Scheller Performance of Corporate Acquisitions over the Medium Term in Germany

Transcription

1 Scheller Performance of Corporate Acquisitions over the Medium Term in Germany

2 GABLER EDITION WISSENSCHAFT

3 K. Randolf Scheller Performance of Corporate Acquisitions over the Medium Term in Germany With a foreword by Prof. Dr. Gunter Dufey Springer Fachmedien Wiesbaden GmbH

4 Die Deutsche Bibliothek - ClP-Einheitsaufnahme Scheller, K. Randolf: Performance of corporate acquisitions over the medium term in Germany / K. Randolf Scheller. With a foreword by Gunter Dufey. - Wiesbaden: Dt. Univ.-Verl.; Wiesbaden: Gabler, 1999 (Gabler Edition Wissenschaft) lug!.: Koblenz, Wiss. Hochsch. fur Unternehmensfuhrung, Diss., 1998 ISBN ISBN (ebook) DOI / Aile Rechte vorbehalten Springer Fachmedien Wiesbaden1999 Urspriinglich erschienen bei Betriebswirtschaftlicher Verlag Dr. Th. Gabler GmbH, und Deutscher Universitcits-Veriag, Wiesbaden GmbH, Lektorat: Ute Wrasmann / Marcus Weber Das Werk einschliel3lich aller seiner Teile ist urheberrechtlich geschutzt. jede Verwertung auf3erhalb der engen Grenzen des Urheberrechtsgesetzes ist ohne lustimmung des Veri ages unzulcissig und strafbar. Das gilt insbesondere fur Vervielfciltigungen, Ubersetzungen, Mikroverfilmungen und die Einspeicherung und Verarbeitung in elektronischen Systemen. Hochste inhaltliche und technische Qualitcit unserer Werke ist unser lie!. Bei der Produktion und Verbreitung unserer Werke wollen wir die Umwelt schonen. Dieses Buch ist deshalb auf saurefreiem und chlorfrei gebleichtem Papier gedruckt. Die Einschweil3folie besteht aus Polyathylen und damit aus organischen Grundstoffen, die weder bei der Herstellung noch bei der Verbrennung Schadstoffe freisetzen. Die Wiedergabe von Gebrauchsnamen, Handelsnamen, Warenbezeichnungen usw. in diesem Werk berechtigt auch ohne besondere Kennzeichnung nicht zu der Annahme, dass solche Namen im Sinne der Warenzeichen- und Markenschutz-Gesetzgebung als frei zu betrachten waren und daher von jedermann benutzt werden durften.

5 FOREWORD Merger and acquisition (M&A) activities of business firms have provided a challenging subject of research to scholars in economics and business for many years. Dr. K. Randolf Scheller's work presents an interesting contribution to the growing body of research in this area. It's focus is on the market for corporate control in Germany which has become much more active. In part, this happened as a result of phenomena that reflect the globalization of markets, both with regard to increased competition and technological change in markets for goods and services as well as financial assets, particularly for equity (risk) capital. These trends in turn have prompted a number of changes in the German legal and regulatory environment, the most important are the takeover code, a selfregulatory measure aiming primarily at preempting legal provisions protecting minority shareholders. By the same token, a number of related laws have been recently passed in Germany, in part prompted by European harmonization initiatives, that change the legal environment for M&A activities. Studies of the value creating/destroying effects of M&A activities provide important Signals to managers and investors about possible (negative) outcomes of such activities, providing some guidance for managerial action. At the same time, sound empirical results further our understanding of the functioning of markets in general, in this case the market for control rights over corporate assets. v

6 Using event study methods, Dr. Scheller reports empirical findings of abnormal (small but statistically significant) positive returns for the target companies at the announcement date, but (large, significant) negative cumulative abnormal returns over the long run (5 years). While these findings are consistent with those reported in both the US and German literature, the author provides additional results that are of interest: (a) smaller stakes obtained by the acquiring company correlate with higher negative returns; (b) acquisitions by companies in unrelated industries are associated with larger negative returns for target company shareholders over the long-run; (c) smaller acquisition target companies performed worse than larger ones. Scheller's work addresses an important issue in the M&A business. By subjecting a unique data set to rigorous empirical analysis, he arrives at interesting, counter-intuitive results. It has been a pleasure to work with the author in this study; we wish him and his publication much success. G.Oufey Visiting Professor, WHU Koblenz Otto Beisheim Graduate School Chair for International Corporate Financial Management Professor of International Business and Finance University of Michigan Business School, Ann Arbor, MI VI

