家族企業與主併公司績效之關聯: 以台灣併購案為例 - 政大學術集成

全文

(1)國立政治大學財務管理研究所碩士論文. ‧ 國. 學. 治政 The Relation between Family 大 Firms and 立 Acquiring Firm Performance: The Cases of ‧. M&A in Taiwan. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 指導教授：湛可南博士研究生：許韶耘. 中華民國一百零三年五月.

(2) 致謝這篇論文能夠完成，首先要感謝我的指導教授湛可南老師。感謝老師在論文指導上所花費的時間與精力，同時也非常感謝老師的支持，贊助經費購買部分資料，讓我得以順利建構樣本，並進行分析。從老師身上，我學到許多寶貴的經驗，不論是老師做研究時的嚴謹態度，抑或是多方面思考與善用各種研究方法，都讓我收穫滿載。在論文的修改上，要特別感謝口試委員王衍智老師及陳鴻崑老師，感謝口委們百忙之中撥空參與我的口試，並給予許多寶貴且專業的意見與指正，使得本論文內容得以更加圓融及充實。. 政治大此外，我也要感謝所有幫助過我的同學與朋友們，除了陪伴我在研究過程中立. ‧ 國. 學. 一起努力，給予我鼓勵與幫我打氣外，也樂於幫助我處理各種疑難雜症，不論是論文架構之建立、樣本建構與整理，抑或是統計軟體的使用，都感謝有你們為我. ‧. 解答。感謝學長姐們經驗的傳承，讓我在撰寫論文時能夠駕輕就熟，準備口試時. sit. y. Nat. 得以周全順利。感謝學弟妹們在我的論文接近尾聲時，挪出與老師討論的時間，. io. n. al. er. 優先讓老師修改我的論文。. i n U. v. 最後，還要感謝我的家人，感謝您們長期的陪伴，以及對我的關心與照顧，. Ch. engchi. 讓我得以用心於課業與研究。我將這份成長與喜悅與您們分享！感謝所有的人！. 許韶耘於民國一百零三年五月十三日. 2.

(3) Abstract This study investigates the relation between family firms and acquiring firm performance for our sample of Taiwanese mergers and acquisitions between 1999 and 2013. We find that cumulative abnormal returns of family acquirers on average outperform those of nonfamily acquirers by 2.17% three days around the announcement. Family acquirers obtain greater abnormal returns even after controlling for both firm characteristics such as firm size, book to market, prior return,. 治政大 and mode of payment. public target and deal characteristics such as year dummy 立 ‧ 國. 學. Furthermore, we explore the potential impact of the deviation between voting rights and cash flow rights on family acquiring performance. In the sample of Taiwanese. ‧. sit. y. Nat. mergers and acquisitions, the deviation is not the significant factor to cause a. n. al. er. io. negative influence. As a result, family acquirers with the advantage of eliminating. i n U. v. agency problems may generate more benefits than nonfamily acquirers.. Ch. engchi. 3.

(4) Contents I.. INTRODUCTION ............................................................................. 6. II. LITERATURE REVIEW .................................................................... 10 A. THE POTENTIAL COSTS OF FAMILY FIRMS .................................................. 10 B. THE POTENTIAL BENEFITS OF FAMILY FIRMS .............................................. 11 C. FAMILY FIRM PERFORMANCE, LARGE-MINORITY SHAREHOLDER CONFLICTS AND THE DEVIATION BETWEEN VOTING RIGHTS AND CASH FLOW RIGHTS ....................... 13 D. RESEARCH FOCUS .............................................................................. 14 III. SAMPLING AND DATA COLLECTION .............................................. 15 A. SAMPLE SELECTION ............................................................................ 15 B. VARIABLE DEFINITION ......................................................................... 18 (a) Dependent variable .................................................................. 18 (b) Independent variable ................................................................ 18 (c) Control Variable ........................................................................ 20. 立. 政治大. ‧ 國. 學. IV. EMPIRICAL RESULT....................................................................... 23. ‧. A. DESCRIPTION STATISTIC ....................................................................... 23 B. EMPIRICAL RESULT ............................................................................. 28 (a) Equation A: simple regression analysis ..................................... 31 (b) Equation B: multiple regression analysis controlling for acquiring firm characteristics .......................................................................... 31 (c) Equation C: multiple regression analysis controlling for firm a l characteristics ........................................... v i characteristics and deal 34 n C (d) Equation D: multiple h regression e n g canalysis h i U controlling for the interaction variable ......................................................................... 37 (e) The relation between family acquiring performance and the deviation between voting rights and cash flow rights...................... 39. n. er. io. sit. y. Nat. V. CONCLUSION ............................................................................... 42 VI. REFERENCE .................................................................................. 43. 4.

(5) List of Table TABLE1. SAMPLE DISTRIBUTION BY ANNOUNCEMENT YEAR AND FAMILY FIRM STATUS .......................................................................... 16 TABLE2. SAMPLE DISTRIBUTION BY INDUSTRIES AND FAMILY FIRM STATUS ............................................................................................... 17 TABLE3. DAILY ABNORMAL RETURNS OVER THE WINDOW [-10, 10] FROM THE ANNOUNCEMENT ............................................................. 24 TABLE4. CUMULATIVE ABNORMAL RETURNS OVER THE WINDOWS [0, 1], [-1, 1] AND [-3, 3] AROUND THE ANNOUNCEMENT ....................... 26 TABLE5. FIRM CHARACTERISTICS ........................................................ 29 TABLE6. CORRELATION DATA .............................................................. 30. 政治大. TABLE7. MULTIPLE REGRESSIONS OF FAMILY FIRM STATUS AND ACQUIRING FIRM PERFORMANCE: CONTROL FOR ACQUIRING FIRM CHARACTERISTICS .............................................................................. 33. 立. ‧ 國. 學. ‧. TABLE8. MULTIPLE REGRESSIONS OF FAMILY FIRM STATUS AND ACQUIRING FIRM PERFORMANCE: CONTROL FOR FIRM CHARACTERISTICS AND DEAL CHARACTERISTICS ................................ 35. er. io. sit. y. Nat. TABLE9. MULTIPLE REGRESSIONS OF FAMILY FIRM STATUS AND ACQUIRING FIRM PERFORMANCE: CONTROL FOR FIRM AND DEAL CHARACTERISTICS WITHOUT EITHER FIRM SIZE OR PUBLIC TARGET ... 36. n. al TABLE10. MULTIPLE REGRESSIONS OF FAMILY FIRM v STATUS AND i n C h CONTROL U ACQUIRING FIRM PERFORMANCE: FOR THE INTERACTION i e h n gc VARIABLE ........................................................................................... 38 TABLE11. MULTIPLE REGRESSIONS OF FAMILY ACQUIRING PERFORMANCE AND THE DEVIATION BETWEEN VOTING RIGHTS AND CASH FLOW RIGHTS............................................................................ 40. 5.

