless the acquirer’s performance (Grinstein and Hribar, 2004; Bliss and Rosen, 2001; Firth, 1991). Grinstein and Hribar (2004) find that measures of the performance are negatively related to the amount of the M&A bonus awarded to CEO. Their results indicate that M&A bonus is inversely related to M&A performance. More recently, Harford and Li (2005) observe, for the 1990s mergers, that acquirer CEOs are richly rewarded for growth through acquisitions with substantial new stock and option grants. CEO pay and wealth are completely insensitive to poor post-acquisition performance but CEO wealth increases with good performance. Consistent with our view that robust corporate governance can moderate inappropriate incentives to CEOs and enhance their sensitivity to performance (see Section 4 above), stronger boards maintain the sensitivity of CEO compensation to poor post-acquisition performance. Table 10.2 summarizes the studies regarding the impact of corporate acquisition performance on executive compensation and the impact of robust corporate governance. These studies suggest that in the context of acquisition executive compensation may be somewhat de-coupled from post-acquisition performance although there is also some evidence that stock-based components lead to value creating acquisitions and strong corporate monitoring enhances pay-toperformance sensitivity. The previous discussion of behavioral biases (see Section 3) suggests that CEO overconfidence may also have had considerable impact upon the 1990s merger wave in the United States. Following Malmendier and Tate (2005b), we use the timing of the CEO option exercise to identify CEO overconfidence. Under the standard assumption of managerial risk aversion and under-diversification of the manager’s wealth portfolio, CEOs are expected to choose an early exercise of their options. Under-diversified CEOs should also minimize their holding of company stock in order to reduce their exposure to firm specific risk. However, their overconfidence may induce managers to postpone the exercise of stock options in order to benefit from expected future gains. Holder67 captures this ‘irrational’ timing of CEO option exercise. Following Malmendier and Tate (2005b), we take 67% of in-the-money option value as the threshold. If the exercisable option is more than 67% in-the-money at the fiscal year end, a riskaverse CEO should have exercised part of the options he holds in his compensation package. We define the CEO as overconfident if he fails to exercise his stock option that is at least 67% in-the-money during the year prior to acquisition announcement. Figure 10.6 shows the number of acquisitions and the number of overconfident CEOs among S&P 500 firms from 1993 to 2004. We can tell from Figure 10.6 that the number of corporate acquisitions and the number of overconfident CEOs follow similar trends from 1993 to 2004. The similarity of trends between the 1990s merger wave and CEO compensation, especially in the form of stock options, in Figure 5 and between the merger wave and CEO overconfidence in Figure 6 warrants a closer look at the joint impact
Table 10.2 Studies of the impact of corporate acquisition on executive compensation Studies Lambert and Larcker (1987)
Sample Characteristics U.S. (1976–1980)
Major Results ●
●
Firth (1991) Khorana and Zenner (1998)
U.K. (1974–1980) U.S. (1982–1986)
●
●
●
●
●
Bliss and Rosen (2001)
U.S. (1986–1995)
●
●
●
Grinstein and Hribar (2004)
U.S. (1993–1999)
●
●
●
Harford and Li (2005)
U.S. (1993–2000)
●
●
●
Increases in executive compensation and wealth observed only if acquisition increases shareholder wealth. Difference in impact of value-increasing and value-decreasing acquisitions on executive compensation and wealth statistically significant. Acquisition leads to increase in managerial remuneration owing to increased firm size. Ex ante compensation-to-size sensitivity makes large acquisition more likely. Ex post, large acquisitions have a small positive effect on total compensation Good acquisitions increase compensation, whereas bad acquisitions do not reduce compensation Compensation Mergers increase compensation, mainly owing to firm-size effect. Compensation increases even if mergers cause acquiring bank’s stock price to decline. CEOs make fewer wealth-reducing mergers when they own more stock. Managerial power explains a large part of the variation in acquisition-related bonus. Bonuses larger with larger deals. Deal announcement performance unrelated to cross-sectional variation in compensation. CEO’s total pay and overall wealth increase substantially following an acquisition. In poorly governed acquirers, CEO’s pay following merger insensitive to performance. CEO’s wealth increases even if he or she makes a poor acquisition.
246
International M&A Activity Since 1990: Recent Research and Quantitative Analysis 600
120
500
100
400
80
300
60
200
40
100
20
0
0 93
94
95
96
Number of deals
97
98
99
00
01
02
03
04
Number of overconfident CEOs
Figure 10.6 Number of corporate acquisitions and number of overconfident CEOs of S&P 500 firms.
of CEO compensation and overconfidence upon corporate acquisition decisions and their shareholder value outcome. Empirical analyzes in Malmendier and Tate (2005a) suggest that overconfident CEOs are, on average, more acquisitive. They find that the odds of an overconfident manager20 making a successful takeover bid are 1.68 times those of a rational manager. According to our data behind Figure 6, the correlation between the number of overconfident managers and the number of M&A deals is 0.84 which is significant at 1% level. This macro level association broadly conforms to the Malmendier and Tate result. Considerable research has been done to understand how CEO overconfidence impacts upon the quality of corporate acquisition decisions. Roll (1986) formulates a hubris hypothesis that is further developed and tested by Hayward and Hambrick (1997). When managers consider taking over another firm, hubris causes managers to underestimate the risks inherent in M&As, leading to the overvaluation of the target. Hubris-infected managers may pay an excessive takeover premium, and consequently destroy value for their shareholders. Although subtle differences exist among these three behavioral biases, they all induce managerial excessive risk taking.21 Therefore, we use the term overconfidence to refer 20
The overconfidence measure using in Malmendier and Tate (2005a) is named Longholder, a binary variable where value of 1 indicates that the CEO at some point during her/his tenure held an option package until the last year before expiration. 21 It may be argued that overoptimism is about one’s excessively sanguine view of future states of nature or environment and overconfidence is about confidence in one’s own ability to manage risk and control adverse outcomes. Hubris is closer to overconfidence than to overoptimism. In the literature there is much ambiguity in the use of these terms or in finding well calibrated proxies to differentiate them. Paredes (2005) and Shefrin (2005) reflect this ambiguity. Kahenman and Lovallo (1993) also argue that optimism may lead to misjudging the odds of success thus linking optimism to risk.
Executive compensation and managerial overconfidence
247
to their common character, i.e., they cause managers to underestimate project risk or overvalue a project and encourage managerial risk taking. Excited by the challenges and aware of the substantial benefits potentially flowing from high-tech acquisitions, overconfident managers may be particularly attracted to such acquisitions because they have the opportunity to demonstrate their capability in ‘creating miracles’ (Kohers and Kohers, 2001).22 Many high-tech acquisitions, during the dotcom bubble of the 1990s, may have been driven by top managers’ optimism or arrogant self-belief, compounding the valuation risk inherent in high-tech acquisitions, leading to overpayment for targets and value losses to acquirer shareholders.23 The disastrous merger of AOL and Time Warner (TW) in 2001, one of the deals from hell as described by Bruner (2005, ch.12), was crafted on the basis of exuberant optimism about internet broadband as a transformative technology and visionary faith in its potential.24 Most of the available empirical evidence points to the negative value impact of different behavioral biases. Using a sample of large acquisitions, Hayward and Hambrick (1997) find that hubris is significantly negatively related to oneyear post-acquisition returns. Hletala, Kaplan, and Robinson (2003) analyze Viacom’s takeover contest for Paramount in 1994 and trace hubris in Viacom’s CEO. They argue that the great success Summer Redstone, Viacom CEO, and the company had enjoyed prior to that acquisition had fostered his hubris that led to the overpayment for Paramount by over $2bn. Viacom performs poorly relative to the S&P 500 index and its three primary competitors after the acquisition. Malmendier and Tate (2005a) find that overconfident CEOs are more likely to conduct mergers, in particular, value-destroying mergers, than are rational ones. They classify a CEO as overconfident if and only if the CEO at some point during her/his tenure held an option package until the last year before expiration. Rau
22
Psychic payoffs from taking risks (Paredes, 2005) and the animal spirits roused by the thrill of the chase may also lead managers to excessively risky acquisitions. 23 The case of Vivendi illustrates such adventurous tendencies (Johnson and Orange, 2003). Jean-Marie Messier transformed a water company into an international high-tech conglomerate through successive acquisitions. He was granted the title of the ‘perfect Frenchman’ by the French media. However, after the telecom bubble burst in 2000, Vivendi fell into substantial financial difficulties. Jean-Marie Messier was sacked and convicted of fraud. “Without his (Messier’s) vision and personality – a strange blend of French technocratic arrogance, wannabe Hollywood showmanship and investment banker charm – Vivendi Universal would never have come into existence. Without Jean-Marie Messier’s weakness – a love of deal-making, self-promotion, obfuscation and risk – the dream of a French champion might have survived” (Johnson and Orange, 2003, p. 3). 24 A respected portfolio manager, Gordon Crawford, reflecting on the AOL-TW merger said of Jerry Levin, the CEO of TW and one of the prime movers behind the deal: “This was a breathtaking deal. Everyone was surprised by it. But it turned out to be the wrong company at the wrong time and at the wrong price. The crux of the issue here is Levin’s ego. … Levin got caught up in his self-image as a leading thinker about transformative technology” (Bruner, 2005, p278) (italics added). Crawford does not fault Levin’s vision about the internet but the timing of the deal, the choice of the merger partner and the price paid to realize the vision.
248
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
and Vermaelen (1998), Kohers and Kohers (2001), and Sudarsanam and Mahate (2003) suggest that glamour stock rating, a plausible proxy for overconfidence, leads to a risky acquisition strategy, excessive acquisition premium and value destruction for acquirer shareholders. Table 10.3 summarizes the abnormal returns to acquirer shareholders reported by these papers in both the announcement periods and in the long term and by Doukas and Petmezas (2006). Most of these report negative returns to acquirer shareholders in the short term and four of them investigating long run returns find value destruction.
10.5.4
Problem of endogeneity
In the study of the pay-for-performance relation, it is possible that, rather than higher equity-based compensation motivating better future performance, firms expecting better future performance grant more equity (Yermack, 1997). Specifically, even though managerial compensation may affect the performance of corporate acquisitions (see Review of prior research in this section above), the contemporaneous or anticipated change in corporate stock performance due to the acquisition may affect both the structure of managerial compensation and the firm’s compensation policy choices. This sort of endogenous relation also applies to the relation between managerial incentives and acquisition-related risk change. Brick, Palia and Wang (2005) study the endogeneity among three corporate governance mechanisms, namely executive compensation, firm leverage, Table 10.3 Impact of overconfidence on acquirer shareholder value Study
Sample characteristics
Acquirer’s share performance (%)
Proxy for overconfidence
Hayward and Hambrick (1997)
U.S. (1989–1992)
⫺4 (11 days) ⫺11 (1 year)
Past performance Media appraisal CEO self-importance
Rau and Vermaelen (1998)
U.S. (1980–1991)
⫺4.04 (3 years)
Book-to-market ratio
Kohers and Kohers (2001)
U.S. (1984–1995)
0.92a (2 day) ⫺18.68a (3 years)
Book-to-market ratio
Sudarsanam and Mahate (2003)
U.K. (1983–1995)
⫺1.39a (2 days) ⫺8.71 (3 years)
Book-to-market ratio
Doukas and Petmezas (2006)
U.K. (1980–2004)
1a (5 days)
Serial acquisition
Malmendier and Tate (2005a)
U.S. (1980–1994)
⫺0.5 (3 days)
Longholder Media appraisal
a, b, c
Significance at 1%, 5 %, and 10% levels, respectively. Note: Except for Sudarsanam and Mahate, who employ, buy, and hold abnormal returns (BHARs), all studies use cumulative abnormal returns to measure shareholder value gains.
Executive compensation and managerial overconfidence
249
and independent board of directors. They subsequently link these three corporate governance mechanisms to shareholder value.
10.5.5
A new integrated framework
According to our survey so far, the extant empirical studies that examine the impact of executive compensation or CEO overconfidence on risk taking and firm performance have not incorporated an integrated framework encompassing both risk and performance and do not explicitly allow for the moderating influence of corporate governance. We find that: ●
●
●
●
●
theoretical models of the substitutive and complementary aspects of the relation between monetary incentives and behavioral drivers in the context of corporate decision makers need to be developed further; prior empirical studies examine the impact of executive compensation on either risk taking or corporate performance and not both simultaneously; prior empirical studies examine the impact of behavioral biases on either risk taking or corporate performance and not both simultaneously; the interactive effects of executive compensation and behavioral biases such as overconfidence on risk taking and performance are not fully explored; and although the somewhat crude proxies for risk incentives provided by compensation contracts in early studies have been replaced in recent studies by more sophisticated measures such as Delta and Vega, other plausible proxies need to be explored.
We, therefore, propose in Figure 10.7 an integrated framework to consider the joint impact of executive compensation, behavioral biases and corporate governance on risk taking behavior and the consequent firm performance.
Executive compensation
Managerial overconfidence
Managerial risk taking
Corporate governance
Investment decisions
Firm risk
Firm performance
Figure 10.7 Impact of executive compensation, behavioral bias, and corporate governance on firm risk and performance.
250
10.6
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Review of recent empirical studies based on the integrated framework
In this section we review the emerging set of empirical studies that have examined some or all of the linkages set out in Figure 10.7. Broadly the hypotheses tested in these studies can be classified into two sets. The first set of hypotheses concerns the impact of executive compensation and managerial overconfidence on risk taking in corporate acquisitions. The second set of hypotheses focuses on the direct or indirect value creation effect of both executive compensation and CEO overconfidence. Indirect estimation of value creation measures the impact of executive compensation and managerial overconfidence on acquisition risk at the first stage and then the impact of the risk level on shareholder value performance at the second stage. Gao and Sudarsanam (LS) (2006) adopt a 2-stage approach. Using a sample of 578 acquisitions of both high-technology targets (proxy for high risk) and low-technology targets (proxy for low risk) in the United Kingdom during the 1990s, they build an empirical acquisition risk model including executive compensation, directors’ overconfidence and corporate governance mechanisms as risk determinants to identify the factors that influence managerial risk taking. This is the empirical version of the conceptual model in Equation 10.1 above and, additionally, includes control variables. They include different components of executive compensation e.g. value of long term incentive plan (LTIP) stock grants and stock options as well the more sophisticated Delta and Vega measures. Proxies for overconfidence include past acquirer stock returns, glamour rating of the acquirer stock and the media profile of the directors in the three years preceding the acquisition. GS report that fixed pay, directors’ wealth and Delta decrease the risk propensity whereas past returns and media praise induce more risk taking. This model is then used to classify the sample acquisitions into ‘over-risk’, ‘under-risk’ and ‘optimal-risk’ acquisitions. LS subsequently examine the 3-year post-acquisition abnormal returns to acquirer shareholders for the three risk levels, using a four factor abnormal returns model.25 They find that high risk acquisitions outperform low risk acquisitions. More interestingly, ‘over-risk’ acquisitions perform as well as ‘optimal-risk’ acquisitions but outperform ‘underrisk’ acquisitions in terms of shareholder value creation.26 These findings challenge the view that overconfidence or monetary risk incentives invariably lead to negative performance outcomes but are consistent with the model of Gervais Heaton, and Odean (2005) and others that posit the beneficial effects of over confidence (see Benefits of overconfidence in Section 3). Sudarsanam and Huang (SH) (2006) adopt a slightly different methodology by examining the impact of executive compensation and CEO overconfidence
25 26
The four factors are: industry, size, book to market and momentum. A direct evaluation of the impact of risk determinants on 3-year abnormal returns yields much weaker but consistent results.
Executive compensation and managerial overconfidence
251
on acquisition risk proxied by the change the acquirer’s stock return volatility resulting from the acquisition. They also examine their impact on 3 year abnormal return performance. Based on a sample of 3007 acquisitions in the United States during the period January 1, 1993 to December 31, 2004, SH find that Vega has a positive impact on acquisition risk and so does managerial overconfidence proxied by Hold67 described in Section 5 above. SH also model the relations among risk incentives Delta and Vega, performance and risk change as endogenous variables within a simultaneous equations system but the results are unchanged. The results of these two studies are summarized in Table 10.4.
10.6.1
Implications of the results
The results from the GS and SH studies suggest that both executive compensation and CEO overconfidence have considerable impact upon acquirer’s risk choices and post-acquisition performance. Thus any study of the effect of executive compensation on managers’ risk preferences should control for the behavioral biases of such managers. Further the interactive effect of high compensation and high overconfidence also merits empirical attention. The SH result indicates that monetary risk incentives like Vega and managerial overconfidence both encourage riskier acquisition choices and thus seem to interact in a complementary and mutually reinforcing way. In Gao and Sudarsanam, however, Vega has little effect but overconfidence encourages more risky acquisitions. The contrasting impact of Vega in the two studies—one based on U.S. data and the other on U.K. data—is perhaps due to the relatively low level of stock options in the total compensation of U.K. managers (Conyon and Murphy, 2000). Thus the risk incentive of stock options in the United Kingdom is weak. In comparison, in both countries, overconfidence does lead to higher acquisition risk. The U.K. results support the view that managerial overconfidence may be a substitute for monetary risk incentives embedded in stock options. Gao and Sudarsanam also report that at higher levels of their wealth managers’ appetite for risk declines. This emphasizes the need to control for managerial wealth in modeling the impact of monetary risk incentives on managerial risk choices. This is in line with the arguments of Ross (2004) and Carpenter (2000) (see Section 2.5 above).