7 AUTHOR'S FOREWORD The study presented in this book has been accepted as my Ph. D. dissertation bearing the title "The German Market for Corporate Control: Performance of Corporate Acquisitions in the Medium-Term" at the Wissenschaftliche Hochschule fur Unternehmensfiihrung (WHU Koblenz), Otto Beisheim Graduate School of Management, in Koblenz, Germany in November In the process of realization, I benefited from a lot of help and from considerable support from many interlocutors, both from inside and outside the WHU. It is always difficult to express one's feelings in words without loosing meaning; but for all those, who have supported me morally, intellectually or have contributed to this work, I would like to say this little word, perhaps banal, but full of significance: "Thanx". I have learned a great deal from Prof. Dr. Gunter Dufey, my Ph.D. advisor. The insight, knowledge, constructive criticism, and experience that he shared with me has made me look at things differently and, thus, has hopefully helped me become a more versatile researcher. lowe a special "Dankeschon" to Prof. Dr. Arnd Huchzermeier, co-advisor of my Ph.D. committee, who has made me think more rigorously by his critical reading and helpful comments. I also appreciate his overwhelming support and encouragement to conduct the foregoing study besides my job as an investment banker. vii

8 Last but not least, I would like to express my gratitude to Eva for all her love, support, and understanding during the long hours which a study like this entails. K. Randolf Scheller viii

9 TABLE OF CONTENTS FOREWORD... V AUTHOR'S FOREWORD... VII LIST OF TABLES, FIGURES, AND GRAPHS... XV LIST OF V ARIABLES... XIX 1. INTRODUCTION AND MOTIVATION THE GERMAN STOCK MARKET, THE GERMAN M&A ENVIRONMENT AND ITS CORPORATE LEGAL AND REGULATORY STRUCTURE Introduction Overview of the German Stock Market Distribution of Ownership of German Public Companies German Corporate Laws as They Pertain to M&A Transactions German Takeover Guidelines Implications ISSUES OF CORPORATE GOVERNANCE Agency Theory The Shareholder-Value-Maximizing Hypothesis ix

10 3.3. The Growth-Maximizing Hypothesis Issues of Corporate Governance related to German M&A Transactions LITERATURE REVIEW Introduction Overview of the Kinds of M&A Research Available Capital Market Oriented Research Sample Analysis Approach Types of M&A Transactions Specifications of Statistical Models The Market Model and Variants thereof The Capital Asset Pricing Model (CAPM) Adjusted Models Multiple Factor Models Problems with Estimating and Interpreting Empirical Results Discussion of Results in M&A Event Study Research Definition of Event Dates Discussion of Results of u.s. M&A Event Study Research Discussion of Results of German M&A Event Study Research Discussion of Results of Long-Run Performance M&A Studies Discussion of Results of Divestitures and Corporate Spin- Offs Event Studies Summary and Conclusion SAMPLE DESIGN AND DESCRIPTION Problem Definition and Selection Criteria Selection Process x

11 5.3. Sample Description Calculations of Daily Discrete Cleaned Returns DATA ANALYSIS... I Introduction Time-Series Plots Testing for Normality Motivation Normal Plots and Shapiro-Wilk Statistic Studentized Range Moment Checking Testing for Independence Motivation Autocorrelation Function Runs Test above and below the Mean Testing for Homoskedasticity of Many Populations Motivation Levene Test Identifying Stationarity Motivation Unit Root Test Testing for Outliers Motivation The Walsh Procedure Testing for Autoregressive Conditional Heteroskedasticity Discussion of Results and Conclusion xi