(6) I. Introduction In the wide-ranging research on mergers and acquisitions (M&A), it has been a central focus to measure and analyze value creation or value destruction obtained by the shareholders of acquiring firms (Andrade, Mitchell and Stafford, 2001). Numerous studies also examine the factors which would affect M&A performance. For instance, Moeller, Schlingemann and Stulz (2004) investigate the size effect of. 政治大 impact of public and private 立targets.. acquiring firms, whereas Faccio, McConnell and Stolin (2006) put emphasis on the. ‧ 國. 學. However, it is almost silent about the relation between family firms and. ‧. acquiring firm performance except the one conducted by Ben-Amar and Andre. sit. y. Nat. (2006). Family firms may operate totally differently from nonfamily firms because of. er. io. the divergence of firm characteristics. For example, family firms exhibit high level of. al. n. v i n ownership concentration, whereas exhibit high level of ownership C hnonfamily firms U engchi dispersion. Thus, it is essential to consider whether a corporation is a family firm or nonfamily firm when discussing acquiring firm performance. With respect to the impact of family firms, there are contrast viewpoints in numerous papers. On the one hand, family firms may have better M&A performance because of their ability to eliminate agency problems which occur when there is separation between control and ownership (Anderson and Reeb, 2003). Agency problems may. 6.

(7) lead value destruction to acquiring firms since managers tend to make M&A decisions from their private perspectives while sacrificing the interests of the shareholders. For example, managers of nonfamily firms may increase firm size by choosing non-optimal M&A projects rather than redistributing free cash flow to the shareholders in order to obtain higher compensation or ensure their job positions (Amihud and Lev, 1981; Avery, Chevalier and Schaefer, 1998; Firth, 1991; Jensen, 1986).. 立. 政治大. On the other hand, some studies raise the opposite arguments. La Porta,. ‧ 國. 學. Lopez-de-Silanes and Shleifer (1999) claim that the main agency problems beyond. ‧. the region of US and UK are not the manager-shareholder conflicts but the ones. Nat. io. sit. y. between large and minority shareholders (Claessens, Djankov, Fan and Lang, 2000;. er. Faccio and Lang, 2002; Dyck and Zingales, 2004). In particular, Zhang (1998) indicates. al. n. v i n that those large shareholders C orhcontrolling shareholders e n g c h i U who hold considerable. proportion of their wealth in a company may make sub-optimal decisions and tend to do corporate diversification at the expense of minority shareholders. It is well-known that diversification always leads to negative abnormal returns (Morck, Shleifer and Vishny, 1988). It is inconclusive that whether family firms have the positive or negative influence on acquiring firm performance in the prior research. The former results 7.

(8) from the elimination of agency problems between mangers and shareholders, and the latter is because of the conflicts between large and minority shareholders. Thus, the impact of family firms on acquiring firms is actually an empirical question. The purpose of our study is to examine whether family acquirers perform better or worse than nonfamily acquirers in the Taiwanese sample. Besides, we also consider the effect of the deviation between voting rights and. 政治大. cash flow rights on family acquiring performance. As La Porta, Lopez-de-Silanes and. 立. Shleifer (1999) show, the corporations in US and UK are characterized by ownership. ‧ 國. 學. dispersion with existence of agency problems; by contrast, the firms in most of. ‧. Continental Europe and Asia are characterized by greater ownership concentration in. Nat. sit er. io. conflicts.. y. the hands of the individuals or families leading to the large-minority shareholder. al. n. v i n Since ownership structureCinh Taiwanese corporations exhibits high levels of engchi U. concentration, the conflicts between large and minority shareholders may probably exist and negatively affect family acquiring performance. The deviation between voting rights and cash flow rights is a common proxy for the presence of the large-minority shareholder conflicts. If the deviation gets higher, large shareholders may tend to exploit the interests of minority shareholders through M&A transactions. Thus, we examine whether family acquirers lose more in M&A transactions when 8.

(9) there exists a conflict interest between large and minority shareholders. To our best of knowledge, this is the first study focusing on the relation between family firms and acquiring firm performance in Taiwan. We find that calculative abnormal returns of family acquirers outperform those of nonfamily firms by 1.69% and 2.17% one day and three days around the announcement respectively. Family firms generate greater abnormal returns even. 政治大. after controlling for firm size, book to market, prior return, public target, year dummy. 立. and mode of payment. Moreover, we find that the deviation between voting rights. ‧ 國. 學. and cash flow rights is not the significant factor to cause a negative influence on. ‧. family acquiring performance in the sample of Taiwanese mergers and acquisitions.. Nat. io. sit. y. Without the conflicts between large and minority shareholders, family acquirers. n. al. er. having the advantage of eliminating agency problems may generate more benefits than nonfamily acquirers.. Ch. engchi. i n U. v. The remainder of our study is organized as follows. Section II reviews the pertinent literatures about the potential costs and benefits of family firms. Section III presents the data and methodology. In section IV, we provide our empirical results. Section V offers a conclusion of this paper.. 9.

(10) II. Literature Review In Taiwan, almost 60% of the public corporations are family firms. These families represent a special class of the shareholders since they are long-term investors even cross multiple generations, with poorly diversified portfolios and often control top management positions (Anderson and Reeb, 2003). Thus, families in a unique position may exert influences on the corporations, causing family firm performance. 政治大 The potential costs of立 family firms. different from nonfamily firm performance.. 學. ‧ 國. A.. Since holding substantial ownership of a firm, families may have incentives and. ‧. power to make decisions that fulfill their own desires. Fama and Jensen (1985). sit. y. Nat. indicate that large undiversified shareholders may make investment decisions totally. er. io. different from those made by atomistic shareholders. Atomistic shareholders tend to. al. n. v i n evaluate investment projects by rules which would maximize the firm C the h emarket ngchi U value; on the contrary, large shareholders may sacrifice greater benefits and pursue the goals like technological innovation, firm growth and survival. Besides, they may emphasize more on firm diversification to reduce the risk. As controlling top management positions, families may probably expropriate wealth from the corporations. Johnson et al. (2000) note that large shareholders are capable of transferring assets or profits to other entities through creating group. 10.

(11) structure. Earlier literatures also report that there are various ways for tunneling such as excessive compensation, special dividend, advantageous transfer-price, loan guaranties and M&A transactions (Anderson and Reeb, 2003; Ben-Amar et al. 2006). Family firms may also cause opportunity costs when top management positions are occupied by families who lack for talent, expertise or competency. Unlike professional managers with incentives to raise competitive abilities, limited. 政治大. personnel to families may generate opportunity costs and meanwhile decrease the. 立. firm value (Peres-Gonzalez, 2001; Ben-Amar and Andre, 2006). Barclay and. ‧ 國. 學. Holderness (1989) also argue that the family firm value would be lower than the. ‧. nonfamily firm value because of families’ intention to create impediments for third. Nat. io. sit. y. parties to participate in the firm and select those top managers to capture firm’s. n. al. er. control. The consistent arguments are raised by Gomez-Mejia, Nunez-Niclel, and. Ch. engchi. Gutierrez (2001); Shleifer and Vishny (1997) as well. B.. i n U. v. The potential benefits of family firms The prior literatures suggest that families may have a negative impact on firm. performance; however, families may also bring benefits to the corporations in a positive way. In this sub-section, we will give three viewpoints about the advantages of family firms. First, large shareholders may have strong economic incentives to eliminate agency problems and maximize the firm value since their wealth is closely 11.