10.7
Unresolved theoretical and empirical issues
There are several issues of both a conceptual and empirical nature that remain to be resolved before we can reach a good understanding of managerial risk taking and its impact on corporate value creation through various financing and investment decisions. Among the conceptual issues are: ●
the absence of a theoretical model of managerial risk taking under the influence of the three factors we have identified in Figure 10.7, namely, compensation contracts, behavioral biases and corporate governance. Such a model will yield predictions
Table 10.4 Empirical studies based on the integrated framework (cf. Figure 10.7) Study
Focus variable/s
Proxies for the focus variable
Sample characteristics
Gao and Sudarsanam (2006)
Executive compensation & CEO overconfidence
CEO overconfidence: media appraisal past performance acquirer Glamour status Executive Compensation: Vega and Delta
U.K., 578 M&A deals (1993–2000)
CEO overconfidence: holder 67 past performance acquirer glamour status Executive compensation: Vega and Delta
U.S., 3007 M&A deals, (1993– 2004)
Main results
●
●
●
●
●
●
●
Sudarsanam and Huang (2006)
Executive compensation & CEO overconfidence
●
●
●
●
●
●
●
●
Equity-based compensation and managerial shareholding have a concave incentive effect on managers’ acquisition risk choice. A high level of stock-based managerial wealth induces managerial risk aversion and weakens risk alignment between shareholders and managers. Managers’ overconfidence induces managers to engage in risky high-tech acquisitions. Over-risk and optimal risk acquisitions perform better than under-risk acquisitions in terms of shareholder value. Increase in Vega is associated with enhancement in acquirer shareholder value while increase Delta is related to reduction in shareholder value. Increase in risk change is associated with increase in Vega; Increase in risk change is associated with managerial overconfidence. Managerial overconfidence provides an alternative solution to the underinvestment problem stemming from managerial risk aversion.
Executive compensation and managerial overconfidence
●
●
253
about the optimal level of risk taking. Whether the three or any a subset of these factors are endogenously determined also needs to be explored by such a model; the absence of a theoretical model that relates managerial risk taking to their wealth (Ross, 2004) and incorporates the implications of the prospect theory (Kahneman and Tversky, 1979); and the absence of a model relating the optimality or suboptimality of risk taking to corporate value change.
There are also significant empirical issues to be addressed. These are: ●
●
●
●
developing a range of credible proxies for the psychological constructs such as overconfidence, overoptimism or hubris. Our review reveals a few of the proxies used for overconfidence e.g. Holder 67, celebrity status, media praise but other measures are possible and a composite index of overconfidence will enhance the reliability of the results. Similarly, we need good proxies for other behavioral bias constructs such as overoptimism or loss aversion; proxies that can reliably differentiate between confidence and overconfidence; in the absence of a theory specifying the relevant observable variables that determine optimal risk choices, the empirical risk model may be subject to an omitted-variable problem; data on managers’ nonfirm-related wealth that will allow an analysis of how their risk preferences change as a function of that wealth; and the problem of endogeneity among monetary incentives, risk taking and performance.
One implication of Carpenter (2000) and Ross (2004) is that the examination of managerial incentives effect should control for CEO wealth level. However, it is difficult to measure the CEO’s non firm-related wealth and recent studies so far use CEO firm specific wealth as a proxy for CEO total wealth (Gao and Sudarsanam, 2005).The investigation of CEO overconfidence relies heavily on the measure of overconfidence chosen. Since there is a thin line between selfconfidence and its extreme case—overconfidence, it is critical to have a sufficiently accurate measure of CEO overconfidence. However, most of the proxies for this bias are noisy, which may render the empirical results weak.
10.8
Conclusion
This chapter examines an area of the traditional agency model which hitherto has been relatively underexplored both theoretically and empirically, namely, the divergence of risk preferences between shareholders and managers. The traditional agency model views managers holding undiversified portfolios with almost all of their money and human invested in their employer firm as risk averse but shareholders, being able to diversify their investments, as risk neutral. One solution to this risk-related agency conflict is to structure the managers’ compensation in such a way that it provides risk incentives. Executive stock option is a device to achieve this aim. This model rests on the assumption of rational and utility maximising managers. We argue that an executive compensation contract is a heterogeneous mix of components with different risk incentives, some risk inducing
254
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
and some reinforcing managerial risk aversion and that we need a good understanding of the impact of these components on risk taking before modeling their impact on performance. In our view risk is a critical variable that links compensation and performance. We also consider the behavioral agency model that posits managers’ behavioral biases as influencing their risk preferences irrespective of the risk incentives offered by compensation contracts. Various biases identified in the behavioral finance literature are explored but we focus on one particular psychological trait i.e. overconfidence that has been the subject of some recent theoretical and empirical work. We extend this work by examining the joint impact of risk incentives from compensation contracts and overconfidence. We review the anecdotal as well as empirical evidence that suggests that both monetary risk incentives and overconfidence can be complementary and lead to excessive risk taking. Avoiding such excessive risk taking or, conversely, inadequate risk taking by risk averse managers calls for appropriate monitoring of managers’ risk incentives and their behavioral biases. We argue that corporate governance has to assume this new monitoring role. We present an integrated model in which risk taking and consequent firm performance are subject to the interacting influences of executive compensation structure, behavioral bias and corporate governance. We present the results from our two recent studies in which this joint impact is empirically tested. The first, based on a U.K. sample of high and low risk acquisitions, finds that certain compensation components like salary and bonus and, at lower levels, stock grants induce risk aversion. Overconfidence leads to more risky acquisitions. The study does not find that corporate governance has any moderating effect on risk taking. This study reports that over-risk acquisitions outperform under-risk acquisitions in terms of three shareholder value gains, consistent with the beneficial impact of overconfidence propounded by some authors. In the second study of U.S. acquisitions from 1993 to 2004, we find evidence that stock options lead to increased risk taking and overconfidence also leads to more risky acquisitions. These results highlight the need to consider both the substitutive and complementary natures of the relation between compensation-based risk incentives and behavioral proclivity toward higher risk caused by overconfidence. Regarding possible future research, the interaction between executive compensation and managerial overconfidence and their impact on corporate acquisition decisions are yet to be fully explored. The intermediate effect of other corporate governance mechanisms, which could be critical in alleviating the negative impact of managerial overconfidence and the agency problem of executive compensation, remains a promising research agenda for both theoretical and empirical study. The measurement of managerial overconfidence also remains an important empirical challenge to be overcome. More accurate proxies are needed to capture overconfidence and other behavioral biases like overoptimism or hubris which, in this study, have been assumed to be the same. More fine-tuning of empirical proxies to reflect the distinction among these is necessary. Other behavioral biases e.g. loss aversion, self-attribution also need to be explored for
Executive compensation and managerial overconfidence
255
a better understanding managerial risk taking and firm performance. The issues raised in this chapter could also be explored in the context of other corporate investment and financing decisions. We identify several limitations in terms of theory developments and empirical methodologies that need to be addressed in future research before a full understanding of managerial risk behavior and its impact on firm performance can be attained. We offer some suggestions for future research.
References Amihud, Y., and Lev, B. (1981). Risk Reduction as a Managerial Motive for Conglomerate Mergers. Bell Journal of Economics, 12:605–617. Baker, G., Jensen, M., and Murphy, K. (1988). Compensation and incentives: practice vs. theory. Journal of Finance, 43:593–616. Barberis, N., and Thaler, R. (2002). A Survey of Behavioral Finance. NBER Working Paper No. 9222. Bebchuk, L., and Fried, J. (2004) Pay without performance, Harvard University Press, US Bénabou, R., and Tirole, J. (2003). Self-Confidence and Personal Motivation. Quarterly Journal of Economics, 3:871–915. Berle, A. A., and Means, G. C. (1932). The Modern Corporation and Private Property. New York: Harcourt, Brace & World. Bertrand, M., and Mullainathan, S. (2001). Are CEOs Rewarded for Luck? The Ones without Principals Are. Quarterly Journal of Economics, 3:901–932. Bhagat, S., and Black, B. (2002). The Non-correlation between Board Independence and Long-Term Firm Performance. The Journal of Corporation Law, 4:232–273. Bliss, R., and Rosen, R. (2001). CEO Compensation and Bank Mergers. Journal of Financial Economics, 61(1):107. Bryan, S., Hwang, L., and Lilien, S. (2000). CEO Stock-Based Compensation: An Empirical Analysis of Incentive-Intensity, Relative Mix and Economic Determinants. Journal of Business, 73:661–93. Bruner, R. (2005). Deals from Hell: M&A Lessons that Rise above the Ashes. John Wiley & Sons. Bryan, S., Hwang, L. S., and Lilien, S. (2000). CEO Stock-Based Compensation: An Empirical Analysis of Incentive-Intensity, Relative Mix, and Economic Determinants. Journal of Business, 73:661–693. Carpenter, J. N. (2000). Does Option Compensation Increase Managerial Risk Appetite? Journal of Finance, 5:2311–2331. Chatterjee, A., and Hambrick, D. (2006). It’s All about Me; Narcissistic CEOs and Their Effects on Company Strategy and Performance. Working Paper. Pensylvania State University. Coakley, J., and Iliopoulou, S. (2006). Bidder CEO and Other Executive Compensation in UK M&As. European Financial Management, 12(4):609–631.
256
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Conyon, M. J., and Murphy, K. J. (2000). The Prince and the Pauper? CEO Pay in the United States and in the United Kingdom. The Economic Journal, 110:F640–671. Coles, J. L., Daniel, N. D., and Naveen, L. (2006). Managerial Incentives and Risk-Taking. Journal of Financial Economics, 79(2):431–468. Core, J., and Guay, W. (1999). The Use of Equity Grants to Manage Optimal Equity Incentive Levels. Journal of Accounting Economics, 28:151–184. Core, J. E., Holthausen, R. W, and Larcker, D. F. (1999). Corporate Governance, Chief Executive Officer Compensation and Firm Performance. Journal of Financial Economics, 51:371–406. Core, J. E., Guay, W. R., and Larcker, D. F. (2003). Executive Equity Compensation and Incentives: A Survey. Economic Policy Review (Federal Reserve Bank of New York), 9(1):27. Datta, S., Iskandar-Datta, M., and Raman, K. (2001). Executive Compensation and Corporate Acquisition Decisions. The Journal of Finance, 56(6):2299. DeFusco, R. A., Johnson, R. R., and Zorn, T. S. (1991). The Association between Executive Stock Option Plan Changes and Managerial Decision Making. Financial Management, 20:40. Dow, J., and Raposo, C. C. (2005). CEO Compensation, Change and Corporate Strategy, Journal of Finance, 60(6):2701–28. Doukas, J. A., and Petmezas, D. (2006). Acquisitions, Overconfident Managers and Self–Attribution Bias. Working Paper. Old Dominion University. Elson, C., (2003). What’s Wrong with Executive Compensation? Harvard Business Review, January 81(1):68–77. Faccio, M., and Lasfer, M. A. (2000). Do Occupational Pension Funds Monitor Companies in which They Hold Large Stakes? Journal of Corporate Finance, 6:71–110. Fama, E. F., and Jenson, M. (1983). Separation of Ownership and Control, Journal of Law and Economics 26:301–325. Feltham, G. A., and Wu, M. G. H. (2001). Incentive Efficiency of Stock versus Options. Review of Accounting Studies, 6:7–28. Franks, J., Mayer, C., and Renneboog, L. (2001). Who Disciplines Management in Poorly Performing Companies? Journal of Financial Intermediation, 10:209–248. Gao, L., and Sudarsanam, P. S. (2006). Executive Compensation, Overconfidence and Corporate Governance: Impact on Managerial Risk Taking and Value Creation in U.K. High-Tech and Low-Tech Acquisitions. Cranfield School of Management Working Paper (Earlier Versions Presented to the European Financial Management Association Annual Meeting, Basle, 2004 and the Financial Management Association Annual Meeting, New Orleans, 2004). Gervais, S., and Goldstein, I. (2004). Overconfidence and Team Coordination. Working Paper. Rodney White Centre for Financial Research. Gervais, S., Heaton, J. B., and Odean, T. (2005). Overconfidence, Investment Policy and Manager Welfare. Working Paper. Duke University.
Executive compensation and managerial overconfidence
257
Goel, A. M., and Thakor, A. V. (2005). Overconfidence, CEO Selection, and Corporate Governance. AFA 2007 Chicago Meetings. Paper.Http://Ssrn. Com/Abstract=890274. Grinstein, Y., and Hribar, P. (2004). CEO Compensation and Incentives: Evidence from M&A Bonuses. Journal of Financial Economics, 73(1):119–143. Guay, W. R. (1999). The Sensitivity of CEO Wealth to Equity Risk: An Analysis of the Magnitude and Determinants. Journal of Financial Economics, 53:43–71. Habib, M., and Ljungqvist, A. (2005). Firm Value and Managerial Incentives: A Stochastic Frontier Approach. Journal of Business, 78(6):2053–2093. Hall, B.J., and Murphy, K.J. (2002). Stock Options for Undiversified Executives. Journal of Accounting and Economics, 33(1):3–42. Harford, J., and Li, K. (2005). Decoupling CEO Wealth and Firm Performance: The Case of Acquiring CEOs. Working Paper. University of Washington Business School. Hayward, M. L. A., and Hambrick, D. C. (1997). Explaining the Premiums Paid for Large Acquisitions: Evidence of CEO Hubris. Administrative Science Quarterly, 42: 103–127. Heaton, J. B. (2002). Managerial Optimism and Corporate Finance. Financial Management, 31:33–45. Hermalin, B. E., and Weisbach, M. S., (1991). The Effects of Board Composition and Direct Incentives on Firm Performance, Financial Management, 20:101–112. Hermalin, B. E., and Weisbach, M. S., (1998). Endogenously Chosen Boards of Directors and Their Monitoring of the CEO. American Economic Review, 88:96–118. Hermalin, B. E. (2005). Trends in Corporate Governance. Journal of Finance, 60(5):2351–2384. Hletala, P., Kaplan, S. N., and Robinson, D. T. (2003). What Is the Price of Hubris? Using Takeover Battles to Infer Overpayment and Synergies. Financial Management, 32:5–31. Ittner, C. D., Lambert, R. A., and Larcker, D. F. (2003). The Structure and Performance Consequences of Equity Grants to Employees of New Economy Firms. Journal of Accounting and Economics, 34:89–127. Jensen, M.C., and Meckling, W. H. (1976). Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics, 3:305–360. Jensen, M., and Murphy, K. J. (1990). Performance Pay and Top-Management Incentives. Journal of Political Economy, 98:225–264. Jensen, M., and Murphy, K. J. (2004). Remuneration: Where We’ve Been, How We Got to Here, What Are the Problems and How to Fix Them. Finance Working Paper No. 44/2004. European Corporate Governance Institute. John, K., Litov, L., and Yeung, B. (2006). Corporate Governance and Corporate Risk Taking: Theory and Evidence. Working Paper. New York University.
258
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Johnson, J., and Orange, M. (2003). The man who tried to buy the world, Jean-Marie Messier and Vivendi Universal, Viking. Kahneman, D., and Lovallo, D. (1993). Timid Choices and Bold Forecast: A Cognitive Perspective on Risk Taking. Management Science, 39:17–31. Kahneman, D., and Tversky, A. (1979). Prospect theory: An analysis of decision under risk, Econometrica, Vol 47, 263–291. Khorana, A., and Zenner, M. (1998). Executive Compensation of Large Acquirers in the 1980’s. Journal of Corporate Finance, 4:209–240. Kirkland, R., and Burke, D. (2006). The Real CEO Pay Problem. Fortune, 154(1):78–86. Knowledge@Wharton. Corporate Fraud on Trail: What Have We Learned. March 30, 2006. Kohers, N., and Kohers, T. (2001). Takeovers of Technology Firms: Expectations vs. Reality. Financial Management, 30:35–54. Lambert, R. A., and Larcker, D. F. (1987). Executive Compensation Effects of Large Corporate Acquisitions. Journal of Accounting and Public Policy, 6:231–243. Lambert, R., Larcker, D., and Verrecchia, R. (1991) Portfolio considerations in valuing managerial compensation, Journal of Accounting Reseach, 29(1):129–149. Malmendier, U., and Tate, G. (2005a). Who Makes Acquisitions? CEO Overconfidence and the Market’s Reaction. Stanford Research Paper 1798. Malmendier, U., and Tate, G. (2005b). CEO Overconfidence and Corporate Investment. Journal of Finance, 60:2661–2700. March, J. G., and Shapira, Z. (1987). Managerial Perspective on Risk and Risk Taking. Management Science, 33:1404–1418. Main, B. G., O’Reilly III, C. A., and Wade, J. (1993). Top Executive Pay: Tournament or Teamwork. Journal of Labour Economics, 11:606. McConnell, J. J., and Servaes, H. (1990). Additional Evidence on Equity Ownership and Corporate Value. Journal of Financial Economics, 27:595–612. Mehran, H, (1995). Executive Compensation Structure, Ownership and Firm Performance. Journal of Financial Economics, 38:163–184. Meulbroek, L. K. (2001). The Efficiency of Equity-Linked Compensation: Understanding the Full Cost of Awarding Executive Stock Options. Financial Management, 30(2):5–44. Moeller, S. B., Schlingemann, F. P., and Stulz, R. M. (2005). Wealth Destruction on a Massive Scale: A Study of Acquiring Firm Returns in the Merger Wave of the Late 1990’s. Journal of Finance, 60:757–82 Morck, R., Shleifer, A., and Vishny, R. W. (1988). Management Ownership and Corporate Performance: An Empirical Analysis. Journal of Financial Economics, 20:293–316. Murphy, K. (1986). Top Executives Are Worth Every Nickel They Get. Harvard Business Review, March–April. Palia, D. (2001). The Endogeneity of Managerial Compensation in Firm Valuation: A Solution. The Review of Financial Studies, 14(3):735–764.