12 7. METHODOLOGY AND HYPOTHESIS Introduction Methodology: The Use of Abnormal Return Techniques in Event Studies Definition of Event Dates and Periods The Mean Adjusted Return ModeL The Industry Market Model Estimation of Residuals Statistical Tests Standardized Models Hypothesis Evaluation of the Success of the Transaction from the Point of View of the Minority Shareholders Evaluation of the Success of the Transaction from the Point of View of the Majority Shareholders Hypothesis Testing STATISTICAL ANALYSIS AND EMPIRICAL RESULTS Preliminary Findings Results for the MAR and the lam Models Comparative Results Segmented by Percentage Shareholdings Comparative Results Segmented by Related or Unrelated Industries Comparative Results Segmented by Size Composition of Targets' Boards Testing if a Shift in Parameters occurred Implementing Trading Strategies Summary and Conclusion Xll

13 9. DISCUSSION OF RESULTS AND CONCLUSIONS... l71 APPENDICES Appendix 1: Appendix 2: List of Publicly Listed German Companies which have acceded to the German Takeover Code List of Target Companies with the Respective Majority Owner Appendix 3: Summary of the Data Set by Transaction Appendix 4: Selected Time-Series Plots Appendix 5: Selected Normal Plots Appendix 6: Tables for Chapter 6: Data Analysis Appendix 7: Appendix 8: Macros for the Moment Checking Statistics gl (Skewness) and g2 (Kurtosis} Selected Autocorrelation Functions and Partial Autocorrelation Functions Appendix 9: Results of the Runs Test Appendix 10: Macro for the Levene Test Appendix 11: Macro for the Walsh Procedure of Outliers' Detection Appendix 12: Appendix 13: Appendix 14: Graphs of Abnormal Returns for the Mean Adjusted Return Model Graphs of Abnormal Returns for the Industry Adjusted Market Model List of Listed Options on the Deutsche Terminborse AG REFERENCES xiii

14 LIST OF TABLES, FIGURES, AND GRAPHS TABLE 2.1: EVOLUTION OF GERMAN M&A ACTIVITY ( ) TABLE 2.2: OWNERSHIP DISTRIBUTION BY SEGMENT (AS OF DECEMBER 31, 1995)14 TABLE 2.3: CONTRAST BETWEEN THE AMENDED GERMAN AND THE BRITISH TAKEOVER CODES TABLE 3.1: ASSUMED OWNERSHIP PATTERN TABLE 4.1: STATISTICAL MODELS FOR CALCULATING ABNORMAL RETURNS TABLE 4.2 TABLE 4.3 MERGER RESULTS OF EMPIRICAL STUDIES IN THE UNITED STATES- CARs OF THE ACQUIRING AS WELL AS ACQUIRED FIRMS MERGER RESULTS OF EMPIRICAL STUDIES IN GERMANY - CARs OF THE ACQUIRING AS WELL AS ACQUIRED FIRMS TABLE 4.4: EMPIRICAL RESULTS OF LONG-RUN M&A STUDIES TABLE 4.5: TABLE 5.1: EMPIRICAL RESULTS OF DIVESTITURES AND SPIN-OFFS EVENT STUDIES MARKET CAPITALIZATION OF THE SAMPLE (FREQUENCY DISTRIBUTION) FIGURE 7.1: PERFORMANCE SCENARIOS FOR THE DIFFERENT SHAREHOLDER GROUPS FIGURE S.l: NUMBER OF COMPANIES WITH AVERAGE RETURNS BEFORE AND AFTER THE ANNOUNCEMENT DATE TABLE S.l: A VERAGE ABNORMAL RETURNS OF THE CROSS-SECTIONAL PORTFOLIO FOR THE MEAN ADJUSTED RETURN MODEL FROM DAY -60 TO xv