(12) connected to the firm welfare (Demsetz and Lehn, 1985). Moreover when monitoring, families with lengthy tenure could potentially offer superior oversights. Second, families relative to managers in nonfamily firms tend to invest the projects from a long-term perspective. It is because families have an enduring presence in the firms rather than a shorter managerial horizon (Anderson and Reeb, 2003). Stein (1988, 1999) also points out that the firms with shareholders having. 政治大. longer investment horizons are unlikely to forgo good investments for current. 立. earning, which is so called managerial myopia. Instead, families judge investments. ‧ 國. 學. such as M&A transactions by the market rules, choosing those with positive NPV. As. ‧. a result, families may invest more efficiently than nonfamily firms (James, 1999).. Nat. io. sit. y. Casson (1999) and Chami (1999) also have consistent arguments that families may. er. view their firm as an asset which would be passed on from one generation to. al. n. v i n C h value maximization another; thus, families are long-term e n g c h i U advocates for firm survival.. Third, families may suffer reputation concerns which result from their enduring presence in the firms. As families tend to maintain the reputation, they are less likely to act harmfully to third parties such as a hostile takeover to the suppliers. Furthermore, external parties such as capital providers may cope with the same governing bodies in family firms for a longer time. As business relation and reputation of the families are built, lasting economic consequences may come into 12.

(13) existence (Anderson and Reeb, 2003). For instance, family firms may enjoy a lower cost of debt financing (Anderson et al. 2003). C.. Family firm performance, large-minority shareholder conflicts and the deviation between voting rights and cash flow rights As La Porta, Lopez-de-Silanes and Shleifer (1999) show, the corporations in most. of Continental Europe and Asia, including Taiwan, are characterized by greater. 政治大. ownership concentration in the hands of the individuals or families. With high levels. 立. of ownership concentration, large shareholders are likely to fulfill their own desires. ‧ 國. 學. at the expense of minority shareholders. Thus, the conflicts between large and. ‧. Nat. io. sit. deviation between voting rights and cash flow rights.. y. minority shareholders may exist. Generally, these conflicts are measured by the. er. As voting rights exceed cash flow rights, large shareholders may tend to exploit. al. n. v i n C h by M&A transactions the interests of minority shareholders e n g c h i U since their wealth will be less influenced by value destruction of M&A. Numerous studies also argue that the deviation between voting rights and cash flow rights may have a negative impact on firm performance (Barontini and Caprio, 2004; Claessens et al., 2002; Lemmon and Lins, 2003; Cronqvist and Nilson, 2003). It concurs with the arguments raised by La Porta et al. (2002) that better performance is gained by those with higher cash flow rights by large shareholders and with better protection of minority shareholders. 13.

(14) D.. Research focus Our study focuses on the relation between family firms and acquiring firm. performance. Through extensive literature review, we could comprehend that the family firm is a less effective organization because of restricted tradability of families’ claims, the diversification or other financial preferences different from atomistic shareholders, the potential tunneling intension for pecuniary or non-pecuniary. 政治大. benefits, and limited talents to a pool of families. However, arguments about. 立. elimination of agency problems, long-term horizon with positive NPV projects. ‧ 國. 學. choosing and the reputation effect suggest that families have strong motivations to. ‧. ensure firm performance on both daily operation and M&A transactions.. Nat. io. sit. y. Accordingly, the effect of family firms on acquiring firm performance is actually. er. an empirical question that we examine in our study. Here we address two specific. al. n. v i n C hor lose more thanUnonfamily firms in the process questions: (1) Do family firms gain engchi of M&A? (2) Is family acquiring performance significantly or insignificantly affected by the deviation between voting rights and cash flow rights? The question (2) focuses on the impact of the deviation to analyze why family acquirers perform better or worse. At the end, we will offer an investigation of these two questions.. 14.

(15) III. Sampling and Data Collection A.. Sample selection The sample of acquiring firms comes from Taiwan Economic Journal’s (TEJ). Event Study Database. We select M&A transactions with the following criteria: (1) The announcement dates are between 1999 and 2013; (2) Acquiring firms are listed Taiwanese companies; (3) Acquiring firms are beyond the financial industry; the. 政治大 during the period are included; 立 (5) Mergers between parent companies and subsidies. amounts of exclusive data are 31 (4) Companies with several M&A transactions. ‧ 國. 學. are excluded. The amounts of the transactions we delete are 188 based on the. ‧. principle (5). Our final sample comprises 171 events (133 companies).. sit. y. Nat. Table 1 reports the annual number of M&A transactions which sorted by family. er. io. and nonfamily firms. Our analysis suggests that companies have done more M&A. al. n. v i n transactions since 2004, the amount as the amount in 2003. The C h in which is as twice engchi U transactions of family and nonfamily acquiring firms are accounted for 55% and 45% respectively. Table 2 further indicates that 85% of the M&A transactions in our sample are done by the firms in the electronics Industries. Half of them are family firms and half of them are nonfamily firms.. 15.

(16) Table1. Sample Distribution by Announcement Year and Family Firm Status The sample contains all Taiwanese M&A transactions between 1999 and 2013 listed on TEJ where acquiring firms are listed domestic companies. However, we exclude the ones between parent companies and subsidies as well as in the financial industry. Family acquiring firms are identified by any of the following conditions is satisfied: (1) A family, an individual or a firm controlled by the family own at least 10% of ultimate share-holdings; meanwhile, there is at least one family member in board of director; (2) More than half of the board members are family members.. Family Firm Status. Announcement year. Family. 1999. 0. 立. 1. 6. 2. 11. 5. a l 10 9C h. 4. 6 6 8. 2007. 14. 2008. sit. y. n. 2006. io. 2005. 5. er. 2004. 3. Nat. 2003. 3. 3. ‧. 2002. 3. 1. 學. 2001. Nonfamily 政治 1大. 0. ‧ 國. 2000. All. n U engchi 11. iv. 16 14 20. 7. 21. 8. 6. 14. 2009. 4. 9. 13. 2010. 5. 7. 12. 2011. 6. 9. 15. 2012. 8. 6. 14. 2013. 5. 3. 8. Total. 94. 77. 171. 16.

(17) Table2. Sample Distribution by Industries and Family Firm Status The sample contains all Taiwanese M&A transactions between 1999 and 2013 listed on TEJ where acquiring firms are listed domestic companies. However, we exclude the ones between parent companies and subsidies as well as in the financial industry. Family acquiring firms are identified by any of the following conditions is satisfied: (1) A family, an individual or a firm controlled by the family own at least 10% of ultimate share-holdings; meanwhile, there is at least one family member in board of director; (2) More than half of the board members are family members. Other includes the industries such as transports, cultural and creative industries and Retail industries.. Family Firm Status. 治 Nonfamily 政 Family 大. Industry. Food Industry. 立. 1. 0. All. 1. 0. Electrical equipment. 0. 1. Biotechnology and Medical Care. 7. n. al. Ch. e n g2 c h i. sit. y. 1. 1. 8. 0. 2. er. io. Paper Industry. 2. ‧. ‧ 國. 學. 2. Nat. Petrochemical plastics. i n U. v. Steel Industry. 1. 1. 2. Electronics Industry. 71. 73. 144. Building Materials. 7. 0. 7. Other. 3. 1. 4. Total. 94. 77. 171. 17.