Executive compensation and managerial overconfidence
259
Paredes, T. A. (2005). Too Much Pay, Too Much Deference: Behavioral Corporate Finance, CEOs and Corporate Governance. Florida State University Law Review, 32:673–762. Parrino, R., Poteshman, A. M., and Weisbach, M. S. (2005). Measuring Investment Distortions When Risk Averse Managers Decide Whether to Undertake Risky Projects. Financial Management, 34(1):21–60. Rajgopal, S., and Shevlin, T. (2002). Empirical Evidence on the Relation between Stock Option Compensation and Risk Taking. Journal of Accounting and Economics, 33:145–171. Rau, P. R., and Vermaelen, T. (1998). Glamour, Value and the Post-Acquisition Performance of Acquiring Firms. Journal of Financial Economics, 49:223–253. Rogers, D. A. (2005). Managerial Risk-Taking Incentives and Executive Stock Option Repricing: A Study of U.S. Casino Executives, Financial Management, Vol. 34(1), Spring. Roll, R. (1986). The Hubris Hypothesis of Corporate Takeovers. Journal of Business, 59:197–216. Ross, S. A. (2004). Compensation, Incentives, and the Duality of Risk Aversion and Riskiness. Journal of Finance, 59:207–225. Rovenpor, J. L. (1993). The Relationship between Four Personal Characteristics of Chief Executive officers and Company Merger and Acquisition Activity. Journal of Business and Psychology, 8(1):27–55. Ryan, H. E., and Wiggins, R. A. (2002). The Interactions between R&D Investment Decisions and Compensation Policy. Financial Management, 31(1):5–29. Sesil, J., Kroumova, D. K., and Blasi, J. (2000). Broad-Based Employee Stock Options in the U.S.: Company Performance and Characteristics. Rutgers University. Shefrin, H. (2005). Behavioral Corporate Finance, Mcgraw-Hill/Irwin. Shleifer, A., and Vishny, R. W. (1986). Large Shareholders and Corporate Control, Journal of Political Economy, 94:461–488. Skeel, D. (2005). Icarus in the Boardroom: The Fundamental Flaws in Corporate America and Where They Came from. Oxford University Press, U.S.A. Smith, C., and Stulz, R. (1985). The Determinants of Firms’ Hedging Policies. Journal of Financial & Quantitative Analysis, 20:391–405. Stathopoulos, K., Espenlaub, S., and Walker, M. (2004). UK executive compensation practices: New economy vs. old economy, Journal of Management Accounting Research, Vol. 16 pp 57–92. Sudarsanam, S. (2000). Corporate Governance, Corporate Control and Takeovers, Advances in Mergers and Acquisitions, Volume 1, pp. 119–155. Sudarsanam, P. S. (2003). Creating Value from Mergers and Acquisitions: The Challenges. London: FT Prentice Hall. Sudarsanam, P. S., P., Holl, and Salami, A. (1996). Shareholder Wealth Gains in Mergers: Effect of Synergy and Ownership Structure. Journal of Business Finance and Accounting 23:673–698.
260
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Sudarsanam, P. S., and Huang, J. (2006). Managerial Incentives, Overconfidence, Risk-Taking, and Acquirer Shareholder Value Creation in Mergers and Acquisitions. European Financial Management Association Annual Meeting, Madrid. Sudarsanam, P. S., and Mahate, A. A. (2003). Glamour Acquirers, Method of Payment and Post-Acquisition Performance: The U.K. Evidence. Journal of Business Finance & Accounting, 30:299–341. Tufano, P. (1996). Who Manages Risk? An Empirical Examination of Risk Management Practices in the Gold Mining industry. Journal of Finance, 51:1097–1137. Weir, C., and Laing, D. (2000). The Performance–Governance Relationship: The Effects of Cadbury Compliance on U.K. Quoted Companies. Journal of Management and Governance, 4:265–281. Weir, C., Laing, D., and Mcknight, P. J. (2002). Internal and External Governance Mechanisms: Their Impact on the Performance of Large U.K. Public Companies. Journal of Business Finance & Accounting, 29:579–611. Welch, J. (2001). Jack, What I’ve Learned Leading a Great Company and Great People. Headline, London. Weston, J. F., Mitchell, M. L., and Mulherin, J. H. (2004). Takeovers, Restructuring and Corporate Governance. New York: Pearson–Prentice Hall. Williams, M. A., and Rao, R. P. (2006). CEO Stock Options and Equity Risk Incentives. Journal of Business Finance & Accounting, 33(1&2):26–44. Wiseman, R. M., and Gomez-Mejia, L. R (1998). A Behavioral Agency Model of Managerial Risk Taking. Academy of Management Review 23:133–153. Wright, P., Ferris, S. P., Sarin, A., and Awasthi, V. (1996). Impact of Corporate Insider, Blockholder, and Institutional Equity Ownership on Firm Risk Taking. Academy of Management Journal, 39:441–463. Yermack, D. (1996). Higher Market Valuation of Companies with a Small Board of Directors. Journal of Financial Economics, 40:185.
11 Opportunistic accounting practices around stockfinanced mergers in Spain María J. Pastor-Llorca and Francisco Poveda-Fuente
Abstract We investigate whether acquiring firms manage accounting earnings through discretionary accruals choices around the stock-financed merger process. We hypothesize that the process of acquisition may provide incentives for managers to use accounting discretion to overstate earnings in order to attempt to increase the share price of the acquiring firm and to reduce the number of shares that must be issued to purchase the target firm. We also examine the market efficiency in processing potentially manipulated accounting reports and if the post-merger performance of acquiring firms could be partially attributable to the reversals of the effect of earnings management practices.
11.1
Introduction
The existence of earnings-management practices around some corporate decisions continues attracting an important debate, especially with the recent concern for transparency. There are some corporate events whereby the motivations to manipulate financial statements are expected to be strong. As noted by Dechow, Sloan, and Sweeney (1996) and Jiambalvo (1996), external financial events provide a setting in which managers have strong incentives to manage earnings. In particular, an interesting case is the stock-financed corporate mergers. Initial and seasoned equity offerings have been largely analyzed,1 whereas stock-financed mergers have received little attention to date in spite of the presence of manipulation incentives for acquiring firm managers. In some acquisitions the acquiring firm buys the target’s shares with cash, but the use of stocks is overwhelming as a method of payment in merger processes. 1
Teoh, Welch, and Wong (1998a), Rangan (1998), Shivakumar (2000), Zhou and Elder (2003), and Heron and Lie (2004) verify that firm commitment offerings in the U.S. market are preceded by significant increases in abnormal accruals. Teoh, Welch, and Wong (1998b) and Teoh, Wong and Rao (1998) for the U.S. market, and Pastor and Poveda (2006) for the Spanish market find that abnormal accruals are unusually high in the IPO year.
262
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Concretely, the acquiring firm issues new shares to exchange them for the target ones. The target shareholders receive a number of acquiring-firm shares for each of their target shares based on the price of the acquiring firm’s share once the acquisition agreement is reached. Thus, managers of acquiring firms have a clear interest to be overvalued, as the higher the price of the acquiring firm the fewer the number of shares that must be issued to purchase the target firm. If managers were convinced that the true value of their firm’s shares is worth more than the current share price, they would prefer not to issue equity to finance the acquisition. From the acquiring firm’s perspective, stock-financed mergers can be viewed as two simultaneous transactions, both an equity offering and a merger. Manipulation incentives surrounding external financial events are stronger around stock-financed mergers as the equity issue finances the purchase of the target firm by overstating earnings to increase the acquiring firm’s stock price. This will reduce the cost of buying the target as the number of shares that must be issued will be lower. Although the incentives to overstate earnings appear to be strong, acquiring firms choose not to do so. Agency theory predicts that earnings management is most feasible in those cases where the costs of miscalculation of earnings management are high, and such a situation is common when the user of accounting information is unsophisticated (Watts and Zimmerman, 1986; Erickson and Wang, 1999). However, this is not the case of stock-financed mergers because, on behalf of target shareholders, target firm managers make an important effort, with expert accountants or auditors, to ensure that the financial statements of the acquirer are free of accounting manipulation. The evidence in the U.S. market is consistent with earnings-management practices for stock mergers. Erickson and Wang (1999) and Louis (2004) find positive and statistically significant unexpected accruals in the merger announcement quarter for stock-financed acquisitions. These results are in line with the manipulation hypothesis obtained by Huang (2004) in the Hong Kong market. Huang (2004) documents a significant increase in abnormal accruals from the year before the merger right up to the year of the merger announcement, followed by a significant decrease in the year following the merger announcement. Moreover, abnormal accruals for stock acquirers are positive and significantly higher than those of cash acquirers in the year the merger is announced. Thus, managers of acquiring firms use accounting discretion to overstate earnings by attempting to increase share price and reduce the cost of buying the target, despite the potential costs of detection. If directors of acquiring firms make use of the discretion allowed in accounting rules to report high earnings—and if investors fail to understand that the high earnings arise from accounting choices and represent a transitory increase—those investors will overvalue acquiring companies. But earnings-management practices cannot be sustained over time, so they will revert in the periods following the merger; then investors will gradually correct previous overvaluation as earnings management reverses. This argument would be consistent with the long-run negative abnormal returns observed in the years following stock-financed mergers in the U.S. market, as documented by Agrawal, Jaffe, and Mandelker (1992),
Opportunistic accounting practices around stock-financed mergers in Spain
263
Loughran and Vijh (1997), and Louis (2004). The evidence obtained by Huang (2004) in the Hong Kong market and Gregory and Matatko (2006) in the United Kingdom is also consistent with the existence of post-merger negative long-run abnormal returns. The interpretation of the long-run abnormal returns is subject to considerable discussion because several authors document important biases in measuring and testing returns in long periods; therefore, long-run abnormal returns could be due exclusively to these biases (Barber and Lyon, 1997; Kothari and Warner, 1997; and Lyon, Barber, and Tsai, 1999). The argument of earnings-management practices as a possible explanation for negative abnormal returns is examined in two studies by Louis (2004) and Huang (2004), wherein the authors find that subsequent returns are negatively correlated with abnormal accruals in the merger announcement year. Our primary objective in this chapter is to examine whether acquiring firms in Spain manipulate accounting earnings upward to reduce the cost of buying the target. Past studies on earnings management have focused on the U.S. market, whereas research on earnings management is absent for numerous countries. In this study we extend the empirical literature on earnings management by testing these accounting practices in a market whose intrinsic characteristics and accounting rules are different from those of the United States. The flexibility of accounting accruals is different between Anglo-Saxon accounting systems (e.g., U.S.) and continental accounting systems (e.g., Spain). In addition, the differences between ownership concentration in the United States and Spain are investor protection and legal enforcement, which give rise to different incentives using accrual discretion to implement earnings management (Leuz, Nanda, and Wysocki, 2003; Hung, 2001). Thus, testing earnings management of stock mergers in a very different market will expand our knowledge on these accounting practices and their role in corporate events. We focus on the market efficiency in interpreting accounting accruals by examining if the post-merger performance of acquiring firms is partially attributable to the reversals of the effect of earnings-management practices. In particular, we first analyze the stock price performance following stock-by-stock mergers incorporating the methodological procedures that minimize the biases observed in the long-term event studies in order to avoid detecting abnormal returns where there are none. Then, we examine the possible relationship between opportunistic earnings management practices and post-merger abnormal returns. Considerable debate remains regarding long-run abnormal returns and the possibility that they originate exclusively from methodological biases. The abnormal pattern present in the Spanish market can be affected by earnings-management practices. The chapter is organized as follows. The next section describes the sample selection and data used. Section 11.3 discusses the measurement of earnings management. Section 11.4 tests the existence of earnings-management practices by acquirer firms during the year of the merger decision. Section 11.5 analyzes the long-run performance of post-merger stock returns and their possible relationship with earnings management practices. Finally, conclusions are presented in Section 11.6.
264
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
11.2
Sample and data
To prepare our database of stock-financed mergers, we employ the review of Bolsa de Madrid and the register of the Comisión Nacional del Mercado de Valores (CNMV), which is the Spanish equivalent of the American SEC. During the 1992–2002 period, 82 stock-financed mergers of companies were listed in the Sistema de Interconexión de las Bolsas Españolas (SIBE). From this initial sample we filter out financial companies, because their accruals are very different from those of industrial firms, hence, narrowing the sample to 68 mergers. We then test the existence of earnings-management practices for the year prior, before, and after the merger year as well as 2 years after the merger. We extend the analysis of the year before the merger and 2 years after the merger to get a better assessment of the pattern and possible reversion in earnings-management practices. To estimate acquirers’ accounting accruals, we use first differences and accounting data from CNMV from year ⫺2 to year ⫹2, where 0 is the year of the merger announcement. Moreover, to compute acquirers’ abnormal accruals, we estimate the normal component of accruals with an estimation sample made up of non-event firms, and use accounting data from the CNMV for matching control firms. We examine stock returns during the 2 years following the merger decision, and calculate monthly returns adjusted by dividends, rights issues, and splits. We obtain this information from the daily closing stock prices of the SIBE register during the investigation period. This further refines our event sample to 46 stock-financed corporate mergers.
11.3
Measuring earnings management
Accounting accruals are the center point of earnings-management tests, and are defined as the difference between earnings before extraordinary items and discontinued operations and cash-flow from operations. The accrual adjustments reflect business transactions which, in turn, affect future cash-flows, even though cash has not currently changed hands. Under generally accepted accounting principles, firms have the discretion to recognize these transactions so that reported earnings reflect the true underlying business conditions of the company. However, managerial flexibility in accruals also opens opportunities for earnings management. In this study, we specifically use the standard definition of current accruals2: ACCit ⫽ (ΔCAit ⫺ ΔCASHit ) ⫺ (ΔCLit ⫺ ΔSTDit ) 2
(11.1)
We employ current accruals, leaving aside depreciations, owing to the aggregation of accounts in the CNMV database. In particular, the level of gross property, plant, and equipment, which is required as a control variable in order to estimate normal depreciation, does not appear. Moreover, as we analyze the pattern of accruals following equity offering, it is advisable to focus on processes with homogeneous reversion.
Opportunistic accounting practices around stock-financed mergers in Spain
265
where ACCit are current accruals, ΔCAit is the change in current assets, ΔCASHit is the change in cash and cash equivalents, ΔCLit is the change in current liabilities, and ΔSTDit is the change in short-term debt. Subscripts i and t refer to company and period, respectively. Observable current accruals, ACCit, can theoretically be broken down into two unobservable components: the nondiscretionary or normal part, NACCit, and the abnormal component, AACCit, which can be used as a proxy for earnings management. Several theoretical models have attempted to obtain this breakdown by estimating the pattern of accruals in the absence of accounting discretion. Thus, the part of accruals not explained by the model is used as a proxy for earnings management as a variation in this component, which represents a manager’s effort to manipulate earnings more than just a change in exogenous economic conditions. The most popular model to estimate the accrual’s normal component is the modified version of the Jones (1991) model proposed by Dechow, Sloan, and Sweeney (1995). We estimate coefficients for each cluster sector-year with the original Jones model (see Eq. 11.2): ACC jt MTAjt
⎛ 1 ⎜ ⫽ αst ⎜⎜ ⎜⎜⎝ MTAjt
⎛ ΔNSALES ⎞⎟ jt ⎟⎟ ⫹ β ⎜⎜ st ⎜ ⎟⎟ ⎜ MTA jt ⎝⎜ ⎠
⎞⎟ ⎟⎟ ⫹ u jt ⎟⎟ ⎠
(11.2)
where j firms are non-event companies belonging to the same two-digit activity sector of acquiring firm i, and the subscript s refers to the activity sector to which the company i belongs. ACCjt and ΔNSALESjt are current accruals and the change in net sales in year t for firm j, respectively, and MTAjt is the mean total asset from year t ⫺ 1 to year t for firm j. Given that current accruals are not homogeneous among different activity sectors, we estimate the coefficients of this model for each activity sector and year. The estimation sample in each cluster sector-year includes exclusively non-event firms, requiring a minimum of 10 observations. This intra-industry cross-sectional regression is estimated for each acquiring firm i and year in the test period, i.e., from year ⫺1 to year ⫹2 relative to the year of the merger. Once the coefficients are estimated, Dechow, Sloan, and Sweeney (1995) suggest an adjustment in the original Jones model in order to avoid errors in the estimation of discretionary accruals when there is discretional behavior through sales. With this modification, the abnormal current accruals are estimated for acquiring firms as follows:
⎛ ACC it AACCit ⫽ ⎜⎜⎜ ⎜⎝ MTAit
⎡ ⎢ ⎞⎟ ⎢ ⎛ 1 ⎟⎟ ⫺ ⎢ αˆ st ⎜⎜ ⎟⎠ ⎢ ⎜⎜⎝ MTAit ⎢ ⎢ ⎣
adjustment ⎞⎤ ⎛ ⎟⎟⎥ ⎜⎜ ⎞⎟ ⎜⎜ ΔNSALESit ⫺ ΔTRit ⎟⎟⎟⎥ ⎟⎟ ⫹ βˆ st ⎜ ⎟⎟⎥ ⎜⎜ ⎟⎟⎥ ⎟⎠ MTAit ⎜⎜ ⎟⎟⎥ ⎟⎠⎥ ⎜⎝ ⎦ (11.3)
266
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
where AACCit is the abnormal component of current accruals and ΔTRit is the change in trade receivable for the acquiring company i in year t. The subscript s refers to the activity sector to which firm i belongs. As pointed out in Fields, Lys, and Vincent (2001), the most relevant problem in using abnormal accrual models to test earnings management is the specification error when sample firms have extreme financial performance.3 In this context, Kothari, Leone, and Wasley (2005) propose controlling for the impact of performance on estimated discretionary accruals using a performance-matched firm discretionary accrual. Their results indicate that matching on ROA provides the best specified and most powerful measures of discretionary accruals. Moreover, Kothari, Leone, and Wasley (2005) show that performance matching is critical to designing a well-specified test of earnings management. The procedure we follow to obtain the performance-matched abnormal accruals consists of matching acquiring firms with nonacquiring firms in the same year and industry and closest in terms of ROA. Therefore, the performance-matched abnormal accrual results from the difference between the abnormal accrual for the acquiring firm and the abnormal accruals for their matching companies. The assumption underlying this procedure, based on Kothari, Leone, and Wasley (2005), is that the real discretionary accruals of the control firm are 0 or near 0, and the single component of discretionary accrual proxy is due to the common measurement error induced by performance:
m MAACCevent firm,t
m MAACCevent firm,t
⎛ ⎞⎟ m AACCev ⎜⎜ entfirm,t ⎟ ⎜⎜ ⎟⎟⎟ m ⎜ ⫽ ⎜ DAeventfirm,t ⫹ ηevent firm,t ⎟⎟ ⎟⎟ ⎜⎜ ⎟⎟ ⎜⎜ Earnings Management ⎟⎠ ⎜⎝ ⫽⬎DA ≠0 ⎛ ⎞⎟ m AACCcontrolfirm, ⎜⎜ t ⎟⎟ ⎜⎜ ⎟ m ⫹ ηcontrol firm,t ⎟⎟⎟ ⫺⎜⎜ DA Acontrol firm,t ⎜⎜ ⎟⎟⎟ ⎜⎜ No Earnings Management ⎟⎟ ⎜⎝ ⫽⬎DA ≅00 ⎠ m m ⫽ DAevent firm,t ⫹ (ηeven ⫺ η t firm,t control firm,t ) ≅ DAevent firm,t
ROA matched⫽⬎≅0
(11.4)
where MAACCitm is the matched-performance abnormal accrual proxy for firm i in year t estimated using model m; AACCitm is the abnormal accrual proxy for firm i in year t estimated using model m; DAit is the discretionary accrual for firm i in year t; and η m it is the measure error induced by model m for firm i in year t. 3
Dechow, Sloan, and Sweenney (1995), Guay, Kothari, and Watts (1996), Healy (1996), and Dechow Kothari, and Watts (1998), among others, have verified the correlation between abnormal accruals estimated by common accruals models and firm performance.