15 TABLE 8.2: CUMULATIVE AVERAGE ABNORMAL RETURNS (CAAR) OF THE CROSS-SECTIONAL PORTFOLIO FOR THE MEAN ADJUSTED RETURN MODEL FROM DAY TABLE 8.3: AVERAGE ABNORMAL RETURNS OF THE CROSS-SECTIONAL PORTFOLIO FOR THE INDUSTRY ADJUSTED MARKET MODEL FROM DAY- 60 TO TABLE 8.4: CUMULATIVE AVERAGE ABNORMAL RETURNS (CAAR) OF THE CROSS-SECTIONAL PORTFOLIO FOR THE INDUSTRY ADJUSTED MARKET MODEL FROM DAY TABLE 8.5: AVERAGE ABNORMAL RETURNS (AAR) AND CUMULATIVE AVERAGE ABNORMAL RETURNS (CAAR) FOR THE MEAN ADJUSTED RETURN MODEL FOR DIFFERENT OWNERSHIP STRUCTURES TABLE 8.6: AVERAGE ABNORMAL RETURNS (AAR) AND CUMULATIVE AVERAGE ABNORMAL RETURNS (CAAR) FOR THE MEAN ADJUSTED RETURN MODEL FOR TRANSACTIONS BETWEEN RELATED AND UNRELATED INDUSTRIES TABLE 8.7: AVERAGE ABNORMAL RETURNS (AAR) AND CUMULATIVE AVERAGE ABNORMAL RETURNS (CAAR) FOR THE INDUSTRY ADJUSTED RETURN MODEL FOR THE SAMPLES SEGMENTED BY SIZE COMPONENTS TABLE AI: LIST OF PUBLICLY LISTED GERMAN COMPANIES WHICH HAVE ACCEDED TO THE PROVISIONS OF THE GERMAN TAKEOVER CODE (AS OF FEBRUARY 26, 1998) TABLE A2: LIST OF TARGET COMPANIES WITH THE RESPECTIVE MAJORITY OWNER TABLE A3: SUMMARY OF THE DATA SET BY TRANSACTION TABLE 6.1: RESULTS OF THE RYAN-JOINER STATISTIC R xvi

16 TABLE 6.2: RESULTS OF THE STUDENTIZED RANGE TEST TABLE 6.3: RESULTS OF THE SKEWNESS TEST TABLE 6.4: RESULTS OF THE KURTOSIS TEST TABLE 6.5: RESULTS OF THE INDEPENDENCE TEST TABLE 6.6: RESULTS OF THE LEVENE TEST TABLE 6.7: RESULTS OF THE LEVENE TEST- SAME TIME PERIOD TABLE 6.8: RESULTS OF THE STATIONARITY TEST TABLE 6.9: RESULTS OF THE FREQUENCY DISTRIBUTION TABLE 6.10: RESULTS OF THE UNIT ROOT TEST TABLE 6.11: RESULTS OF IDENTIFYING OUTLIERS TABLE 6.12: RESULTS OF THE OUTLIERS' TEST TABLE 6.13: ESTIMATES OF THE ARCH PARAMETERS TABLE 6.14: RESULTS OF THE ARCH DIAGNOSIS TABLE A9: RUNS TEST RESULTS GRAPH 8.1: CUMULATIVE AVERAGE ABNORMAL RETURNS FOR CROSS-SECTIONAL PORTFOLIO FOR THE PERIOD HALF A YEAR BEFORE THE ANNOUNCEMENT DATE (FROM DAY -120 TO 0 FOR MEAN ADJUSTED RETURN MODEL) GRAPH 8.2: CUMULATIVE AVERAGE ABNORMAL RETURNS FOR CROSS-SECTIONAL PORTFOLIO FOR ONE YEAR AFTER THE ANNOUNCEMENT DATE (FROM DAY -60 TO +250 FOR MEAN ADJUSTED RETURN MODEL) GRAPH 8.3: CUMULATIVE AVERAGE ABNORMAL RETURNS FOR CROSS-SECTIONAL PORTFOLIO OVER FIVE YEARS AFTER THE ANNOUNCEMENT DATE (MEAN ADJUSTED RETURN MODEL) xvii

17 GRAPH 8.4: AVERAGE ABNORMAL RETURNS OF CROSS-SECTIONAL PORTFOLIO FOR MEAN ADJUSTED RETURN MODEL FROM DAY -60 TO GRAPH 8.5: CUMULATIVE AVERAGE ABNORMAL RETURNS FOR MEAN ADJUSTED RETURN MODEL SEGMENTED BY ANNOUNCEMENT DATE RETURN GRAPH 8.6: CUMULATIVE AVERAGE ABNORMAL RETURNS FOR CROSS-SECTIONAL PORTFOLIO FOR THE PERIOD HALF A YEAR BEFORE THE ANNOUNCEMENT DATE (FROM DAY -120 TO 0 FOR THE INDUSTRY ADJUSTED RETURN MODEL) GRAPH 8.7: CUMULATIVE AVERAGE ABNORMAL RETURNS FOR CROSS-SECTIONAL PORTFOLIO FOR ONE YEAR AFTER THE ANNOUNCEMENT DATE (FROM DAY -60 TO +250 FOR INDUSTRY ADJUSTED RETURN MODEL) GRAPH 8.8: CUMULATIVE AVERAGE ABNORMAL RETURNS FOR CROSS-SECTIONAL PORTFOLIO OVER FIVE YEARS AFTER THE ANNOUNCEMENT DATE (INDUSTRY ADJUSTED RETURN MODEL) GRAPH 8.9: AVERAGE ABNORMAL RETURNS OF CROSS-SECTIONAL PORTFOLIO USING THE INDUSTRY ADJUSTED RETURN MODEL FROM DAY -60 TO TABLE A14: LIST OF PUBLICLY LISTED GERMAN COMPANIES WHICH HAVE OPTIONS TRADED ON THE DEUTSCHE TERMINBORSE AG xviii