(18) B.. Variable definition. (a) Dependent variable Andrade et al. (2001) argue that companies could examine value creation or value demolition of M&A by observing the market reaction around the announcement. Thus, we use the event-study methodology to obtain the Cumulative Abnormal Return (CAR) as the indicator of M&A performance. Our calculations are. 政治大. based on OLS Risk-Adjusted Model, with the benchmarks of TSE and OTC market. 立. index return. The estimation period is [-375, -15] from the announcement and. ‧ 國. 學. abnormal returns are accumulated over the different windows [0, 1], [-1, 1] and [-3, 3]. ‧. around the announcement.. Nat. io. sit. y. (b) Independent variable. er. Independent variables include family firm status and the deviation between. al. n. v i n Ch voting rights and cash flow rights. We manually identify e n g c h i U family or nonfamily acquiring firms based on the information of the board structure, board members and share-holdings from TEJ as well as the news, articles and magazines. Moreover, the value of the deviation is calculated by TEJ. 1.. Family firm status The principles for identifying family firms are quite inconsistent. Some studies. focus on the board structure since board members have the rights to determine the 18.

(19) policies and the top executives of the firms; others put emphasis on the corporate ownership of voting rights (Lee and Liao, 2004; Barnes and Hershon, 1976; La Porta et al., 1999; Claessens et al., 2000; Faccio and Lang, 2002). La Porta et al. (1999), Faccio and Lang (2002) argue that it is a family firm when a family or an individual own at least 20% of either the voting rights or ultimate control stakes. The threshold is 20% of the votes since this is generally enough to. 政治大. effectively control a firm (La Porta et al., 1999). Anderson and Reeb (2003) use both. 立. the fractional equity ownership of the families and (or) the board presence of family. ‧ 國. 學. members to identify family firms.. ‧. Yeh1 (1999) mentions that it is a family firm if any of the two following. Nat. io. sit. y. conditions is satisfied: (1) A family, an individual or a firm controlled by the family. er. own at least 10% of ultimate share-holdings; meanwhile, there is at least one family. al. n. v i n C h(2) More than half member in the board of director; e n g c h i U of the board members are. families. In our study, we use the principle raised by Yeh2 (1999) since their sample include Taiwanese corporations. Family firm status is defined as a dummy variable of the value equal to 1 when it is a family firm; 0 otherwise.. 1 2. Please refer to ”葉銀華 (1999)，’家族控股、董事會組成與經營績效─台灣家族企. 業管治機制之研究’，行政院國家科學委員會專題研究計畫成果報告。”. 19.

(20) 2.. Deviation There are three ways to measure the deviation between voting rights and cash. flow rights: (1) a dummy variable of the value equal to 1 when there is separation between control rights and ownership, 0 otherwise; (2) the difference between the level of voting rights and the level of cash flow rights; (3) the ratio of the level of voting rights respective to the level of cash flow rights (Claessens et al., 2002; Faccio. 政治大. and Lang, 2002). In our study, we use method (2) to calculate the deviation.. 立. If the level of control rights exceeds the level of ownership, which means the. ‧ 國. 學. higher deviation, large shareholders may tend to exploit the interests of minority. ‧. shareholders through M&A transactions. It is because that the wealth of large. Nat. io. sit. y. shareholders will be less influenced by value destruction of M&A. Thus, the deviation. er. between voting rights and cash flow rights is a common proxy variable to measure. n. al. i n C the conflicts between large and minority h e nshareholders. gchi U. v. (c) Control Variable In our multivariate analysis, we control for several variables which are identified in the papers as having the impacts on CARs. These variables can be broken down into two groups: (1) Firm characteristics of acquirers and targets, the former include firm size, book to market and prior return; the latter include public target; (2) Deal characteristics, such as year and mode of payment. 20.

(21) 1.. Firm size There are numerous studies indicating that the CARs of acquiring firms would be. negatively affected by their firm size, which is called the size effect (Loderer and Martin, 1990; Schwert, 2000). Moeller, Schlingemann and Stulz (2004) also argue that small firms fare significantly better than large firms when making M&A announcement. On average, the abnormal returns obtained by small acquiring firms. 政治大. exceed the ones of large acquiring firms by 2.24%. In our study, we use natural log of. 立. to measure firm size (Anderson and Reeb, 2003).. ‧. 2.. 學. ‧ 國. the book value of total assets as of the calendar year-end prior to the announcement. Book to market (BM). Nat. io. sit. y. In our study, we use the equity book to market ratio as of the calendar year-end. er. prior to the announcement. Some studies argue that firms with low BM. al. n. v i n C hacquisition; on theUcontrary, firms with high BM underperform after a merger and engchi experience superior performance. They conclude that it is because of “performance extrapolation” and “hubris”, which lead glamour firms to make relatively poorer acquisitions (Rau and Vermaelen, 1998). 3.. Prior return Our control variables include prior return, which is the sum of acquirers’. abnormal returns over the period from fifteen trading days prior to the 21.

(22) announcement to three days prior to the announcement (Faccio, McConnell and Stolin, 2006). By controlling for this variable, we can exclude extraordinary returns caused by the events occurred prior to the announcement. 4.. Public target Weather target firms are public or private will influence the CARs of acquiring. firms. Faccio, McConnell and Stolin (2006) show that average abnormal returns of. 政治大. acquirers of public targets are insignificantly negative, whereas average abnormal. 立. returns of acquires of private targets are significantly positive. The earlier research. ‧ 國. 學. also takes the listed effect into consideration (Moeller et al., 2004). We set a dummy. ‧. variable which equals to 1 for acquiring public targets; 0 for acquiring private targets.. Nat. y. Year after 2007. io. sit. 5.. er. We control for year dummy which equals to 1 when the acquisitions occurred. al. n. v i n C h happened between during 2007 to 2013; and 0 for those e n g c h i U 1999 and 2006. 6.. Cash payment There are evidences showing that mode of payment will influence acquirers’. performance (Andrade, Mitchell and Stafford, 2001). The average CAR gets higher when paying in stock for private targets and cash for public targets respectively (Chang, 1998; Travlos, 1987). Thus, we control for the payment by setting a dummy variable equal to 1 when the transactions are paid in cash; 0 for the stock payment. 22.

(23) IV. Empirical Result A.. Description statistic Table 3 presents daily abnormal returns over the window [-10, 10] around the. announcement. Particularly, the ARs of our sample are highly significant which are 0.56% and 0.54% in day [0] and day [1] respectively. Besides compared to nonfamily firms with ARs closed to zero, family firms have significantly greater ARs on the. 政治大 difference of ARs between立 family and nonfamily firms. As ARs are accumulated over. announcement date and one day after announcement. Thus, there is a noticeable. ‧ 國. 學. the windows [0, 1], [-1, 1] and [-3, 3] around the announcement, we obtain Table 4.. ‧. As we can see in Table 4, Taiwanese M&A transactions between 1999 and 2013. sit. y. Nat. lead significant value creation to acquiring firms one day and three days around the. er. io. announcement. The outcome differs from American results, often finding negative or. al. n. v i n returnsCof acquiring firms (Bruner, hengchi U. insignificant abnormal. 2002). However, it is. consistent with numerous European studies, which claim that the shareholders of acquiring firms often obtain positive abnormal returns around the announcement (Bigelli and Mengoli, 2004; Bohmer, 2000; and Ben-Amar and Andre, 2006). Furthermore, Table 4 also reports that on average the shareholders of family acquiring firms benefit from M&A transactions over the windows [0, 1], [-1, 1] and [-3, 3] around the announcement.. 23.