Opportunistic accounting practices around stock-financed mergers in Spain
267
Finally, to extend the evidence obtained in the paper, we use a second matched procedure consisting of calculating the matched abnormal accrual proxy as the excess of the abnormal accrual, estimated for firm i in year t, in relation to the median abnormal accrual of the non-event firms in the same cluster sector-year. This matched-sector abnormal accrual is used as an additional test to assess the robustness of the analysis.
11.4
Accruals pattern around the time of the merger
11.4.1
Current accruals around the merger decision
As a first approximation, we examine the pattern of current accruals without getting into the estimation of the normal component. To control for temporal effects of this profile with regard to the specific year in which the merger is made, or related to isolated sector patterns, we also analyze the performance of current accruals in relation to non-event control samples. To ensure the robustness of results, control firms are selected following the two alternative criteria described in the previous section. Table 11.1 displays the results of this first approximation. In Panel A of Table 11.1 we observe how the mean current accrual for merger firms is calculated using Eq. (11.1), which is positive and significantly different from 0 in the merger year and a year prior to the merger with p-values less than 1%. The maximum value is reached in the event year (0.1172) with a p-value close to 0. In the following years the values are not statistically different from 0, suggesting the existence of an accrual reversion after the merger. This reversion pattern is also observed in medians with a maximum value of 0.0618 in the event-year, which is statistically significant at the 1% level using the Wilcoxon nonparametric test. In Panel B of Table 11.1, the MACC variable measures the excess of acquiring firms’ current accruals in relation to non-event firms with similar ROA, in the same sector-year. The mean pattern of this adjusted variable reaches the maximum value (0.0741) in year 0, where it is statistically significant with a p-value of 7% in a two-sided student t-test. In the post-merger years, there is a clear decline and values are not statistically different from 0. If we focus on medians, the maximum value of performance-adjusted accruals is located in the event year (0.0418) with a Wilcoxon p-value of 2%. From the medians, we observe the reversion of accruals in subsequent years. Finally, to provide an alternative adjusted measure of matched accruals, in Panel C of Table11.1, the “MACC” variable measures the excess of acquirers’ current accruals in relation to the non-event sector-year medians. By observing the means, the highest level is reached in the event year with an excess of 0.0911, statistically significant at a level of 1%. If we focus on the years after the merger, the pattern of adjusted accruals reflects a progressive decline. Using nonparametric tests, the results are qualitatively similar with a maximum of 0.0427 in the event year with a Wilcoxon p-value of 2%. Thus, the pattern of current accruals
268
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 11.1 Current accruals around Spanish mergers PANEL A: Current accruals
Year ⫺1 Event Year Year ⫹1 Year ⫹2
Obs.
Mean ACC
p-Value mean ACC
Median ACC
p-Value median ACC
46 46 46 46
0.0611 0.1172 0.0252 ⫺0.0131
0.01 0.00 0.26 0.58
0.0309 0.0618 0.0084 ⫺0.0056
0.01 0.00 0.88 0.66
PANEL B: Matched performance current accruals
Year ⫺1 Event year Year ⫹1 Year ⫹2
Obs.
Mean MACC
p-Value mean MACC
Median MACC
p-Value median MACC
42 42 42 42
0.0298 0.0741 0.0290 0.0225
0.25 0.07 0.31 0.40
0.0158 0.0418 0.0264 0.0037
0.76 0.02 0.36 0.88
PANEL C: Matched sector current accruals
Year ⫺1 Event year Year ⫹1 Year ⫹2
Obs.
Mean MACC
p-Value mean MACC
Median MACC
p-Value median MACC
46 46 46 46
0.0557 0.0911 0.0313 ⫺0.0109
0.00 0.00 0.10 0.63
0.0174 0.0427 0.0072 ⫺0.0052
0.37 0.02 0.55 1.00
Obs: Number of observations; ACC: current accruals; MACC: matched current accruals by performance (panel B) or by sector-median (panel C); p-value mean: p-value of the two-sided test of the null hypothesis than Mean (ACC) ⫽ 0; p-value median: p-value of the Wilkoxon robust test of the null hypothesis than Median (ACC) ⫽ 0; Event year: Year of merger; Year ⫹j: Year j relative to the event year; Significant coefficients, at least at 10% nominal size, are remarked in bold.
experiences a clear peak in the year of the merger which would be consistent with managers overstating earnings in this date.
11.4.2
Abnormal accruals analysis
In this subsection we analyze whether the accrual pattern holds if we break down accruals into their normal and abnormal components using the procedure presented in Section 11.3. Table 11.2 shows abnormal accruals around the merger date. We notice how the highest level of abnormal accruals is located in the event year regardless of the matching procedure employed. This maximum is supported by low p-values at the 1% significance level, which allows us to reject the absence of earnings management. Without using any matching procedure (Panel A), the
Opportunistic accounting practices around stock-financed mergers in Spain
269
Table 11.2 Abnormal accruals around Spanish mergers PANEL A: Abnormal accruals from cross section modified Jones model
Year ⫺1 Event Year Year ⫹1 Year ⫹2
Obs
Mean AAC
p-Value mean AACC
Median AACC
p-Value median AACC
41 41 41 41
0.0223 0.0828 0.0259 ⫺0.0078
0.25 0.02 0.28 0.74
0.0052 0.0370 ⫺0.0010 ⫺0.0063
0.53 0.01 1.00 1.00
PANEL B: Matched performance abnormal accruals from cross section modified Jones model
Year ⫺1 Event year Year ⫹1 Year ⫹2
Obs
Mean MAACC
p-Value mean MAACC
Median MAACC
p-Value median MAACC
37 37 37 37
⫺0.0026 0.0966 0.0314 0.0147
0.91 0.02 0.32 0.58
⫺0.0166 0.0401 0.0101 0.0029
0.87 0.02 0.26 0.75
PANEL C: Matched sector abnormal accruals from cross section modified Jones model
Year ⫺1 Event year Year ⫹1 Year ⫹2
Obs
Mean MAACC
p-Value mean MAACC
Median MAACC
p-Value median MAACC
41 41 41 41
0.0187 0.0863 0.0229 ⫺0.0046
0.30 0.01 0.32 0.85
0.0048 0.0462 ⫺0.0014 ⫺0.0001
0.53 0.01 1.00 1.00
Obs: Number of observations; AAC: Abnormal accruals estimated from the cross-sectional version of the modified Jones model; MAACC: Matched abnormal accruals estimated from the crosssectional version of the modified Jones model with the matching procedure indicated in each panel; p-value mean: p-value of the two-sided test of the null hypothesis than Mean (MAAC) ⫽ 0; p-value median: p-value of the Wilkoxon robust test of the null hypothesis than Median (MAACC) ⫽ 0; Event year: Year of merger; Year ⫹j: Year j relative to the event year; Significant coefficients, at least at 10% nominal size, are remarked in bold.
maximum mean value (0.0828) is reached in year 0 with a p-value of approximately 2%, and the following years show a clear decline. If we focus on medians, results are qualitatively quite similar, with a peak (0.0370) in the event year, which is statistically significant at the 1% level using the Wilcoxon nonparametric. Panel B of Table 11.2 displays the results with the performance-matching procedure suggested by Kothari, Leone, and Wasley (2005). We observe how the highest levels of matched-performance abnormal accruals, both in the mean and median, are positive and statistically significant in year 0. The mean reaches a value of 0.0966 in the event year, which is statistically significant at a 2% level and clearly reverses in the following years. With the nonparametric tests,
270
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
the medians support the same pattern with a peak of 0.0401 (Wilkoxon p-value of 2%) and an immediate reverse to 0 values. Finally, in Panel C of Table 11.2, we replicate the same analysis but measure the excess of abnormal accruals in relation to the median of the non-event firms in the same sector-year. Results are highly significant and are similar with a mean and a median pattern showing the maximum value in the event year, followed by a subsequent decrease.4 Results are robust using different models and several matching procedures, and they are similar to the evidence found by Erickson and Wang (1999) and Louis (2004) for the U.S. markets. Our results confirm that acquiring firms in the Spanish market use discretion to overstate earnings around the merger decision. The question is, do these accounting practices detected in the merger year have any relationship with the post-merger stock price performance.
11.5
Earnings management and post-merger stock price performance
In this section, we first analyze the long-run abnormal returns the years after the merger, and then we apply regression analysis to examine whether the unusually high abnormal accruals detected in the year of the event affect post-merger returns. To test the predictive power of accruals in future returns, we must take into account the fact that the return analysis period should start after the market receives accruals information; therefore, we examine post-merger returns beginning the month after the publication of the event-year financial statement.
11.5.1
Stock returns performance
We study stock returns in the 1- and 2-year periods following the publication of the event-year financial statement. To examine abnormal returns in these periods, we apply two alternative procedures, an event-time analysis and a calendartime methodology. In the event-time analysis we examine returns adjusted by the market portfolio. Specifically, we calculate the abnormal return of company i in the post-merger period τ, ACoRi,τ, as the compound return of the acquiring firm i minus the compound market return: ACoRi τ ⫽
4
τ
τ
t ⫽1
t ⫽1
∏ (1 ⫹ Rit ) ⫺ ∏ (1 ⫹ RMt )
(11.5)
Tests have been replicated using the Poveda (2005) model and estimating each model by cross section and with panel data. Results are qualitatively similar to those presented in the tables so they are robust to model specification and estimation procedures. These results are available from the authors.
Opportunistic accounting practices around stock-financed mergers in Spain
271
where Rit and RMt are the returns of firm i and the market portfolio in month t, respectively, and τ is the number of months in the post-merger period analyzed. To test the statistical significance of these abnormal returns, we use the bootstrapping technique to simulate the empirical distribution of the traditional t and the skewness-adjusted t statistic. Kothari and Warner (1997) and Lyon, Barber, and Tsai (1999) show that this procedure has fewer misspecification problems and that it is the most appropriate for long-run event studies. Panel A of Table 11.3 reports average abnormal compound returns for the 1- and 2-year period following the publication of the event year financial statement as well as the p-values of the traditional and the skewness-adjusted t tests using the bootstrap technique. For the 1-year period the mean abnormal return is ⫺9.09%, statistically significant with a p-value of 7% for both tests. Abnormal returns worsen for the 2-year period, reaching the mean a value of ⫺17.75%, statistically significant with a p-value of 4% and 3% for both tests, respectively. Thus, results in this panel show that acquirers seem to experience negative abnormal returns in the years following stock finance mergers. The alternative procedure to compute and test post-merger abnormal returns is to examine the abnormal monthly mean return by applying a calendar-time portfolio approach. This methodology analyzes the strategy of holding a portfolio in each calendar month made up of stocks affected by the event over the last τ months. In particular, we apply this methodology with a time horizon τ of 12 and 24 months, as well as in the event-time analysis. This calendar-time approach enables us to check the robustness of results and to avoid different problems of Table 11.3 Stock return performance following the year 0 financial statement PANEL A: Abnormal compound return in the post-merger period Analysis period
AACoR (%)
Traditional t statistic
Bootstrap p-value
Skewness adjusted t
Bootstrap p-value
12 months 24 months
⫺9.09% ⫺17.75%
⫺1.94 ⫺2.35
(0.07) (0.04)
⫺1.85 ⫺2.14
(0.07) (0.03)
PANEL B: Abnormal monthly mean return with Fama-French model in calendar-time regressions Analysis period
α ˆ p(%)
t Statistic
p-value
12 months 24 months
⫺0.69 ⫺0.57
⫺2.62 ⫺2.31
(0.01) (0.02)
Panel A shows the mean abnormal compound return for acquiring firms in the one and two year period following the year 0 financial statement. This panel also reports results of traditional t test and skewness-adjusted t test using the bootstrap technique to evaluate statistical significance. Panel B shows the abnormal monthly mean return in the post-merger period estimated with the Fama-French model in calendar-time regressions. We estimate the Fama-French regression with weighted least squares estimation where the weighting factor is based on the number of securities in the portfolio in each calendar month. Significant coefficients, at least at 10% nominal size, are remarked in bold.
272
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
accumulating returns over long periods. We can compute the monthly return of the calendar portfolio p in each month t as N pt
Rpt ⫽
∑ Rjt j⫽1
N pt
(11.6)
where Rjt is the return of firm j in month t and Npt is the number of stocks in the portfolio in month t. So we obtain the time series of the calendar-portfolio monthly returns and apply the Fama and French (1993) three-factor model to this time series to test the portfolio’s abnormal monthly mean return: Rpt ⫺ Rft ⫽ αp ⫹ β1p ⋅ (RMt ⫺ Rft ) ⫹ β2 p HMLt ⫹ β3p SMBt ⫹ εpt
(11.7)
where Rft is the 1-month Treasury Bill (risk-free) rate of return, HMLt is the difference in returns between portfolios made up of stocks with high and low book-to-market ratios and SMBt is the difference in returns between portfolios made up of stocks with high and low trading volumes, both orthogonalized. The Jensen’s alpha, αp, measures the abnormal monthly mean return of the calendar portfolio p; that is, the abnormal monthly mean return of acquirers over the investigation period—in our case, the 12 and 24 months following the merger. Panel B of Table 11.3 reports the results of the Fama–French calendartime regression. The change in the composition of the portfolio each month could lead to heteroskedasticity problems, as the variance depends on the number of firms in the portfolio. Following Lyon, Barber, and Tsai (1999), to correct for heteroskedasticity, we use a weighted least-squares estimation, where the weighting factor is based on the number of securities in the portfolio in each calendar month. The intercept from the Fama–French model, αp, is negative and significant for the two time periods analyzed. Thus, this evidence indicates that the abnormal monthly mean return of acquiring firms in the 1- and 2-year post-merger period is statistically negative, which accords with the negative abnormal compound returns illustrated in Panel A.
11.5.2
Post-merger returns and previous earnings management
In this section our objective is to evaluate the extent to which unusually high abnormal accruals detected in the year of the merger (year 0) have an influence on subsequent abnormal returns. To address this possible relationship, we carry out a regression analysis of post–merger-adjusted returns on year 0 abnormal accruals: ACoRi τ ⫽ γ0 ⫹ γ1AACCi0 ⫹ μi τ
(11.8)
Opportunistic accounting practices around stock-financed mergers in Spain
273
where ACoRiτ is the abnormal compound return of company i in period τ, τ being the 1- and 2-year periods following the merger. AACC 0i is the abnormal accruals for year 0, information that is available for investors at the beginning of the period over which abnormal returns are calculated. If the market understands the implications of these discretionary adjustments, the coefficient on AACC 0i will be 0. However, if investors misinterpret these accounting practices, believing that high earnings owing to discretionary accruals reflect good expectations, they could overvalue acquirers and gradually correct this overvaluation in the post-merger period. According to this argument, a negative relationship would be expected between the level of year 0 discretionary accruals and postoffering abnormal returns. In other words, the greater the earnings management at the time of the merger, the larger the post-issue price correction. We first run regression (11.8) calculating AACC 0i using the modified Jones model to estimate the normal component of the accrual as we explained in Section 11.3 and without matching. Second, we use as explanatory variable the abnormal accrual matched by ROA (MAACC) as Kothari, Leone, and Wasley (2005) propose. Finally, we employ abnormal accruals matched by sector medians. Coefficient estimations of this equation as well as their p-values in brackets are reported in panel A of Table 11.4. We observe how the three procedures are used to calculate abnormal accruals for both temporal horizons, where the coefficient of year 0 abnormal accruals (AACC0i ) is not statistically different from 0. Therefore, there is no relationship between post-merger performance and the unusually high abnormal accruals detected in the year of the merger announcement. These results are different from those obtained by Louis (2004) and Huang (2004), who find a negative relationship between abnormal accruals and post-merger adjusted returns. Neither are they in line with the negative relationship found between abnormal accruals and returns following seasoned and initial public offerings (Rangan, 1998; Teoh, Welch, and Wong, 1998a,b; Teoh, Wong, and Rao, 1998, for the U.S. market; and Pastor and Poveda, 2006, for initial public offerings in Spain). To investigate further if abnormal accruals affect post-merger returns, we develop an alternative regression analysis, employing a framework similar to that of Rangan (1998). The author links abnormal returns with earnings changes and abnormal accruals by regressing the following equation: ACoRi τ ⫽ γ0 ⫹ γ1UEi τ ⫹ γ 2 AACCi0 ⫹ μi τ
(11.9)
where UEiτ is the unexpected change in ROA not explained by the previous earnings management. The argument is that abnormal returns should be affected by the change in ROA, which measures the firm’s profitability. A positive relationship between adjusted returns and post-merger change in ROA is expected, but the change in ROA could be related to accounting accruals, so we include in the regression the change in ROA not explained by year 0 abnormal accruals (UEiτ).