18 LIST OF VARIABLES 1. Introduction the stochastic return of the share of company during period t period of the return, here assumed to be a day the expected value of a variable the expected value of the return of a share of company i during period t 4.3. Specifications of Statistical Models the stochastic return of the share of company i during period t the stochastic return of the market portfolio m or the market index m in period t the stochastic return of industry portfolio k or the industry index k in period t the stochastic return of the size portfolio s in period t the unsystematic risk specific to share i the systematic risk specific to share i the systematic industry risk the systematic risk specific to the size factor s systematic risk specific to share i in book-tomarket portfolio s, systematic risk specific to share i in size portfolio s, the residual value or abnormal return specific to share i in period t the risk-free rate the high-min us-low book-to-market portfolio return in period t, the small-minus-large size portfolio return in period t, xix

19 the abnormal return of share i in period t the average return of share i in period t rmt the average return of the market m in period t 4.4. Problems with Estimating and Interpreting Empirical Results CAR cumulative abnormal returns or cumulative average residuals (generic) 5.4 Calculations of Daily Discrete Cleaned Returns r t Pt the simple stochastic return in period t the price of a share at the end of period t Normal Plots and Shapiro-Wilk Statistic n Zj w A z R number of observations an index of values values of the (i/n)th quartile of the empirical distribution function values of the (i/n)th quartiles of the standardized normal distribution function (N(O,l)) sample cumulative (empirical) distribution function the Shapiro-Wilk statistic a vector of tabulated coefficients especially derived to be used in computing the Shapiro-Wilk statistic the vector of ordered Zj'S the usual estimator of 0'2, the standard deviation the Ryan-Joiner statistic Studentized Range 0' the standard deviation of a population mean xx

20 the mean of a population Moment Checking A.I A.z the skewness measure of a distribution the kurtosis measure of a distribution the skewness statistic the kurtosis statistic Autocorrelation Function the kth autocorrelation (or serial correlation) of a time series the estimated autocorrelation coefficient Runs Test above and below the Mean p the number of runs the number of observations above the mean the number of observations below the mean Levene Test L Levene test statistic the absolute differences of I rji - mean(r) I Unit Root Test Ut statistic to find the unit root in a time series classical stochastic error term with zero mean, constant variance 0 2, and non-autocorrelated denominated for an autoregressive regressive process xxi

21 6.7 Testing for Outliers o = outlier variable 6.8 Testing for Autoregressive Conditional Heteroskedasticity estimated square residuals of an autoregressive process The Mean Adjusted Return Model a constant value for a time series of company i the mean of a time-series, an estimate of the expected return of that series the estimate of abnormal return E it, the abnormal return the covariance Estimation of Residuals AA~ CA~/e = T average abnormal returns or average residuals cumulative average abnormal returns (or cumulative average residuals) starting from period b and ending in period e period over which abnormal returns are cumulated Standardized Models SA~ ASAR t CASARb,e standardized abnormal returns average standardized abnormal returns cumulative average standardized abnormal returns starting from period b and ending in period e xxii

22 8.3 Comparative Results Segmented by Percentage Shareholdings Xa na Calculating the Weir test statistic estimated mean of sub-sample a estimated mean of sub-sample b estimated standard deviation of sub-sample a estimated standard deviation of sub-sample b number of observations of sub-sample a number of observations of sub-sample b xxiii