(24) Table3. Daily Abnormal Returns over the window [-10, 10] from the Announcement Please refer sample collection to Table1. Family acquiring firms are identified by the principle of any of the following conditions is satisfied: (1) A family, an individual or a firm controlled by the family own at least 10% of ultimate share-holdings; meanwhile, there is at least one family member in board of director; (2) More than half of the board members are family members. ARs denote abnormal returns (%) which are measured by OLS Risk-Adjusted Market Model, with the benchmarks of TSE and OTC market index return and estimation period over [-375, -15] from the announcement. Column 4 indicates the difference between ARs of family and nonfamily firms. The difference tests are based on t-tests for equality in means. P-values are in brackets. ***Statistical significance at the 1 % level. **Statistical significance at the 5% level. *Statistical significance at the 10% level.. AR. All. (%). (1). (0.017). (0.226). (0.318). 0.10. 0.33. -0.18. 0.51. (0.583). (0.172). (0.560). 0.11. 0.23. -0.04. (0.574). (0.366). (0.895). 0.14. -0.30. -5. -4. -3. -2. al. n. -6. -0.06. (0.733) -0.05. 0.38. y. sit. er. ‧ 國. (0.008). io. -7. (4)=(2)-(3). 0.30. Nat. -8. Difference. 0.69**. ‧. -9. 立. 學. -10. 0.52***. Family Nonfamily 政(2) 治大(3). iv. (0.253) n C h(0.539) U e n g c h i 0.33 -0.36. (0.187) 0.27 (0.492) 0.44 (0.204) -0.69*. (0.772). (0.107). (0.279). (0.069). 0.18. 0.30. 0.02. 0.28. (0.332). (0.255). (0.925). (0.440). 0.09. 0.10. 0.08. 0.03. (0.565). (0.670). (0.693). (0.935). 0.11. 0.05. 0.18. -0.14. (0.535). (0.839). (0.502). (0.690). -0.02. 0.37. -0.51*. 0.88**. (0.913). (0.143). (0.087). (0.025). 24.

(25) Table3. (continued) AR. All. Family. Nonfamily. Difference. (%). (1). (2). (3). (4)=(2)-(3). 0.16. 0.36. -0.10. 0.45. (0.425). (0.161). (0.748). (0.247). 0.56***. 0.90***. 0.14. 0.76*. (0.005). (0.001). (0.640). (0.057). 0.54**. 0.95**. 0.05. 0.90*. (0.037). (0.010). (0.896). 治 -0.83*** 政 -0.54* 大. (0.088). (0.076). (0.009). (0.502). -0.31*. -0.39. -0.22. -0.16. (0.092). (0.134). (0.406). (0.663). 0.02. -0.34. 0.46*. (0.928). (0.169). (0.064). 0.18. 0.10. 0.29. (0.740). (0.305). -1. 0. 1. io. al. n. (0.370) -0.08 6 (0.675). y. Nat. 5. ‧. 4. 學. 3. 立. ‧ 國. (0.002). 0.30. sit. 2. er. -0.67***. iv. n0.04 C h-0.17 U engchi (0.534) (0.864). -0.80** (0.025) -0.19 (0.640) -0.21 (0.559). -0.01. 0.05. -0.08. 0.13. (0.954). (0.863). (0.734). (0.723). 0.02. -0.09. 0.16. -0.25. (0.908). (0.750). (0.577). (0.537). 0.00. -0.16. 0.19. -0.35. (0.989). (0.584). (0.463). (0.371). 0.14. 0.11. 0.18. -0.06. (0.443). (0.663). (0.498). (0.867). 7. 8. 9. 10. 25.

(26) Table4. Cumulative Abnormal Returns over the Windows [0, 1], [-1, 1] and [-3, 3] around the Announcement Please refer sample collection to Table1. Family acquiring firms are identified by the principle of any of the following conditions is satisfied: (1) A family, an individual or a firm controlled by the family own at least 10% of ultimate share-holdings; meanwhile, there is at least one family member in board of director; (2) More than half of the board members are family members. CARs denote cumulative abnormal returns (%) which are measured by OLS Risk-Adjusted Market Model, with the benchmarks of TSE and OTC market index return and estimation period over [-375, -15] from the announcement. Column 4 indicates the difference between CARs of family and nonfamily firms. The difference tests are based on t-tests for equality in means. P-values are in brackets. ***Statistical significance at the 1 % level. **Statistical significance at the 5% level. *Statistical significance at the 10% level.. CAR. All. (%). (1). 1.10***. 1.85***. 0.16. 1.69**. (0.001). (0.000). (0.734). (0.013). 1.24***. 2.20***. 0.04. (0.002). (0.000). (0.941). n. [-3, 3]. (0.556). y. 2.17***. sit. (0.006). er. io. al. 0.36. (4)=(2)-(3). ‧. ‧ 國. (3). Nat. [-1, 1]. Difference. (2). 學. [0, 1]. 立. 政Family治 Nonfamily 大. Ch. 1.70*. e(0.069) ngchi. i n U. v. -1.29*. 3.00**. (0.063). (0.011). By comparison, the CARs of family firms outperform those of nonfamily firms by 1.69%, 2.17% and 3% one day, three days and seven days around the announcement. It concurs with the consequences obtained by Ben-Amar and Andre (2006), claiming that family firms on average obtain greater abnormal returns (2.1%) than nonfamily firms (0.2%) in Canada. Besides, it can be inferred from Table 4 that the shareholders of family acquirers do not perceive families expropriate their interests and cause 26.

(27) value destruction by mergers and acquisitions. Table 5 further indicates that there are important differences between family and nonfamily firms with respect to firm size and book to market. Moreover, the intension of family firms and nonfamily firms to purchase public targets also exhibits inconsistence. In Table 5, firm size of family acquirers is on average NT 36,234 thousand dollars smaller than firm size of nonfamily acquirers. In our sample, BM of. 政治大. family firms is higher than BM of nonfamily firms, which means family firms are tend. 立. to be valued corporations. Besides, family firms have lower intention to acquire. ‧ 國. 學. public targets. As we can see, family firms acquire 22% of public targets which is less. ‧. than 47% of public targets acquired by nonfamily firms.. Nat. io. sit. y. Furthermore, table 6 also indicates that the correlation between family firm. er. status and firm size, BM as well as public target are -0.274, 0.164 and -0.264. al. n. v i n C hthat the size effect,Uthe listed effect and lower BM respectively. Earlier studies argue engchi. may negatively influence M&A performance of acquirers (Faccio, McConnell and Stolin , 2006; Loderer and Martin, 1990; Moeller, Schlingemann and Stulz, 2004; Rau and Vermaelen ,1998; Schwert, 2000). Thus, do family firms in our sample obtain higher CARs because of their smaller firm size, private target acquiring or higher BM? We wonder whether family firm status actually has a positive impact on M&A performance. 27.

(28) Besides in Table 5, the deviation between voting rights and cash flow rights of family acquirers presents insignificantly unequal to the one of nonfamily acquirers. It means that compared to nonfamily firms, family acquirers do not have more serious conflicts between large and minority shareholders. However, does the deviation cause no impacts on family acquiring performance? We wonder whether the large-minority shareholder conflicts have such an insignificant effect that family. 政治大. acquirers in our sample outperform nonfamily acquirers.. 立. Empirical result. 學. ‧ 國. B.. In this section, we will do the regressions step by step to examine the effect of. ‧. family firm status on M&A performance and analyze why the effect exists. First, it’s. Nat. io. sit. y. simple regression between family firm status and the CARs of acquirers. Second, we. er. will control for acquiring firm characteristics and test weather family firm status still. al. n. v i n affects the CARs. Third, we willCextra h einclude h i Utarget and deal characteristics, n g cpublic repeating the third step. Forth, we will create an interaction variable between family firm status and firm size to make robust analysis on the influence of family firm status. Finally, we will regress on CARs of family acquirers and the deviation between voting rights and cash flow rights.. 28.