274
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 11.4 Abnormal accruals in the merger year and subsequent stock return performance PANEL A: Abnormal accruals and subsequent stock returns Abnormal accruals without matching
Obs. Coeff. AACC0 p-Value
ROA matched abnormal accruals
Sector matched abnormal accruals
ACoR1
AcoR2
ACoR1
AcoR2
ACoR1
AcoR2
36 ⫺0.0917 (0.70)
36 ⫺0.1056 (0.82)
33 0.0703 (0.73)
33 0.0436 (0.98)
36 ⫺0.0538 (0.82)
36 ⫺0.0598 (0.90)
PANEL B: Abnormal accruals and subsequent stock performance: Rangan’s approach Abnormal accruals without matching
Obs. Coeff. AACC0 p-Value Coeff. UE p-Value
ROA matched abnormal accruals
Sector matched abnormal accruals
ACoR1
AcoR2
ACoR1
AcoR2
ACoR1
AcoR2
36 ⫺0.0900 (0.71) 0.2176 (0.86)
36 ⫺0.1042 (0.83) 0.2359 (0.92)
33 0.0717 (0.73) 0.3729 (0.75)
33 0.0437 (0.92) 0.0172 (0.99)
36 ⫺0.0517 (0.83) 0.2382 (0.85)
36 ⫺0.0580 (0.91) 0.2396 (0.92)
Panel A reports coefficients of Eq. (8) as well as their p-values in brackets. ACoR1 and ACoR2 are abnormal returns for the 1 and 2-year post-merger period, respectively. N is the number of observations in each regression. AACC0 is the level of abnormal accruals in the merger year computed with the matching procedure indicated. Panel B reports coefficients of Eq. (9) as well as their p-values in brackets. UE is the unexpected ROA variation
If the market is efficient, it will react to unexpected changes in earnings only and γ1 will be positive; however, we do not find any relationship between returns and abnormal accruals because the market will interpret correctly that these accounting practices do not affect the real future cash flows of the company. If investors fail to understand the implications of discretionary adjustments, γ2 will be negative. The variable AACC 0i in regression (11.9) is calculated following the three procedures explained previously, without matching abnormal accruals, matching by ROA, and matching by sector medians. The coefficient estimations of this equation and their p-values (in brackets) are reported in panel B of Table 11.4. As in Panel A, results in Panel B show that any procedure employed to calculate abnormal accruals and for both temporal horizons, the coefficient of year 0 abnormal accruals (AACC i0) is not statistically significant. Thus, the market seems to be interpreting correctly discretionary accounting practices, as the level of earnings manipulation does not affect post-merger returns. Stock-price underperformance in the years following the merger decision does not seem to be related to
Opportunistic accounting practices around stock-financed mergers in Spain
275
previous overvaluation due to earnings-management practices. However, in our opinion, these results should be interpreted with caution because there is no relationship between the unexpected change in ROA and returns. Perhaps variables not included in the regression were affecting post-merger return performance and affecting our results.
11.6
Conclusion
The growing concern for transparency and quality of financial information revealed by firms has become an important debate regarding the existence of earnings-management practices around some corporate events. An interesting case is stock-financed mergers, in which managers of acquiring companies have incentives to overstate earnings because higher earnings can push up stock prices and yield more favorable exchange ratios for stock takeovers. Therefore, the overvaluation of equity provides acquirers with a low-cost method to pay for their acquisitions. We conclude that the evidence for the Spanish market is similar to that of directors of acquiring firms who make use of the accounting discretion to report high earnings at the time of the merger decision, verifying that stock-financed mergers produce similar accruals patterns in very different markets. This similarity suggests a global anomaly, in line with the idea that not only accounting and market rules, but also other firm characteristics, shape the final accounting policy choice (Leuz, 2005; Ball, Robin, and Wu, 2003; Holthausen, 2003). The fact of detecting earnings-management practices is a genuinely relevant question for consideration by Spanish market regulators, as these institutions should make an effort for the transparency and quality of financial information revealed by firms. As in the U.S. market, Spanish stock-financed acquirers experience negative long-run abnormal returns following the merger decision. Therefore, it does not appear to be a local anomaly, but rather is present in markets with different characteristics. However, opportunistic earnings management practices do not lead to this abnormal pattern. Abnormal accruals detected in the merger year are not a significant determinant of the long-term performance of stockfor-stock acquirers. Thus, additional research is necessary to examine the possible explicative factors of the negative long-run abnormal returns following stock-financed mergers in the Spanish market.
References Agrawal, A., Jaffe, J., and Mandelker, G. (1992). The Post-Merger Performance of Acquiring Firms: A Re-examination of an Anomaly. The Journal of Finance, 47:1605–1621.
276
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Ball, R., Robin, A., and Wu, J. (2003). Incentives versus Standards: Properties of Accounting Income in Four East Asian Countries. Journal of Accounting & Economics, 36:235–270. Barber, B. M., and Lyon, J. D. (1997). Detecting Long-Run Abnormal Stock Returns: The Empirical Power and Specification of Tests Statistics. Journal of Financial Economics, 43:341–472. Dechow, P., Sloan, R., and Sweeney, A. (1996). Causes and Consequences of Earnings Manipulation: An Analysis of Firms Subject to Enforcement Actions by SEC. Contemporary Accounting Research, 13:1–36. Dechow, P., Kothari, S., and Watts, R. (1998). The Relation between Earnings and Cash Flows. Journal of Accounting and Economics, 25:133–168. Dechow, P. M., Sloan, R. G., and Sweenney, P. (1995). Detecting Earnings Management. The Accounting Review, 70:193–225. Erickson, M., and Wang, S. (1999). Earnings Management by Acquiring Firms in Stock for Stock Mergers. Journal of Accounting and Economics, 27:149–176. Fama, E. F., and French, K. R. (1993). Common Risk Factors in the Returns on Stocks and Bonds. Journal of Financial Economics, 33:3–56. Fields, T., Lys, T., and Vincent, L. (2001). Empirical Research in Accounting Choice. Journal of Accounting and Economics, 31:255–307. Gregory, A., and Matatko, J. (2006). Long-Run Abnormal Returns to Acquiring Firms: The Form of Payment Hypothesis, Bidder Hostility and Timing Behavior. Working Paper. Centre for Finance and Investment, University of Exeter. Guay, W., R. (1996). A Market-Based Evaluation of Discretionary Accrual Models. Journal of Accounting Research, Vol. 34(S):83–105. Healy, P. (1996). Discussion of a Market-Based Evaluation of Discretionary Accrual Models. Journal of Accounting Research, 34(S):107–111. Heron, R. A., and Lie, E. (2004). A Comparison of the Motivations for and the Information Content of Different Types of Equity Offerings. Journal of Business, 77:605–632. Holthausen, R. W. (2003). Testing the Relative Power of Accounting Standards versus Incentives and Other Institutional Features to Influence the Outcome of Financial Reporting in an International Setting. Journal of Accounting & Economics, 36:271–283. Huang, Y. (2004). Earnings Management in Stock Financed Takeovers: The Role of Analysts as Information Intermediary. Working Paper. Hong Kong University of Science and Technology. Hung, M. (2001). Accounting Standards and Value Relevance of Financial Statements: An International Analysis. Journal of Accounting and Economics, 30:401–420. Jiambalvo, J. (1996). Discussion of Causes and Consequences of Earnings Manipulation: An Analysis of Firms Subject to Enforcement Actions by SEC. Contemporary Accounting Research, 13:37–47. Jones, J. (1991). Earnings Management during Import Relief Investigations. Journal of Accounting Research, 29:193–228
Opportunistic accounting practices around stock-financed mergers in Spain
277
Kothari, S. P., and Warner, J. B. (1997). Measuring Long-Horizon Security Price Performance. Journal of Financial Economics, 43:301–339. Kothari, S. P., Leone, A. J., and Wasley, C. E. (2005). Performance Matched Discretionary Accrual Measures. Journal of Accounting and Economics, 39:163–197. Leuz, C. (2005). Cross Listing, Bonding and Firms’ Reporting Incentives: A Discussion of Lang, Raedy and Wilson. Journal of Accounting and Economics. Forthcoming. Available at SSRN: http://ssrn.com/ abstract⫽800304 Leuz, C., Nanda, D., and Wysocki, P (2003). Investor Protection and Earnings Management: An International Comparison. Journal of Financial Economics, 69:505–527. Loughran, T., and Vijh, A. M. (1997). Do Long-Term Shareholders Benefit from Corporate Acquisitions? Journal of Finance, 52:1765–1790. Louis, H. (2004). Earnings Management and the Market Performance of Acquiring Firms. Journal of Financial Economics, 74:121–148. Lyon J. D., Barber, B. M., and Tsai, C. (1999). Improved Methods for Tests of Long-Run Abnormal Stock Returns. Journal of Finance, 54:165–201. Pastor, M. J., and Poveda, F. (2006). Earnings Management and the Long-Run Performance of Spanish Initial Public Offerings. Initial Public Offerings: An International Perspective (Gregoriou, G. N., ed.), Elsevier/ Butterworth–Heinemann. Poveda, F. (2005). Nuevo Enfoque en la Estimación del Componente Anormal del Resultado. Moneda y Crédito, 221:105–148. Rangan, S. (1998). Earnings Management and the Performance of Seasoned Equity Offerings. Journal of Financial Economics, 50:101–122. Shivakumar, L. (2000). Do Firms Mislead Investors by Overstating Earnings before Seasoned Equity Offerings? Journal of Accounting and Economics, 29:339–371. Teoh, S., Welch, I., and Wong, T. J. (1998a). Earnings Management and the Underperformance of Seasoned Equity Offerings. Journal of Financial Economics, 50:63–99. Teoh, S., Welch, I., and Wong, T. J. (1998b). Earnings Management and the Long-Run Market Performance of Initial Public Offerings. Journal of Finance, 53:1935–1974. Teoh, S., Wong, T. J., and Rao, G. R. (1998). Are Accruals during Initial Public Offerings Opportunistic? Review of Accounting Studies, 3:175–208. Watts, R., and Zimmerman, J. (1986). Positive Accounting Theory. New York: Prentice-Hall. Zhou, J., and Elder, R. (2003). Audit Quality and Earnings Management by Seasoned Equity Offering Firms. Working Paper. School of Management, Binghamtom University.
This page intentionally left blank
12 Size does matter—firm size and the gains from acquisitions on the Dutch market Roman Kräussl and Michel Topper
Abstract Based on a sample of 269 M&A transactions that involved a Dutch buy-side over the years 1980 to 2003, this study presents evidence of a significant size effect on the Dutch market. Small companies earn returns about 2.45% larger than those of large companies over the three-day event window around the announcement of an acquisition. Our empirical findings prove that the size effect holds over a number of different time periods and exists independent of (i) the method of payment, (ii) the organizational form of the target company, (iii) the nature of the deal, (iv) the geographic scope and (v) the mode of acquisition.
12.1
Introduction
A common method to determine the short-term success of an acquisition is to look at the cumulative abnormal return around the day of the acquisition. Although many researchers have investigated the returns to the acquirer, their analysis has yielded no clear outcome. For example, Berkovitch and Narayanan (1993) as well as Andrade, Mitchell and Stafford (2001) find negative bidder returns. On the contrary, Bradley, Desai, and Kim (1987) as well as Moeller, Schlingemann, and Stulz (2004) find positive returns. In addition to the fact that returns from M&A transactions vary over different kinds of companies, they also vary over different kinds of deal characteristics. Numerous hypotheses have been developed in order to explain the large variation among the abnormal return of acquisitions. The five major deal characteristics that have proved to influence the abnormal returns are (i) the method of payment (Travlos, 1987), (ii) the organizational form of the target company (Fuller, Netter, and Stegemoller, 2002), (iii) whether an acquisition is industrially diversifying or not (Morck, Shleifer, and Vishny, 1990), (iv) whether an acquisition is internationally diversifying or not (Moeller and Schlingemann, 2004), and (v) the mode of acquisition (Jensen and Ruback, 1983).
280
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Moeller, Schlingemann, and Stulz (2004) conclude that, independent of the deal and target characteristics, another factor exists that influences the abnormal returns upon announcement of a transaction—size. They study a sample of 12,023 U.S. acquisitions between 1980 and 2001 and show that small acquirers earn higher abnormal returns than do larger acquirers. Independent of the organizational form of the target and method by which the acquisitions are paid for, small companies earn about 2% higher returns than do large companies. The authors argue that this significant difference arises from a size effect. This chapter analyzes whether this U.S. size effect can also be found on the Dutch M&A market. We analyze 269 successful M&A transactions over the period from 1980 until 2003. The empirical results show that small companies earn higher returns upon announcement of an acquisition than large companies do. For the whole period, small companies are found to earn a 2.65% return upon announcement of a transaction while large companies earn an insignificant 0.2%. This difference in returns is found to hold over a number of different time periods. We also control our empirical results for the above-mentioned five major deal characteristics. Though the magnitude of the difference in returns varies over different deals, we show that large companies never earn larger returns than small companies. We conclude that, consonant with research performed in the United States, evidence of a statistically and economically significant size effect can be found on the Dutch market for corporate takeovers. The remainder of this chapter is organized as follows: Section 12.2 presents the empirical analysis by discussing the hypotheses, the building of the data set and the event study approach. Section 12.3 discusses the empirical results. Section 12.4 concludes and gives an outlook for future research.
12.2
Empirical analysis
12.2.1
Hypotheses
Moeller, Schlingemann, and Stulz (2004) provide evidence of a size effect on the U.S. market for mergers and acquisitions: Smaller companies earn significant larger returns upon announcement of an acquisition than do larger companies. To analyze whether Dutch acquirers deal with a similar size effect, we have to investigate whether there is a significant difference in cumulative abnormal returns (CARs) between small and large companies upon announcement of an acquisition. This results in our hypothesis: H1: Small companies earn significant larger CARs than large companies when announcing an acquisition. However, the fact that small companies earn larger returns than large companies does not automatically prove the existence of a size effect. It might well be that the difference in returns is caused by the fact that small and large companies engage in
Size does matter—firm size and the gains from acquisitions on the Dutch market
281
deals with different characteristics. Numerous academic papers have shown that certain deal and target characteristics can influence the abnormal returns; these characteristics include the method of payment, the target’s organizational form, whether or not a deal is diversifying (industrially or internationally), or the mode of acquisition. To prove the existence of a size effect, the difference in returns will have to hold independent from the deal and target characteristics. In order to check whether evidence of a size effect is indeed found, we introduce five control hypotheses. Travlos (1987) shows that acquisitions financed with cash have higher returns upon announcement than acquisitions financed with equity. It might be the case that small companies more often use cash as the method of payment than large companies do. Cash payment results in higher abnormal returns for a transaction, so this fact could explain why small companies enjoy higher CARs upon announcement of a transaction. Hence the first control hypothesis: H2a: Small companies earn significant larger CARs independent of the method of payment. Fuller, Netter, and Stegemoller (2002) find that the announcement returns are lower when companies announce the acquisition of a public company than when they announce the acquisition of a nonlisted company. Small companies are two times more likely to take over private companies. This implies that a potential size effect might be due merely to the fact that small companies engage in acquisitions with nonpublic targets and leads to the second control hypothesis: H2b: Small companies earn significant larger CARs independent of the organizational form of the target. Morck, Shleifer, and Vishny (1990) conclude that the announcement returns for companies that engage in diversifying acquisitions are lower than the announcement returns for companies that do acquisitions in a related industry. It might be the case that a potential size effect exists just because smaller companies acquire targets from the same industry. Hence the third control hypothesis for the size effect: H2c: Small companies earn significant larger CARs independent of whether the deal is an industrial diversifying acquisition or not. Moeller and Schlingemann (2004) document significantly lower announcement returns for companies that announce a cross-border transaction. This phenomenon would especially apply to large companies, as it is probable that they “outgrow” their domestic market more quickly than do small companies. Thus, the fourth control hypothesis explicitly tests whether a potential size effect is due to the fact that small companies engage in domestic transactions: H2d: Small companies earn significant larger CARs independent of whether the deal is an international diversifying acquisition or not. Jensen and Ruback (1983) show that the returns to bidders are small and insignificantly positive for successful merger bids, while successful tender offers are found to result in small but significant positive returns. The final control hypothesis investigates whether a potential size effect exists just because small companies engage more often in tender offers: H2e: Small companies earn significant larger CARs independent from the mode of acquisition.
282
12.2.2
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Data
The sample of transactions originated from the Securities Data Company’s (SDC) Merger and Acquisitions Database. We selected only those M&A transactions that include a Dutch acquirer and that were announced between January 1, 1980, and December 31, 2003. We selected only those transactions in which the acquirer ended up with a 100% stake in the target. These criteria ensured that all transactions resulted in full control of the acquired company and that share buyback programs were eliminated from the sample. Transactions in which the acquirer possessed a stake larger than 50% before the transaction was announced were also excluded from the sample. In addition to this, transactions have to satisfy the following five conditions in order to be included in the sample: 1. The bidding firm is a Dutch public firm, and its daily returns must be available in Datastream from 138 trading days prior to the takeover announcement. 2. The deal value has to be larger than $1 million. 3. The target has to be a public firm, a private firm, or a nonpublic subsidiary of a public or private firm. 4. Only the announcement returns of deals that were completed were included in the sample; unsuccessful bids and pending transactions were excluded. 5. Investment companies and (real estate) funds were excluded from the sample because the characteristics of these companies differ from the other companies.
The initial sample obtained from SDC held a number of 711 M&A transactions that involved a Dutch buy-side over the period 1980–2003. After applying the above criteria to the initial sample of transactions, a sample of 269 successful offers resulted. Table 12.1 shows that 49.4% of the successful transactions were done by large companies, which are defined as companies that belong to the 10% largest public Dutch companies by market capitalization, one week prior to the day they announce a transaction. All other companies are referred to as small companies. Table 12.1 indicates that during the 1980s the number of successful acquisitions was fairly low. It took until the 1990s before M&As started to gain popularity in The Netherlands. As the Internet bubble grew during the second half of the 1990s, the volume of M&A transactions rose as well. The glory days of the Dutch M&A market took place at the height of the Internet hype during the years 1999 and 2000, resulting in a record of 47 transactions in 2000. After the bursting of the Internet bubble in 2000, the deal volume dropped dramatically and small companies accounted for the largest share of M&A activity in the years 2001–2003. Table 12.2 provides an overview on the transaction statistics. The information is reported for the sample as a whole and for small and large companies independently. Not surprisingly, the average transaction value for large companies involved in acquisitions is larger than the transaction value for small companies. During the period 1980–2003, the average transaction size of a transaction done
Size does matter—firm size and the gains from acquisitions on the Dutch market
283
Table 12.1 Transactions by year and acquirer size Year
Acquirer size Large
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 All
Small
All
No.
%
No.
%
No.
0 1 1 0 0 1 3 1 2 5 6 5 2 7 3 4 6 9 17 19 25 8 4 4
(0%) (50%) (50%) (0%) (0%) (100%) (100%) (50%) (100%) (71.4%) (46.2%) (55.6%) (25%) (77.8%) (75%) (50%) (42.6%) (52.9%) (65.4%) (41.3%) (53.2%) (40%) (22.2%) (36.7%)
0 1 1 0 0 0 0 1 0 2 7 4 6 2 1 4 8 8 9 27 22 12 14 7
(0%) (50%) (50%) (0)% (0)% (0)% (0)% (50)% (0)% (28.6)% (53.9)% (44.4)% (75%) (22.2%) (25%) (50%) (57.1%) (47.1%) (34.6%) (58.7%) (46.8%) (60%) (77.8%) (63.6%)
0 2 2 0 0 1 3 2 2 7 13 9 8 9 4 8 14 17 26 46 47 20 18 11
133
(49.4%)
136
(50.6%)
269 (100%)
Notes: The sample contains all completed M&A transactions between 1980 and 2001 with a Dutch buy-side that are listed on SDC. Only those transactions where the acquirer obtains full control of a public, private, or subsidiary target are included. Values in parentheses are percentages of total deals.
by a large company was almost 700% larger than the average transaction size of a small company. Table 12.2 indicates that the percentage of pure cash deals is very high. A common explanation for this is that European countries are more debt- than equity-oriented when it comes to corporate financing. The figures also show that the percentage of cross-border deals in the dataset is remarkably high. Almost 90% of the transactions done by large companies are international diversifying transactions. This finding can be explained by the fact that The
284
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 12.2 Transaction statistics
Transaction value Pure-cash deals Pure-equity deals Mixed deals Public target Private target Subsidiary target Diversifying acquisitions Cross-border acquisitions Tender offers
Small
Large
All
$131 m N ⫽ 136 75.7% N ⫽ 103 5.1% N⫽7 14.0% N ⫽ 19 10.3% N ⫽ 14 31.6% N ⫽ 43 55.9% N ⫽ 76 52.2% N ⫽ 71 63.2% N ⫽ 86 8.8% N ⫽ 12
$903 m N ⫽ 133 82% N ⫽ 109 8.3% N ⫽ 11 4.5% N⫽6 31.6% N ⫽ 42 15.8% N ⫽ 21 50.4% N ⫽ 67 33.1% N ⫽ 44 89.5% N ⫽ 119 22.6% N ⫽ 30
$513 m N ⫽ 269 78.8% N ⫽ 212 6.7% N ⫽ 18 9.3% N ⫽ 25 20.8% N ⫽ 56 23.8% N ⫽ 64 53.2% N ⫽ 143 42.8% N ⫽ 115 76.2% N ⫽ 205 15.6% N ⫽ 42
Notes: The transaction value is the total value of consideration paid by the acquirer excluding fees and expenses. Pure-cash (equity) deals are transactions that are 100% financed with cash (equity). Mixed deals are transactions that were financed with a mixture of cash and equity. The information about whether a target is public, private or subsidiary is obtained from the SDC database; the same applies to tender offers. A deal is considered as diversifying when the two-digit SIC codes of the acquirer and the target company differ. Please note numbers might not always add up to 100% because the category “other” is not shown in the table.
Netherlands is a small country, so that growth opportunities are sought outside the domestic area. Table 12.2 also shows that but a limited number of small companies use a tender offer to make a bid. The large companies in the dataset are almost three times more likely to launch a tender offer.
12.2.3
Methodology
In order to investigate a potential size effect and to test the five control hypotheses, a classical event study was performed. To measure the economic impact of an acquisition the abnormal returns during the announcement of this event were calculated. The abnormal return is the actual ex-post return of the security over the event window minus the normal return of the firm over the event window. The normal return is defined as the return that would be expected
Size does matter—firm size and the gains from acquisitions on the Dutch market
285
if the event had not taken place. The deviation from the normal return can be attributed to the announcement of an acquisition. The abnormal returns reflect the market’s assessment of the acquisition: Positive abnormal returns represent a positive reaction by the market, while negative abnormal returns reflect a negative market reaction. To calculate the abnormal percentage returns, we follow the market model approach of Brown and Warner (1985). The returns for the individual securities were obtained from Datastream, while we use the CBS-share price index as the market model benchmark return. This index is drawn up by the Dutch Central Statistic Agency (CBS) and reflects the price development of all shares listed at the Amsterdam Stock Exchange. The market model is given as R jt ⫽ α j ⫹ β j Rmt ⫹ ε jt
(12.1)
where: Rjt ⫽ the returns for security j at time t α ⫽ the model’s constant β j ⫽ the beta factor of security j given as COV(Rjt, Rmt)/VAR(Rmt) Rmt ⫽ the market returns at time t (CBS-share price index) εjt ⫽ the disturbance term for security j at time t
The coefficients αj and βj are estimated with OLS. The estimation takes places over a period from 138 days to 6 days prior to the announcement of the bid. This 6-months time frame (based on a 22-day trading month) was chosen because it represents a time frame long enough to get a solid estimation of the model’s betas without running the risk of calculating an outdated β. In the next step we calculate the abnormal returns Ajt: Ajt ⫽ rjt ⫺ aj ⫺ bj Rmt
(12.2)
The abnormal returns Ajt are the returns rjt achieved during the announcement period minus the estimated historical returns. In this model, aj and bj are OLS estimates of the parameters αj and βj from Eq. (12.1). To calculate the average abnormal returns at t ⫽ 0 for N acquirers the following equation (12.3) is used: A0 ⫽
1 N
n
∑ Aj 0 j⫽1
(12.3)
The abnormal returns are calculated over a three-day event window, starting one day before the announcement date and ending one day after the announcement
286
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
date. A three-day event window [⫺1, ⫹1] is chosen because it is unknown whether a transaction was announced before opening, during a trading day, or after closing of trading at the Amsterdam Stock Exchange. The cumulative abnormal returns (CARs) can then be calculated by using Eq. (12.4): CAR(⫺1, ⫹1) ⫽
1
∑ At
(12.4)
t ⫽⫺1
Finally, a student t-test is used to test whether the results significantly differ from 0.
12.3
Discussion of results
Table 12.3 gives an overview of the CARs over the period 1980–2003. The first column shows the CARs for all 269 companies in the dataset, columns 2 and 3 show the CARs separately for the 133 large and for the 136 small companies while the difference between the returns is presented in column 4. The top row of Table 12.3 shows that small companies earn significantly higher returns than large companies. Small companies earn a highly significant 2.65% upon announcement of a transaction while large companies earn an insignificant 0.20%. This result is in line with Moeller, Schlingemann, and Stulz (2004) who find 2.31% returns for small companies and 0.08% returns for large companies
Table 12.3 Cumulative abnormal returns by acquirer size
1980–2003 1980–1997 1998–2000 2001–2003 1998–2003
All (1)
Large (2)
Small (3)
Difference (2) ⫺ (3)
1.33%*** N ⫽ 269 0.85%* N ⫽ 101 1.97%*** N ⫽ 119 0.60% N ⫽ 49 1.62%*** N ⫽ 168
0.20% N ⫽ 133 0.42% N ⫽ 56 0.26% N ⫽ 61 ⫺0.90% N ⫽ 16 0.09% N ⫽ 77
2.65%*** N ⫽ 136 1.64%** N ⫽ 45 4.18%*** N ⫽ 58 1.33%*** N ⫽ 33 3.15%*** N ⫽ 91
⫺2.45%*** ⫺1.29% ⫺3.92%*** ⫺2.23%** ⫺3.06%***
Notes: A company is classified as large if it belongs to the 10% of the largest public Dutch companies by market capitalization one week prior to the day it announces a transaction; otherwise it is classified as small. *, **, *** Statistical significance at the 10%, 5%, and 1% level.
Size does matter—firm size and the gains from acquisitions on the Dutch market
287
on the U.S. market. The results presented in Table 12.3 imply that our hypothesis holds: Small companies earn significant larger CARs than large companies do when announcing an acquisition. However, these significant differences in returns could be influenced by market trends or other relevant events that are taking place. To ensure that this size effect is consistent over time, we split the sample into a number of subsamples and investigated whether the empirical findings are robust. The ranges for the subsamples were chosen based on events that took place on the M&A market. The first subsample covers the period from 1980 until just before the beginning of the Internet bubble years (1997). The second subsample covers the bubble years (1998–2000) in which the number of M&A deals was the highest. The third subsample captures the years after the Internet bubble until the end of 2003, while the fourth subsample is a combination of subsample two and three. Table 12.3 indicates that, when looking specifically at the period 1998–2000, the difference in returns between small and large companies is the largest. This significant difference is mainly caused by the large returns that small companies earned in these years (4.18%). The returns earned by large companies (0.26%) are not particularly small compared to the returns large companies earn in other periods. Apparently, the market assessed acquisitions done by small companies as very positive events during this period. After the heyday of M&A, it seems that skepticism rules the market. The combined announcement returns show a strong decrease during the years 2001–2003 to a level even lower than before the Internet bubble years. A reasonable explanation for this might be that many of the deals initiated during the bull market failed signally to create the promised returns. Our empirical findings provide evidence that the size effect on the Dutch market also holds over a number of different time periods. To make sure that these significant differences are independent of deal and target characteristics, the five control hypotheses from Section 12.2.1 are tested. First, we investigate whether this size effect exists just because small companies engage more often in pure-cash acquisitions. Therefore, we divide the acquisitions after the method of payment (mixed, equity, or cash) and calculate the CARs separately for small and large companies. Table 12.4 displays the CARs for each of the three methods of payment. The three columns on the right present the differences between the methods of payment. The bottom row shows the difference in returns between acquisitions announced by small and large companies for each method of payment. Table 12.4 indicates that, when looking at the whole sample, a mixture of equity and cash is the method of payment that results in the highest abnormal returns (2.26%), followed by pure cash (1.33%), and pure equity (0.01%). These results are in line with Moeller, Schlingemann, and Stulz (2004) who argue that, for both small companies and large companies, equity is the method of payment that results in the lowest abnormal returns upon announcement of a transaction. Small companies earn the highest returns when announcing a transaction when the method of payment is a mixture of cash and equity (3.14%). Whether small companies pay purely in cash or equity does not seem to have a large
288
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 12.4 Cumulative abnormal returns by acquirer size and method of payment Mixed (1) All
2.26% N ⫽ 25 Small 3.14%* N ⫽ 19 Large ⫺0.52% N⫽6 Difference 3.66%
Equity (2)
Cash (3)
All (4)
Difference
0.01% N ⫽ 20 2.34% N⫽7 ⫺1.44% N ⫽ 13 3.78%
1.33%*** N ⫽ 212 2.38%*** N ⫽ 103 0.33% N ⫽ 109 2.05%**
1.33%*** 2.25% N ⫽ 269 2.65%*** 0.80% N ⫽ 136 0.20% 0.92% N ⫽ 133 2.45%***
(1) ⫺ (2) (2) ⫺ (3) (1) ⫺ (3) ⫺1.32% 0.93% ⫺0.04% 0.76% ⫺1.77% ⫺0.85%
Notes: *, **, ***Statistical significance at the 10%, 5%, and 1% level.
impact on the announcement returns; the returns for equity (2.34%) are based on a very small number of cases, though. Large companies earn the highest returns when they announce a transaction paid for in cash (0.33%), followed by mixed payment (⫺0.52%), and equity payment (⫺1.44%). The largest difference in returns between small and large companies is found when the acquisition is financed with equity (3.78%), followed by mixed payment (3.66%), and cash (2.05%). Although these percentages are different from the values found by Moeller, Schlingemann, and Stulz (2004), the order of differences is the same. These empirical findings indicate that regardless of the method of payment, small acquirers earn higher returns than large acquirers. However, the difference between small and large acquirers is statistically significant only when the acquisition is paid in cash. It might well be the case that the empirical findings of a size effect are due to the fact that small companies more often engage in acquisitions with private targets. To test whether small companies earn significantly larger CARs independent of the organizational form of the target, we split our sample into three subsamples (private, public, and subsidiary). Table 12.5 displays the calculated CARs based on the organizational form of the target for both small and large companies. Table 12.5 shows that for the whole sample the largest returns are earned when a subsidiary target is acquired (1.64%). This result is in line with the empirical findings of Fuller, Netter, and Stegemoller (2002). Contrary to their results, our analysis shows that the acquisition of public targets results in higher returns (1.49%) than does the acquisition of a private company (0.59%). It is remarkable that this result is found for both large (0.33%) and small companies (4.83%). Based on previous research by Moeller, Schlingemann, and Stulz (2004), we would have expected higher returns when a small company acquires a nonlisted company. Subsidiary targets are for small companies the targets that result in the second highest abnormal returns (2.67%) while the announcement of a private target earns the lowest returns (1.03%). Large companies earn the highest returns
Size does matter—firm size and the gains from acquisitions on the Dutch market
289
Table 12.5 Cumulative abnormal returns by acquirer size and organizational form of the target
All Small Large Difference
Private (1)
Public (2)
Subsidiary All (3) (4)
0.59% N ⫽ 64 1.03% N ⫽ 43 ⫺0.30% N ⫽ 21 1.33%
1.49%** N ⫽ 56 5.86%** N ⫽ 14 0.03% N ⫽ 42 5.83%***
1.64%*** N ⫽ 143 2.67%*** N ⫽ 76 0.47% N ⫽ 67 2.20%**
Difference (1) ⫺ (2) (2) ⫺ (3) (1) ⫺ (3)
1.33%*** ⫺0.9% ⫺0.15% ⫺1.05% N ⫽ 269 2.65%*** ⫺4.83% 3.19% ⫺1.64% N ⫽ 136 0.20% ⫺0.33% ⫺0.44% ⫺0.77% N ⫽ 133 2.45%***
Notes: *, **, *** Statistical significance at the 10%, 5%, and 1% level.
when they announce the acquisition of a subsidiary target (0.47%), acquisitions of public targets result in the second highest announcement returns (0.03%), and the acquisitions of private targets are the transactions that earn the lowest returns (⫺0.30%). The bottom row of Table 12.5 shows the differences in abnormal returns between small and large companies for the different organizational forms of the target. It is not surprising that the difference in abnormal returns between small and large companies is the largest when a public firm is acquired (5.83%), followed by subsidiary targets (2.67%) and private firms (1.33%). These empirical findings indicate that independently of the organizational form of the target small companies earn significantly higher abnormal returns than large companies upon announcement of an acquisition. The results by Morck, Shleifer, and Vishny (1990) indicate that the CARs are higher for industry-related acquisitions than for companies that engage in industrydiversifying acquisitions. To test whether the size effect exists just because smaller companies acquire targets from the same industry, we divided the original sample of 269 transactions into industry-diversifying and nondiversifying deals. To determine whether a transaction is diversifying or not, we used the Standard Industrial Classification (SIC) Codes of both the target and the acquirer. A deal is considered diversifying when the two-digit SIC codes of the involved companies differ. Subsequently, the CARs for both types of transactions were calculated for small and large companies. Table 12.6 gives an overview of the difference in abnormal returns between diversifying transactions and nondiversifying transactions. The first column shows the returns earned for all the companies. The second and third column display separately the returns that are earned by large and small companies, while the most right column presents the difference in returns between small and large companies. Table 12.6 shows that the difference between the CARs for diversifying and nondiversifying deals is very small. While diversifying acquisitions earn 1.39%
290
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 12.6 Cumulative abnormal returns by acquirer size and nature of the deal
All Diversifying transactions Nondiversifying transactions
All (1)
Large (2)
Small (3)
Difference (2)⫺(3)
1.33%*** N ⫽ 269 1.39%*** N ⫽ 115 1.37%*** N ⫽ 154
0.20% N ⫽ 133 ⫺0.11% N ⫽ 44 0.37% N ⫽ 89
2.65%*** N ⫽ 136 2.61%*** N ⫽ 71 2.76%*** N ⫽ 65
⫺2.45%*** ⫺2.72%** ⫺2.39%***
Notes: A deal is considered a diversifying transaction when the two-digit SIC codes of the involved companies differ. *, **, *** Statistical significance at the 10%, 5%, and 1% level.
upon announcement, nondiversifying acquisitions earn 1.37%. This is not in line with research done by Morck, Shleifer, and Vishny (1990), who document that diversifying acquisitions in the United States earn significantly lower returns during the 1980s. Still, both large and small companies earn the largest CARs when they engage in nondiversifying transactions. The far right column of Table 12.6 shows that, once again, small companies earn significantly higher returns than large companies. The difference for diversifying transactions is 2.72%, and for nondiversifying transactions the difference is 2.39%. These empirical findings prove the existence of the size effect. Small companies earn, independent of the nature of the deal, significantly larger returns than large companies. In order to test whether the size effect is just due to the fact that small companies engage more often in domestic acquisitions, we made a distinction between international diversifying and domestic transactions. Whether a deal is international diversifying or not is determined by looking at the nation of the target. All deals in which targets are acquired that are not situated in The Netherlands are international diversifying deals. To get an even clearer picture, we divide the 205 cross-border deals also into continental, U.S., and intercontinental (non-U.S.) transactions. Table 12.7 shows the CARs sorted by acquirer size and geographic scope. The rows hold transactions with a different geographic scope, while columns 1, 2, and 3 present the CARs, respectively, for all deals, large company, and small company deals. The difference between small- and large-company transactions is presented in the far right column. Table 12.7 shows that the returns earned in domestic deals (1.87%) are larger than the returns earned in cross-border deals (1.19%) for all companies in the sample, as shown in column 1. This empirical finding is in line with the results found by Moeller and Schlingemann (2004). To get a clearer picture on the influence of potential cross-border effects, the international diversifying transactions are divided by geographic scope. It appears that continental deals are the transactions that result in the highest abnormal returns (1.71%). A possible explanation for this is that the European countries do not differ much from a cultural,
Size does matter—firm size and the gains from acquisitions on the Dutch market
291
Table 12.7 Cumulative abnormal returns by acquirer size and geographic scope
All Dutch Cross-border Continental U.S. Intercontinental (non-U.S.)