(29) Table5. Firm characteristics Please refer sample collection and family definition to Table1. Firm size is the average book value of total assets with a unit of thousand NT dollars as of the calendar year-end prior to the announcement. BM is the average equity book to market ratio as of the calendar year-end prior to the announcement. Prior return is the average CAR of acquirers over the period from 15 trading days prior to the announcement to 3 days prior to the announcement. Public target is the proportion of public targets to the total targets. Year after 2007 is the proportion of the transactions occurred in 2007~2013 to those occurred in 1999~2013. Cash payment is the proportion of transactions paid in cash to all transactions paid both in cash and stock. The deviation is the average difference between voting rights and cash flow rights. Column 4 indicates the difference between each variable of family and nonfamily firms. The difference tests are based on t-tests for equality in means. P-values are in brackets.. 政治大. ***Statistical significance at the 1 % level. **Statistical significance at the 5% level. *Statistical. Nonfamily (3). 23,014. 59,248. 0.72. 0.60. y. al. sit. 0.67. io. Book to Market. Nat. NT dollars). 39,330. Family (2). ‧. (unit: thousand. 立. er. Firm Size. All (1). 學. Summery Description. ‧ 國. significance at the 10% level.. 33%. 22%. 47%. Year after 2007(%). 57%. 53%. 61%. Cash Payment (%). 17%. 17%. 17%. Deviation. 7.85. 7.97. 7.71. n. 1.85. iv n U -0.34. Prior Return (%). 0.87. Public Target (%). Ch. engchi. 29. Difference (4)=(2)-(3) -36,234*** (0.005) 0.12** (0.029) 2.19 (0.120) -24%*** (0.000) -8% (0.306) 0% (0.981) 0.25 (0.869).

(30) Table6. Correlation Data Please refer sample collection to Table 1. Family firms are identified by the principle of any of the following conditions is satisfied: (1) A family, an individual or a firm controlled by the family own at least 10% of ultimate share-holdings; meanwhile, there is at least one family member in board of director; (2) More than half of the board members are family members. Firm size is the log of book value of total assets as of the calendar year-end prior to the announcement.. 治政 from the announcement. Public target is the dummy variable which is equal to 1 for public 大 target acquiring; 0 otherwise. Year after 2007 is equal to 1 for the 立 transactions occurred in 2007~2013; and is equal to 0 for those occurred in 1999~2006. Cash payment is the dummy variable equal to 1 for the transactions 學. ‧ 國. BM is the equity book to market ratio as of the calendar year-end prior to the announcement. Prior return is the CARs of acquirers over the period [-15, -3]. paid in cash; 0 for those paid in stock. The deviation is the difference between the level of control rights and the level of cash flow rights.. Family Firm. BM. Prior Return. Public Target. Year after 2007. Family Firm. 1.000. Firm Size. -0.274. 1.000. BM. 0.164. 0.012. Prior Return. 0.120. -0.163. 0.190. Public Target. -0.264. 0.338. 0.018. Year after 2007. -0.087. -0.059. 0.021. -0.069. 0.188. 1.000. Cash Payment. -0.001. -0.079. 0.051. -0.090. -0.190. 0.330. 1.000. Deviation. 0.008. -0.179. 0.090. -0.031. -0.113. -0.028. -0.004. Deviation. Nat. io. sit. y. Cash Payment. 1.000. n. al. er. Firm Size. ‧. Variables. Ch. 1.000. e-0.098 ngchi. 30. i n U. v. 1.000. 1.000.

(31) (a) Equation A: simple regression analysis Table 7 reports that family firm status has a significant impact on M&A performance of acquiring firms. Regression (1) and (2) indicate that the CARs of family acquirers exceed nonfamily acquirers by 1.69% and 2.17% respectively over the windows [0, 1] and [-1, 1] around the announcement. Totally speaking, the outcome of simple regression is consistent with the result obtained by the. 政治大. two-sample mean-comparison test.. 立. (b) Equation B: multiple regression analysis controlling for acquiring firm. ‧ 國. 學. characteristics. ‧. From Equation A, we find that family firm status has a significant positive impact. y. Nat. al. er. io. sit. on the CARs of acquirers. However as Table 5 and Table 6 indicate, do family firms. v. n. perform better because of smaller firm size, private target acquiring or higher BM? In. Ch. engchi. i n U. this section, we will control for acquiring firm characteristics such as firm size, BM and prior return to examine whether the effect of family firm status persists. As we can see in Table 7, family firms generate greater benefits by M&A transactions than nonfamily firms over the windows [0, 1] and [-1, 1] around the announcement. However, the coefficients of family firm status decrease from 1.69% and 2.17% in Equation A to1.26% and 1.58% in Equation B respectively when taking firm characteristics into consideration. 31.

(32) The control variables generally have expected signs. However, only the coefficients of firm size are significantly different from zero. As in most of the literatures, large acquiring firms make poor M&A decisions than small acquiring firms (Loderer and Martin, 1990; Schwert, 2000). Other control variables including BM and prior return have an insignificant impact on the CARs. Although BM of family acquirers apparently differs from that of nonfamily acquirers, there is insufficient. 政治大. evidence to attribute the higher CARs to higher BM.. 立. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 32. i n U. v.

(33) Table7. Multiple Regressions of Family Firm Status and Acquiring Firm Performance: Control for Acquiring Firm Characteristics Please refer sample collection to Table1. CARs denote cumulative abnormal returns (%) which are measured by OLS Risk-Adjusted Market Model, with the benchmarks of TSE and. OTC market index return and estimation period over [-375, -15] from the announcement. Independent variable is family firm status which is equal to 1 for family acquiring firms; 0 otherwise. The identification of family firms are based on any of the two criteria: (1) A family, an individual or a firm controlled by the family own at least 10% of ultimate share-holdings; meanwhile, there is at least one family member in board of director; (2) More than half of the board members are family members. Control variables including firm size, BM and prior return: firm size is the log of book value of total assets as of the calendar year-end prior to the announcement; BM is the equity book to market ratio as of the calendar year-end prior. 政治大. to the announcement; prior return is the CARs of acquirers over the period [-15, -3] from the announcement. P-values are in brackets. ***Statistical significance at the 1 % level. **Statistical. 立. -. (2). 1.69**. 2.17***. (0.012). (0.006). 1.26* (0.075). y. Firm Size. (1). CAR [0, 1] (3). io. sit. +/-. CAR [-1, 1]. n. al. Ch. engchi. i n U. v. CAR [-1, 1] (4) 1.58* (0.053). -0.93* (0.051). -1.22** (0.026). er. Family Firm Status. CAR [0, 1]. ‧. Expected sign. Nat. Independent Variable. 學. ‧ 國. significance at the 5% level. *Statistical significance at the 10% level.. Book to Market. +. 0.92 (0.315). 1.09 (0.305). Prior Return. +/-. -0.03 (0.444). -0.02 (0.654). 0.16. 0.04. 6.26*. 8.15**. (0.754). (0.946). (0.069). (0.040). No. of observations. 169. 169. 169. 169. Adj-R2. 0.031. 0.039. 0.041. 0.055. Intercept. 33.