All (1)
Large (2)
Small (3)
Difference (2)⫺(3)
1.33%*** N ⫽ 269 1.87%** N ⫽ 64 1.19%*** N ⫽ 205 1.71%*** N ⫽ 101 0.66% N ⫽ 78 0.78% N ⫽ 26
0.20% N ⫽ 133 0.45% N ⫽ 14 0.18% N ⫽ 119 ⫺ 0.01% N ⫽ 42 0.37% N ⫽ 59 ⫺ 0.04% N ⫽ 18
2.65%*** N ⫽ 136 2.51%*** N ⫽ 50 2.72%*** N ⫽ 86 3.12%*** N ⫽ 59 1.58% N ⫽ 19 2.61% N⫽8
⫺2.45%*** ⫺2.06% ⫺2.54%*** ⫺3.12%*** ⫺1.21% ⫺2.65%**
Notes: Deals are considered cross-border deals when the target nation is any nation other than The Netherlands. *, **, *** Statistical significance at the 10%, 5%, and 1% level.
social-political, and economic point of view. Besides this, separate European national markets are becoming more and more integrated into one European market. These factors might result in a situation in which the market sees a crossborder European deal more like a domestic one, which has a higher chance to be successful. The intercontinental non-U.S. deals consist of deals that took place in countries like Canada, Australia, and Chile. These deals earn higher returns (0.78%) compared to U.S. deals (0.66%). The empirical findings in Table 12.7 indicate that large companies gain the most when they engage in domestic deals (0.45%). In comparison to small companies, their performance is the best in U.S. deals (0.37%) when the difference is relatively small (1.21%). Small companies enjoy the highest announcement returns when they undertake a nondomestic European deal (3.12%). These empirical findings provide evidence for the size effect. Small companies earn significantly higher returns than large companies independent of the geographic scope of the deal. Finally, we test whether the size effect is due to the mode of acquisition, i.e., due to the fact that small companies engage more often in tender offers, which result in significant positive returns. Therefore, we split the sample into two subsamples and calculated the CARs. Table 12.8 presents the empirical findings for tender offers and non-tender offers. When assessing the whole sample, tender offers earn smaller returns (0.65%) than non-tender offers (1.45%), despite the previous results by Jensen and Ruback (1983). A possible explanation could be that, usually, tender offers are unfriendly and as such do not fit with the typical Dutch consensus mentality. Although based on a limited number of deals, our empirical analysis shows that small companies do earn larger returns when announcing a tender offer
292
International M&A Activity Since 1990: Recent Research and Quantitative Analysis
Table 12.8 Cumulative abnormal returns sorted by acquirer size and mode of acquisition
All Tender offers Non-tender offers
All (1)
Large (2)
Small (3)
Difference (2)⫺(3)
1.33%*** N ⫽ 269 0.65% N ⫽ 42 1.45%*** N ⫽ 227
0.20% N ⫽ 133 ⫺0.15% N ⫽ 30 0.31% N ⫽ 103
2.65%*** N ⫽ 136 2.91% N ⫽ 12 2.63%*** N ⫽ 124
⫺2.45% ⫺3.06%* ⫺2.32%***
Notes: *, **, *** Statistical significance at the 10%, 5%, and 1% level.
(2.91%) compared to when they announce a non-tender offer (2.63%). On the contrary, large companies obtain better returns when they do not use a tender offer (0.31%) compared to when they do use this instrument (⫺0.15%). The far right column of Table 12.8 shows that the difference in returns between small companies and large companies is the largest for tender offers (⫺3.06%). The average difference between non-tender offers is based on a larger number of cases and is smaller (⫺2.32%). The empirical results in Table 12.8 prove that the existing size effect is not due to the mode of acquisition.
12.4
Conclusion
The objective of this chapter is to investigate a potential size effect on the Dutch M&A market. To determine if this phenomenon exists, we conducted a thorough empirical study to test whether there is a significant difference in cumulative abnormal returns between small and large companies when they announce M&A transactions. Based on earlier research done on the U.S. M&A market, we would expect that small companies earn larger returns than large companies when they announce an acquisition. Indeed, the empirical results from this study prove that small companies earn significantly higher returns than large companies upon announcement of a transaction. Over the 1980–2003 timeframe and for a total of 269 Dutch deals, small companies earned a cumulative abnormal return of 2.65% over the three-day event window around the announcement. Large companies earn just 0.20% over the same period. This result does not only hold for the whole timeframe of the dataset, but this finding is also consistent over a number of smaller timeframes. The question arises: Is this significant size effect independent of deal and target characteristics? To answer this question we tested five different control hypotheses. The empirical findings prove that the size effect exists independent
Size does matter—firm size and the gains from acquisitions on the Dutch market
293
of (i) the method of payment, (ii) the organizational form of the target company, (iii) the nature of the deal, (iv) the geographic scope, and (v) the mode of acquisition. Although the small sample size of some of the subsamples sometimes makes it hard to obtain significant results, we can conclude that the returns earned by small companies are significantly larger than the returns earned by large companies independent of any of the deal or target characteristics that were tested. The results of the empirical study do support the hypothesis that there is a size effect on the Dutch M&A market. Our conclusions are in line with the results found by Moeller, Schlingemann, and Stulz on the U.S. M&A market. For future research, we suggest an investigation of why small companies earn larger returns than large companies do when they announce an acquisition. A different ownership structure between small and large companies could be the source of more agency problems for large companies. Synergies might be harder to achieve for larger companies, or managers of large corporations might be more vulnerable to hubris than managers of small companies.
References Andrade, G., Mitchell, M., and Stafford, E. (2001). New Evidence and Perspectives on Mergers. Journal of Economic Perspectives, 15:103–120. Berkovitch, E., and Narayanan, M. P. (1993). Motives for Takeovers: An Empirical Investigation. Journal of Financial and Quantitative Analysis, 28:347–362. Bradley, M., Desai, A., and Kim, E. H. (1988). Synergistic Gains from Corporate Acquisitions and Their Division between the Stockholders of Target and Acquiring Firms. Journal of Financial Economics, 21:3–40. Brown, S. J., and Warner, J. B. (1985). Using Daily Stock Returns: The Case of Event Studies. Journal of Financial Economics, 14:3–32. Fuller, K., Netter, J., and Stegemoller, M. (2002). What Do Returns to Acquiring Firms Tell Us? Evidence from Firms That Make Many Acquisitions. Journal of Finance, 57:1763–1793. Jensen, M. C., and Ruback, R. S. (1983). The Market for Corporate Control. Journal of Financial Economics, 11:5–50. Moeller, S. B., and Schlingemann, F. P. (2005). Are Cross-Border Acquisitions Different from Domestic Acquisitions? Evidence on Stock and Operating Performance for U.S. Acquirers. Journal of Banking and Finance, 29: 533–564. Moeller, S. B., Schlingemann, F. P., and Stulz, R. M. (2004). Firm Size and the Gains from Acquisitions. Journal of Financial Economics, 73:201–228. Morck, R., Shleifer, A., and Vishny, R. W. (1990). Do Managerial Objectives Drive Bad Acquisitions? Journal of Finance, 45:31–48. Travlos, N. G. (1987). Corporate Takeover Bids, Methods of Payment, and Bidding Firms’ Stock Returns. Journal of Finance, 42:943–963.
This page intentionally left blank
Index Note: Page numbers accompanied by n refer to footnotes. A Abbey National plc, 152, 153 Abnormal bond returns, 129, 130n Abnormal return, 8, 136, 181, 185, 188, 271, 280, 284, 285, 289 average abnormal return (AAR), 54, 188, 285 cumulative abnormal returns (CARs), 280, 281, 289 long-run abnormal return, 263, 275 Accounting accruals, 264 definition, 264 United States and Spain comparison of, 263 Accounting disclosure and treatment acquisitions details, 150–153 annual accounts and reports, 154–156 creditors’ disclosures, 153–154 deal, tracing, 148–150 Accruals pattern, 267 abnormal components, 268–270, 274 current accruals, 267–268 Acquirer cash reserves, 86, 105 Dutch acquirer, 279 growth of, 72–74 leverage, 86, 101–105 short-term performance, 61, 62, 64, 68, 72 valuation of, 64–67 value-destroying acquirers, 62, 63, 71, 72, 73 value-enhancing acquirers, 51 Acquiring countries, 38–39, 40 Acquiring firms, 93n, 173, 261 abnormal current accrual, 265–266 normal current accrual, 265, 266 Acquisition targets, in IPO market predictors, 170 Acquisitions, 195
Agency costs, 58–59, 239, 240 Agency theory, 262 Airtel, 89 Alberta (AB), 214 Allied Leisure plc, 149 Amadeus Extended Database, 88n Amsterdam Stock Exchange, 285, 286 Anglian Water plc. See AWG plc Anglo American plc, 149, 155 Argentina, 39 Arlen plc, 154 Ash & Lacy plc, 154 Asia, 4, 5 AT&T, 229 AustralAsia (AAS), 33, 34, 36 VC exits, 203 Australia, 16, 30, 33, 39, 81, 291 Average abnormal return (AAR), 54, 188, 285 AWG plc, 149, 154, 155 B Behavioral agency model, 224, 254 on managerial risk taking, 236–238 Belgium, 30, 31 BellSouth, 229 Berkeley Group plc, 149 Bid premium, 67–69 Bidder bondholders, 120, 124, 129, 131 cash-for-stock versus stock-for-stock, 124 Bidder shareholders, 8, 185 Bidders, performance of, 54–58, 62, 64 Bidder’s perspective, 146–148 deferring cash payable, 146 in loan notes issuing costs, 146–148 Bilton plc, 150 Bloomberg, 25, 52 Bolsa de Madrid, 264
296
Bondholders of bidders, 120, 124 governance and legal standards, cross-country differences, 125–131 cross-border deals, 126–128 European study, 128–131 relevance of, 125–126 of investment-grade targets, 124 of junk-grade targets, 120, 124 theory and empirics, 118–125 background, 118–120 in United States domestic deals, 120–125 Bootstrap technique, 271 Brazil, 39 British Columbia (BC), 213 British Fittings Group plc, 151 BSS Group plc, 154 Bull market, 4, 5, 6, 7, 52, 287 Bunzl plc, 150, 151 Business cycle and merger waves, 29, 30, 45 Buybacks, 15, 195, 196, 197, 202, 203, 204, 215 C Calendar-time portfolio approach, 271 of stock returns performance, 270–272 Canada, 2, 30, 39, 195, 203, 213, 291 corporate funds, 199, 201 foreign funds, 199, 200 government funds, 199, 200 hybrid funds, 199, 200 investment size, 204, 216–217 labour-sponsored venture capital corporations (LSVCCs), 200 private independent funds, 199, 200 VC exits, 195 acquisitions, 195 buybacks, 196 initial public offerings (IPO), 195 outcomes, 204, 205, 212–217 secondary sales, 195–196 theory and evidence, 196–199 write-offs, 196 Canadian Venture Capital Association (CVCA), 199, 201 Capital gain, 145, 147, 156, 157, 158, 195
Index
Cash, 26, 58, 61, 62, 80, 91, 105, 117, 119, 126, 137, 141, 146, 155, 156, 227, 283, 284 versus equity, 101 versus stock, 81, 86 versus stock-financed deals, 124 compensation, 232 and performance, 230 popularity of, 136–137 Cash flow measures, 91–94, 97 Cash, loan notes and ordinary shares, 139, 141 Cash and loan notes, 139, 141 Cash and ordinary shares, 138, 141 Cash-financed acquisitions, 62 Cash-financed deals, 13, 129 Cater Allen Holdings, 152, 153 Central Statistic Agency (CBS), 285 share price index, 285 CEO, 227, 229, 234, 239–240, 241, 244, 246, 249, 251, 253 behavioral biases. See Manager’s behavioral biases compensation and performance, 229–230 and S&P 500 index, 228 in United States, 227 Change model, 94, 99, 113 versus intercept model, 100 Chargeable gain, 145, 156, 158 Chile, 291 China, 5, 25, 31, 39 C.K. Coffee Ltd., 89 Clustering, of M&A, 5–7 hubris and herding hypotheses, and agency problem, 6–7 managerial market timing model, 7 neoclassical models, 5–6 Coburg Group, 89 Continental Europe, 1, 4, 81, 88, 118, 129, 131 Combined firm, profitability of, 80, 93–94, 95, 113, 114 change model, 94, 99, 113 intercept model, 94, 99, 113 Comisión Nacional del Mercado de Valores (CNMV), 264 Corporate acquisitions, 223 and corporate governance, 240–249
Index
executive compensation, 238–240 integrated framework, 249 reviews, 250–251 and managerial overconfidence, 244–246, 248, 249 managers’ behavioral biases and risk taking, 236–238 shareholder and managerial interests, risk alignment, 225–236 excessive risk taking, 234–236 executive compensation, 226–229 executive pay and shareholder value, link, 229–230 large-shareholder monitoring, 230–231 misalignment, 231–234 traditional agency model, 225–226 theoretical and empirical issues, 251–253 Corporate bond market, 4 Corporate VC funds, 200, 204 Corporate governance managerial risk taking model, 240 mechanism, 224, 225–226, 248–249, 250 role for, 239–240 Corporate takeovers profitability changes, 99–113 Creditors’ disclosure, 153–154 Cross-border acquisitions, 4, 12, 87 Cross-border deals, 14, 29, 117, 118, 126 creditor protection, 126–128 versus domestic deals, 74, 87–88, 106–107 Cross-border M&As basic characteristics, 25–30 and countries, 38–40 in 2005, 30–32 features, 46 individual firms, 45–46 inequality, 41–45 international economics, guide, 23 inter-regional connections, 37 Lorenz curves, 42 regional distribution, 32–38 values, 41 Cross-border versus domestic acquisition, 74
297
Cross-sectional regression model, 189 Cumulative abnormal returns (CARs), 280, 281, 289 by acquirer size, 286 and acquisition mode, 292 and deal nature, 290 and geographic scope, 291 and organizational form, of target, 289 and payment method, 288 calculation, 286 in small and large companies, 280–281 Cumulative average abnormal return (CAAR), 54, 55, 188 Cumulative average standardized abnormal returns (CASAR), 188 Current accruals, 264n, 265 of Spanish mergers, 267–268 unobservable components, 265 Czech Republic, 31 D DBS Management plc, 152 Deals, 14, 88, 120, 137, 139, 140–141, 142, 289 announcement, indicators, 54, 55 breakdown, by country, 52, 53 breakdown, by year, 52, 53 cash versus stock-financed, 124 cross-border, 14, 117, 118, 126–128 diversifying versus nondiversifying, 124 domestic versus cross-border, 74, 87–88, 106–107 hostility versus friendly, 86, 101, 103 involving loan transactions, 144 motives for, 58–59 tracing details, 148–150 United States, 120–125 Debt capacity, 58, 70, 118–119, 120 Debt financing, 51, 62, 81, 119 Delta value, 235 Denmark, 30 Deutsche Boerse AG, 172 Diversifying acquisitions, 86–87 versus nondiversifying mergers, 64–65 versus nondiversifying deals, 289–290 versus nondiversifying transactions, 124
298
Dividend yield, 67 Domestic versus cross-border deals, 87, 106–107, 108 Double-exit strategy, 14, 169 Dual-tracking firms, 171 Dutch acquirer M&A transactions, 282 between 1980 and 2001, 283 statistics, 284 Dutch market, acquisitions, 279 empirical analysis, 280–286 data, 282–284 event study approach, 284–286 hypotheses, 280–281 results, 286–292 E Earnings management abnormal accrual component, 265–266 agency theory of, 262 measurement, 264–267 normal accrual component, 265 and post-merger stock price performance, 270–275 East Asia and Pacific (EAP), 32, 33, 34 East Europe and Central Asia (ECA), 32, 33 EBITDA, 80, 91–92, 95–97, 189 EBITDA-to-assets ratio, 91 Egypt, 30, 31, 33 Elf Aquitaine, 89 Empirical evidence of bondholders in United States domestic deals, 120, 123–124 on M&A profitability, 7 English insolvency law, 126 English law, 128n Enron, 150 Equity and cash combination, 61, 287 Equity-financed deal, 62, 77 Equity-based compensation, 227, 230, 234, 241, 242, 248 Eurobonds, 128n, 130n Europe, 33, 34, 36, 41 bondholder wealth effects, 128–131 VC exits, 203 European corporate bond market, 128
Index
European Insolvency Regulation (EIR), 127 European M&A, 1, 5, 88, 89, 118 bondholder wealth, 128–131 long-term operating performance in, 79 European Union (EU), 36, 127 Event-time analysis of stock returns performance, 270–271 Evode plc, 145n Excess return, 59 calculation, 54 short-term market, 59–60 Exchange FS Group plc, 149 Executive compensation, 59, 226 acquisition impact, 243–248 and CEO behavioral biases, 227, 228 corporate acquisition, 240–249 endogeneity problem, 248–249 impact, 243–248 integrated framework, 249 fixed compensation, 226 long-term incentive plan (LTIP), 227 and overconfidence, 238–240 corporate governance, 239–240 risk-taking model, 240 risk-taking behavior, 239 S&P 500 index, 227, 228 short-term incentive plan, 227 Executive pay and shareholder value, link pay and performance, 229–230 Expected utility maximization (EU) model, 237n External block (share) holders, 226, 231, 240 External financing cost, 58 F Fama–French model calendar-time regression, 271, 272 Fifth M&A wave, 3–5, 241 Finance theory, 118 asset risk versus financial risk effect, 119 co-insurance effect, 119 Financial risk effect, 119 Financial synergies, 58, 86, 87 Financial Times, 186, 236 Firm characteristics, 172, 175 in IPO versus acquisitions, 197–198
Index
Firm performance, 254 large-shareholder monitoring, 230–231 Firm size effect in Dutch market, 279 in United States market, 280 First M&A wave, 2 Fixed compensation, 226, 227, 232 Focus versus diversifying acquisitions, 86–87, 105–106 Forecasting targets, 52 Foreign direct investment (FDI) distribution, types, 24 and firm heterogeneity, 45–46 horizontal FDI, 28, 29, 31, 39 vertical FDI, 31, 40 Foreign funds, 200 Fourth M&A wave, 3 France, 20, 30, 33, 38, 39, 52, 126 Free cash flow theory, 86, 105 French civil-law, 125 G Germany, 2, 15, 30, 33, 39, 41 data and sample, 172–174 empirical study/regression analysis, 174–177 firm characteristics, 172–174, 173 IPO markets, 169 Gini coefficient, 13, 42–44 Glaxo Wellcome plc, 141 Global Mergers and Acquisitions Database, 4, 5, 25 Global regions, 13, 32–33, 37, 38, 47 Governance and legal standards, crosscountry differences cross-border deals, 126–128 European study, 128–131 relevance of, 125–126 Government funds, 200 Greene King plc, 149 Greenfield FDI, 24, 40 H Hartlepool Water plc, 154 Herding, 5–6 Hewlett-Packard, 229 High-income regions, 33, 34, 37, 39 Hill & Smith Holdings plc, 154 Hillsdown Holdings plc, 155
299
Historical background, of M&As fifth wave, 3–5 first wave, 2 fourth wave, 3 second wave, 2 sixth wave, 5 third wave, 2 Hobson plc, 155 Home Depot, 229 Hong Kong, 30, 262, 263 Horizontal FDI, 31 cross-border M&A, 28 Hubris, 59, 64, 77, 236, 246n Hunters Armley Group plc, 146 Hybrid funds, 201 I Imperial Metals Industries (IMI) plc, 146 Indexation allowance, 158, 160 Industry relatedness focus versus diversifying acquisitions, 86–87, 105–106 Intercept model, 94, 99 versus change model, 100 Initial public offerings (IPO) in Germany, 169, 172 data and sample, 172–174 empirical study, 174–177 financial research on, 170 firm characteristics, 172–174 and mergers and acquisitions, 169 takeover targets, 170–172 Israel, 30, 31 Italy, 30, 39 J James Crosby Group plc, 149 Japan, 30, 31, 39, 125 Jensen’s alpha, 272 Jones model, 265, 269, 273 Jurisdiction shopping, 127–128 K Kidder Peabody, 238 L Labour-sponsored venture capital corporations (LSVCCs), 200, 201, 202, 203, 204, 209, 213, 215, 216, 217
300
Laporte plc, 145n Large companies, 160, 286n, 287, 288–289 and CARs, 280–281, 286 Latin America and Caribbean, 33 Leverage, 70, 72, 77–78, 101, 103, 120 Lloyds Bank plc, 147 Loan notes, 144 accounting treatment, 148–156 bidder’s perspective, 146–148 disclosure of, 150–154 mix-and-match facilities, 135 target shareholder’s perspective, 145–146 tax-choices, 156 modeling, 157–161 versus ordinary shares and cash, 144 Logistic regressions, 174–175 London Stock Exchange (LSE), 137, 182, 186, 188 Long-run abnormal return, 263, 275 Long-term performance, 54, 59, 76 cash payment, 61–62 highest and lowest premium, 68 measurement, 189–190 overseas acquisition, 74 Long-term returns, 56, 62, 76 Long-term track record, 56–57, 67 Lorenz curves, 42 Lucent Technologies, 229 Luxembourg, 30 Luxembourg Stock Exchange, 128n M M&G Group plc, 149 Mann-Whitney test, 103, 106, 107, 110 Management efficiency and operating performance, 70–72, 73 Managerial hubris models, 6 Manager’s behavioral biases. See also CEO corporate governance, 239–240 corporate performance hubris, 236, 237, 238 overconfidence, 240–250 overoptimism, 236, 237 self-confidence, 238, 238 benefits, 238
Index
and executive compensation, 236–237, 238–240 impact on risk taking, 237–238, 239 model, 240 Managerial shareholdings, 233 Manitoba (MB), 216 Mannesmann, 30, 41, 89 Market capitalization, 52, 62–64, 173–174 Market index, 54, 56, 188 Market model, 187–188, 285 for abnormal percentage return, 285 Market reactions, 60 long-term, 56–57 overseas acquisition, 74 short-term, 56–57, 59–60, 76, 77 Market value of assets, 93n, 99 Market-to-book ratio, 174 Marlborough Stirling plc, 149 Median post-acquisition performance, 94, 102, 103, 104, 108 Merck, 229 Merger and Acquisition Database, 282 Method of payment cash, 62, 81, 101, 136–137, 281 equity, 62, 101 share, 135 stock, 81 Middle East and North Africa (MNA), 33, 34, 36, 47 Misys plc, 152, 153 Mix-and-match facilities, in acquisitions characteristics cash, loan notes and ordinary shares, 139, 141 cash and loan notes, 139 cash and ordinary shares, 139, 141 deal sizes, 139, 140–141 take-up of alternatives, 141–143 data sources and sample selection, 137–138 disclosure and treatment, 148–156 literature review, 136–137 loan notes, 144 Mixed deals, 284n Morland plc, 149 Morrison plc, 149, 155 Motives of managers, 59
Index
for reverse merger, 184 for takeover deal agency costs reduction, 58–59 synergies creation, 58 Multiple bidders, 69, 70 Multivariate analysis, 107, 111–112, 113 N Negative abnormal return, 262, 263, 271, 285 Neoclassical model of M&A clustering, 5–6 Net acquiring countries, 38–39 Net operating loss (NOL), 58 Net target countries, 38–39 Netherlands, The, 30, 39, 126, 282, 284, 290 New Brunswick (NB), 214 Newfoundland (NF), 214 Nondiversified mergers versus diversified mergers, 64–65 Nonexecutive directors, 226 Normal return, 284–285 North America (NAM), 13, 33, 34, 36, 37, 184 Northern Territories (NT), 197, 214 Norway, 30 Nova Scotia (NS), 214 O Ontario (ON), 214 Operating performance, in European M&A, 8–12, 191, 192 changes in, 95–99 definition, 91 deterioration, 87 and management efficiency, 70–72, 73 measurement, 189–190 measures of, 80 cash flow, 91–94 results of analysis, 95–99 median changes in, 94, 102, 103, 104, 106, 107, 108 peer companies, selection of, 89–91 sample description, 89, 90 selection, 82–85, 88–89 Operational synergy
301
diversified versus nondiversified mergers, 64–65 Outperforming acquirers. See valueenhancing acquirers Overconfidence, 223, 238, 240, 246n, 248, 254 Overoptimism, 236, 246n, 254 Overseas acquisition, 74 P Pan-European S&P/Citigroup BMI, 54 Paramount, 247 Pay for performance, 229–230, 248 Pay without performance, 229–230 Payment method, 60–62, 281, 288 cash versus equity, 101 cash versus stock, 81–86 Peer companies, selection of, 89–91 Pfizer, 229 Plasmec plc, 154 Polypipe plc, 147 Pooling of interest, 92n, 93n Positive abnormal return, 285 Post-acquisition performance, 81 changes in profitability, 95–99 prior research, overview, 81 determinants, 81–88, 99–113 empirical studies, 81, 82–85 sample selection procedure, 82–85, 88 See also Takeovers Post-acquisition profitability, 93, 99, 105 Post-merger profitability, 80, 101, 106, 107, 109 changes assessment, 94 takeover characteristics, 99, 113 Post-merger returns, 270 and previous earnings management, 272–275 Post-merger stock price performance and earnings management previous earnings management, 272–275 stock returns performance, 270–272 Poveda model, 270n Pre-acquisition leverage, 101–105 Pre-acquisition profitability, 92–93, 99 Primesight plc, 150 Prince Edward Island (PEI), 214
302
Private independent funds, 201 Private targets, 169, 171 for small companies, 288, 289 Profitability, of M&A empirical evidence, 7 operating performance, 8, 12 short-term effects, 8, 9–11 Provend Group plc, 150 Prudential plc, 149 PTS Group plc, 154 Public targets, 185 for small companies, 288 Purchase method, 92n, 93n Pure-cash deals, 284n Pure-equity deals, 284n Q Quebec (QC), 197, 214 R Recent activity, in M&A in the 1990s, 3–5 from 2003, 5 Regression analysis, 175, 176n, 177, 270, 272, 273 of acquisition, 174–177 Retail Price Index, 158, 160n Return on assets (ROA), 266, 273 Reverse mergers abnormal stock price reactions, 185 advantages, 182 attraction of, 183–184 cross-sectional regression model, 189 results, 190–192 drawbacks, 184 event study methodology, 187–188 results, 190–192 hypotheses, 185–186 long-term performance measurement, 189–190 mechanics, 183–185 motives for, 184 sample description, 186–187 in United Kingdom, 181 in United States, 183 Rhone-Poulenc, 89 Risk alignment corporate governance mechanism, 225–226 managerial stock ownership, 225–226
Index
shareholder and managerial interests, 225–236 traditional agency model, 223, 225–226 Risk incentives, measures of, 234–236 Risk preference compensation contracts, 231 Robustness check, 98–99, 105n Russia, 30, 32 S S&P 500 index, 227, 228, 235n, 241, 244, 246 CEO compensation, 227–229 equity-based compensation, 230, 241, 242 Delta and Vega value, 234–236 overconfidence in corporate acquisitions, 244–246 Safeway plc, 138, 229 Sample selection procedure European acquisitions, 82–85, 88–91 Saskatchewan (SK), 214 Scottish Media Group plc, 150 SDC platinum database, 137 Second M&A wave, 2 Secondary sales, 195–196, 204, 212 Securities Data Company (SDC), 88, 186, 282 Share price weakness, 56, 62, 77 Shareholder and managerial interests, risk alignment, 225–236 excessive risk taking, 234–236 executive compensation, components, 226–229 executive pay and shareholder value, link, 229–230 large-shareholder monitoring, 230–231 misalignment in, 231–234 traditional agency model, 225–226 Shell company, 182, 184 Short-term market reactions, 59–60, 76 Short-term performance, 62 of acquirers, 61, 62, 64, 68, 72 Short-term returns, 54, 55, 76 Short-term wealth effects, 8, 9–11 Sistema de Interconexión de las Bolsas Españolas (SIBE), 264 Sixth M&A wave, 5 Slough Estates plc, 150
Index
Small companies, 282, 284, 286n and CARs, 280–281, 286–287, 288–289, 290, 291–292 SmithKline Beecham plc, 141 South Korea, 31 Spain, 30, 39, 261 accruals pattern, 267 abnormal accrual analysis, 268–270 current accruals, 267–268 acquiring firms in, 263, 267, 268 earnings management, measurement, 264–267 post-merger stock price performance previous earnings management, 272–275 stock returns, 270–272 stock-financed corporate mergers, 261 sample and data, 264 St Ives Group plc, 146 Standard event study, 187 market model, 187–188 results, 190–191 Standard Industrial Classification (SIC) Codes, 289, 290n Standardized abnormal returns, 188 Stock options dark side, 234–236 and managerial risk taking, 233, 234 Stock returns performance, 270–272 calendar-time methodology, 270, 271–272 event-time analysis, 270–271 Sub-Saharan Africa (SSA), 34 Subsidiary targets for large companies, 288, 289 for small companies, 288, 289 Swallow Group plc, 150–151 Sweden, 30, 31, 40 Switzerland, 30, 39 T T-statistics, 59, 63, 66, 71, 72, 73, 74, 94, 130, 141, 173, 189, 267 market capitalization factor, 62–64 value-destroying acquirers, 62 value-enhancing acquirers, 62 Takeovers acquirers leverage, 101–105 acquirer’s cash reserves, 105 deal atmosphere, 101
303
domestic versus cross-border deals, 106–107, 108 industry relatedness, 105, 106 multivariate analysis, 107, 111–112, 113 payment method, 101 univariate analysis, 107, 109–110 targets, 51–52, 72, 176 prediction, 170–172 size, 106 transactions, 59, 67, 106, 136 valuation, 54–58 Taper relief, 158–159 Target countries, 38–39, 40 Target size, 62–64, 87, 106, 107 valuation, 67 Target shareholders bidder shares, 143 cash, 143, 156–157 and shares, 156, 157 cash or loan notes, 157 considerations offered, 137, 138, 139, 140 deal size, 140, 141 interest rate, 145 loan notes, 145–146, 157, 160 repayment, 145 shares versus loan notes, 143 tax management, 145–146 Tarmac plc, 149, 155 Tax benefits, 58 Tax choices, loan notes, 156 modeling, 157–161 corporation tax, 159–160 tax liability, 156, 159, 160 Taxation of Chargeable Gains Act (TCGA) 1992, 156 Taxmodels individuals and corporate bodies, 157–161 Taxpayers, 157 corporate bodies, 157, 158, 159–160, 161 individuals, 157, 158, 159, 160–161 Tech, 57 Telcos, 57 Third M&A wave, 2 Thomson Financial Securities Data Corporation (SDC), 52, 186
304
Time Warner, 229, 247 TotalFina, 89 Traditional agency model, 223, 225 corporate governance mechanism, 225, 226 managerial stock ownership, 225, 226 Transaction value, 52, 241, 242, 243, 282, 284n Transactions continental, 290–291 cross-border, 290, 291n diversifying, 289, 290 domestic, 290, 291 intercontinental, 291 nondiversifying, 289, 290 in United States., 290 Turkey, 31 U Underperforming. See value-destroying acquirers United Arab Emirates, 30, 31 United Kingdom, 30, 39, 129, 137, 226, 251 fifth M&A wave, 4, 29–30 LTIP share awards, 232n reverse mergers, 181 targets and private acquirers, 181 third M&A wave, 2 United States, 30, 39, 226, 251, 263 bondholder wealth effects, 120, 121–123 cross-border target, 126 domestic deals, 120–125 fifth M&A wave, 4 IPO markets, 174 reverse mergers, 182, 183 third M&A wave, 2 VC exits, 203 Univariate analysis, 107, 109–110 U.S. VC index, 202n V Valuation acquirer’s valuation, 64–67 bid premium, 67–69 multiple bidders, 69, 70 target’s valuation, 67 Value-Added Tax (VAT), 144
Index
Value-destroying acquirers, 62, 63, 71, 72, 73 Value-enhancing acquirers, 51 characteristics affecting factors, 59–69 combination, of facotrs, 74–77 non-affecting factors, 69–74 motives for acquisitions, 58–59 takeover value creation, 54–58 Vega value, 235, 243, 251 Venture capitalists (VCs), 15, 171, 176 in Canada, 199–202 investment, 1991–2004, 203–217 corporate funds, 200 exits, 196, 206–211 acquisitions, 195 buybacks, 196, 197 initial public offerings (IPO), 195, 197, 198 secondary sales, 195–196 write-offs, 196, 197 foreign funds, 200 government funds, 199, 200 hybrid funds, 200 labour-sponsored venture capital corporations (LSVCCs), 200 private independent funds, 200 role, 198 VC-backed IPOs, 196–197 Viacom, 247 Verizon Communications, 229–230 Vertical FDI, 31 Vodafone, 30, 41, 89 W Wal-Mart, 230 Wall Street Journal, The, 229 Waterfall Holdings plc, 149 Western Europe (EUR), 13, 33, 34, 37, 38 Whitbread plc, 150 “Winner’s curse”, 59, 69, 77 Wilcoxon signed rank test, 94, 97, 98, 102, 103, 104, 106, 108, 110, 189, 267, 268, 269 WM Morrison Supermarkets plc, 138 Wolseley plc, 151 Working capital, 91, 92, 97n, 98 Write-offs, 15, 196, 197, 204