(34) (c) Equation C: multiple regression analysis controlling for firm characteristics and deal characteristics According to Equation B, it is convinced that family firm status actually has the positive influence on acquiring firm performance even after controlling for acquiring firm characteristics. However, do family firms still generate greater CARs when controlling for more variables such as public target or cash payment? In this section,. 政治大. we will conduct the regressions including all the control variables: : firm size, book to. 立. market, prior return, public target, year after 2007 and cash payment.. ‧ 國. 學. Table 8 indicates that as we control for more variables, the coefficients of family. ‧. firm status decrease from 1.26% and 1.58% in Equation B to 1.11% and 1.49% in. Nat. io. sit. y. Equation C one day and three days around the announcement respectively. Besides,. er. the impact of family firm status on acquiring firm performance is significant merely in. al. n. v i n regression (2). The insignificantC coefficient U status in regression (1) may h e n gofcfamily h i firm. be caused by two reasons: First, the CAR over the window of [0, 1] around the announcement is lower and less significant than the CAR three days around the announcement (See Table 4); Second, the correlation between firm size and public target leads to errors (See Table 6). Furthermore, Table 9 also reports that the coefficients of family firm status in regression (1) and (2) are significantly positive when dropping either public target or firm size. 34.

(35) Table8. Multiple Regressions of Family Firm Status and Acquiring Firm Performance: Control for Firm Characteristics and Deal Characteristics Please refer sample collection and family definition to Table1; calculations of the CARs to Table 4. Independent variable is family firm status which is equal to 1 for family acquiring firms; 0 otherwise. Control variables including the followings: firm size is the log of book value of total assets as of the calendar year-end prior to the announcement; BM is the equity book to market ratio as of the calendar year-end prior to the announcement; prior return is the CARs of acquirers over the period [-15, -3] from the announcement; public target is a dummy variable equal to 1 for public targets; 0 otherwise. Year after 2007 is a dummy variable equal to 1 for the transactions occurred in 2007~2013 and equal to 0 for those occurred in 1999~2006.Cash payment is a dummy variable equal to 1 for the transactions paid in cash; 0 otherwise. P-values are in brackets. ***Statistical significance at the 1 % level.. 政治大 CAR [0, 1]. **Statistical significance at the 5% level. *Statistical significance at the 10% level.. 立 Expected sign. (1) 1.11 (0.126). Firm Size. -. -0.84* (0.095). 1.49* (0.076) -1.08* (0.063). y. n. al. (0.255). Ch +/-. i n U. -0.04 (0.358). engchi. sit. 1.05. +. er. io. Prior Return. ‧. +/-. Nat. Family Firm Status. Book to Market. CAR [-1, 1] (2). 學. ‧ 國. Independent Variable. v. 1.16 (0.278) -0.02 (0.611). Public Target. -. -0.79 (0.328). -0.80 (0.391). Year after 2007. +/-. 0.13 (0.859). 0.64 (0.458). Cash Payment. +/-. -1.08 (0.269). -0.85 (0.455). Intercept. 6.00* (0.098). 7.21* (0.085). No. of observations. 169. 169. Adj-R2. 0.034. 0.044. 35.

(36) Table9. Multiple Regressions of Family Firm Status and Acquiring Firm Performance: Control for Firm and Deal Characteristics without either Firm Size or Public Target Please refer sample collection and family definition to Table1; calculations of the CARs to Table 4. Independent variable is family firm status which is equal to 1 for family acquiring firms; 0 otherwise. Control variables including the followings: firm size is the log of book value of total assets as of the calendar year-end prior to the announcement; BM is the equity book to market ratio as of the calendar year-end prior to the announcement; prior return is the CARs of acquirers over the period [-15, -3] from the announcement; public target is a dummy variable equal to 1 for public targets; 0 otherwise. Year after 2007 is a dummy variable equal to 1 for the transactions occurred in 2007~2013 and equal to 0 for those occurred in 1999~2006.Cash payment is a dummy variable equal to 1 for the transactions paid in cash; 0 otherwise. P-values are in brackets. ***Statistical significance at the 1 % level.. 政治大 CAR [0, 1]. **Statistical significance at the 5% level. *Statistical significance at the 10% level.. 立 Expected sign. (1) 1.23* (0.084). Firm Size. -. -0.98** (0.042). io. n. al. (0.282). Ch +/-. i n U. -0.03 (0.381). engchi. sit. 0.99. +. er. Nat. Prior Return. y. +/-. 1.36* (0.056). ‧. Family Firm Status. Book to Market. CAR [0, 1] (2). 學. ‧ 國. Independent Variable. v. 0.94 (0.310) -0.03 (0.491) -1.19 (0.129). Public Target. -. Year after 2007. +/-. -0.07 (0.922). 0.31 (0.672). Cash Payment. +/-. -0.82 (0.383). -1.10 (0.265). Intercept. 6.77* (0.056). 0.14* (0.876). No. of observations. 169. 169. Adj-R2. 0.034. 0.023. 36.

(37) In Table 8, the coefficients of firm size significantly differ from zero in regression (1) and (2) with other control variables insignificantly effective. As we include public target, the size effect is partially diminished. The result concurs with the arguments which indicate that large acquirers tend to purchase public targets and have negative CAR, whereas small acquirers tend to buy private targets and have positive CAR (Faccio, McConnell and Stolin, 2006). Although there are apparent differences of BM. 政治大. and public target between family and nonfamily acquiring firms, there is no sufficient. 立. evidence to attribute the higher CARs of family acquiring firms to higher BM or. ‧ 國. 學. private targets acquiring.. ‧. (d) Equation D: multiple regression analysis controlling for the interaction variable. Nat. io. sit. y. From Equation A, B and C, we find that both family firm status and firm size. er. have a significant impact on the CARs of acquirers. However as we can see in Table 5,. al. n. v i n C h than nonfamilyUfirm size. Beside, Table 6 also family firm size is significant smaller engchi shows that the correlation between family firm status and firm size is -0.274. There is probability that the influence of family firm status is a size effect in disguise. To address this concern, we create an interaction variable between family firm. status and firm size to examine the effect of family firm status. As Table 10 reports, the coefficients of family firm status are significantly positive, which means that family firms persistently obtain better M&A performance without those effect 37.

(38) simultaneously caused by firm size. We can also find in Table 10 that the coefficients of family firm status in regression (1) and (2) are 7.63% and 11.43% respectively. It is much higher than those in Equation A because of the negative impact from the interaction between family firm status and firm size. Thus, it is convinced that the effect of family firm status on acquirers’ stock returns is not a size effect in disguise. Table10. Multiple Regressions of Family Firm Status and Acquiring Firm. 政治大. Performance: Control for the Interaction Variable. 立. Please refer sample collection and family definition to Table1; calculations of the CARs to Table 4. Independent variable is family firm status which is equal to 1 for family acquiring. ‧ 國. 學. firms; 0 otherwise. Interaction Variable is the one between family firm status and firm size. Firm size is the log of book value of total assets as of the calendar year-end prior to the. ‧. announcement. P-values are in brackets. ***Statistical significance at the 1 % level. **Statistical significance at the 5% level. *Statistical significance at the 10% level.. CAR [0, 1]. n. al. (1). Ch. e 7.63* ngchi. CAR [-1, 1]. er. io. Variable. sit. y. Nat. Independent. i n U. v. (2). (0.093). 11.43** (0.029). -0.88 (0.185). -1.37* (0.074). 0.16. 0.04. (0.754). (0.946). No. of observations. 169. 169. Adj-R2. 0.035. 0.052. Family Firm Stats. Family Firm Stats * Firm Size. Intercept. 38.

(39) (e) The relation between family acquiring performance and the deviation between voting rights and cash flow rights Based on the results in Equation A to Equation D, we find that family acquirers persistently obtain higher CARs no matter what control variables we include. In this section, we wonder whether the deviation between voting rights and cash flow rights of family acquirers causes such an insignificant impact that family firms in our. 政治大. sample perform better than nonfamily firms.. 立. As La Porta, Lopez-de-Silanes and Shleifer (1999) show, the firms in most of. ‧ 國. 學. Continental Europe and Asia are characterized by greater ownership concentration in. ‧. the hands of the individuals or families leading to the large-minority shareholder. Nat. io. sit. y. conflicts. Since ownership structure in Taiwanese corporations exhibits high levels of. er. concentration, the conflicts between large and minority shareholders may probably. al. n. v i n Cacquiring exist and negatively affect family performance. h e n gfirmc h i U. Although families may also bring benefits to the corporations such as elimination of agency problems, long-term horizon with positive NPV projects choosing and the reputation effect (Anderson et al. 2002; Anderson and Reeb, 2003; Casson, 1999; Chami, 1999; Demsetz and Lehn, 1985; Stein, 1988, 1999). However, these benefits may hard to realize if the large-minority shareholder conflicts exist.. 39.

(40) Table11. Multiple Regressions of Family Acquiring Performance and the Deviation between voting rights and cash flow rights The sample contains all Taiwanese M&A transactions between 1999 and 2013. Please refer detail sample selection and family definition to Table1; calculations of the CARs to Table 4. Independent variable is the deviation which is calculated by the difference between the level of voting rights and the level of cash flow rights. Control variables including the followings: firm size is the log of book value of total assets as of the calendar year-end prior to the announcement; BM is the equity book to market ratio as of the calendar year-end prior to the announcement; prior return is the CARs of acquirers over the period [-15, -3] from the announcement; public target is a dummy variable equal to 1 for public targets; 0 otherwise. Year after 2007 is a dummy variable equal to 1 for the transactions occurred in 2007~2013. 政治大 1 for the transactions paid in cash; 0 otherwise. P-values are in bracket. 立. and equal to 0 for those occurred in 1999~2006.Cash payment is a dummy variable equal to ***, ** and *. represent statistical significance at the 1 %, 5% and 10% level.. (0.310). n. al. Ch -. sit. io Firm Size. y. -0.05. -. i n U. -1.21 (0.12). engchi. er. Nat. Deviation. CAR [0, 1] (1). Expected sign. ‧. ‧ 國. 學. Independent Variable. v. CAR [-1, 1] (2) -0.07 (0.267) -1.68* (0.065). Book to Market. +. -0.06 (0.961). 0.10 (0.950). Prior Return. +/-. 0.01 (0.825). 0.00 (0.993). Public Target. -. 1.06 (0.390). 0.93 (0.518). Year after 2007. +/-. 0.02 (0.988). 0.15 (0.906). 40.

(41) Table.11 (continued) -1.28 (0.351). -0.31 (0.845). Intercept. 10.43* (0.056). 13.82** (0.031). No. of observations. 94. 94. Adj-R2. -0.021. -0.025. Cash Payment. +/-. Thus in our study, we focus on the impact of the large-minority shareholder conflicts.. 政治大. We use the factor of the deviation between voting rights and cash flow rights to. 立. examine why family firms perform better. The deviation is a common proxy for the. ‧ 國. 學. existence of the conflicts between large and minority shareholders. As the deviation. ‧. gets higher, large shareholders have greater intension to exploit the interests of. y. Nat. affect the acquiring performance of family firms.. n. al. Ch. engchi. er. io. sit. minority shareholders through M&A transactions. Thus, the deviation may negatively. i n U. v. As Table 5 reports that the deviation between voting rights and cash flow rights of family acquirers presents insignificantly unequal to the one of nonfamily acquirers. It means family acquirers compared to nonfamily firms do not have more serious conflicts between large and minority shareholders. Furthermore, Table 11 indicates that the deviation insignificant affect family acquirers’ CARs over the windows [0, 1] and [-1, 1] around the announcement. Thus, there is insufficient evidence to the presence of the large-minority shareholder conflicts in Taiwanese family acquirers. 41.

(42) V. Conclusion Our study focuses on the relation between family firms and acquiring firm performance. As numerous studies have contrast arguments, the effect of family firm is not a theoretical but an empirical question that we examine in our study. The results of our regressions indicate that the CARs of family acquirers exceed those of nonfamily firms by 1.69%, 2.17% and 3% one day, three days and seven days. 政治大 acquiring performance persists 立 even after we control for the variables including firm around the announcement respectively. The positive effect of family firms on. ‧ 國. 學. size, BM, prior return, public target, year and mode of payment.. ‧. Besides, we analyze the relation between family firm status and the deviation. sit. y. Nat. between voting rights and cash flow rights to examine why family acquirers perform. er. io. better. In the sample of Taiwanese M&A, the deviation is not the significant factor to. al. n. v i n negatively influence family acquiring C h performance. AsUa result, family acquirers with engchi advantage of eliminating agency problems may generate more benefits than nonfamily acquirers. In our study, we do not analyze the positive factors such as long-term perspectives on investments and the reputation effect caused by family ownership. Besides, we do not examine the M&A performance between family and nonfamily acquirers from a long-term perspective. This remains a challenge for further studies.. 42.

(43) VI. Reference 1.. 葉銀華 (1999)，’家族控股、董事會組成與經營績效─台灣家族企業管治機制之研究’，行政院國家科學委員會專題研究計畫成果報告。. 2.. Amihud, Y. and B. Lev (1981), ‘Risk Reduction as Managerial Motive for Conglomerate Mergers’, Bell Journal of Economic, Vol. 12, pp. 605-17.. 3.. Anderson, R.C., and D.M. Reeb (2003), ‘Founding-Family Ownership and Firm Performance: Evidence from the S&P 500’, Journal of Finance, Vol. 58, No. 3, pp. 1301-1328.. 4.. Anderson, R., S. Mansi, and D. Reeb (2003), ‘Founding Family Ownership and the Agency cost of debt’, Journal of Financial Economics, Vol. 68, pp. 263-285.. 政治大. 5.. Andrade, G., M. Mitchell and E. Stafford (2001), ‘New Evidence and Perspectives on Mergers’, Journal of Economic Perspectives, pp. 103-20.. 6.. Avery, C., J.A. Chevalier and S. Schaefer (1998), ‘Why Do Managers Undertake Acquisitions? An Analysis of Internal and External Rewards for Acquisitiveness’, Journal of Law, Economics and Organizations, Vol. 14, pp. 24-43.. 7.. Barclay, M., and C. Holderness (1989), ‘Private Benefits from Control of Public Corporations’, Journal of Financial Economics, Vol. 25, pp. 371-396.. 8.. Barnes, L.B., and S.A. Hershon (1976), ‘Transferring Power in the Family Business’, Harvard Business Review, Vol. 54, No. 4, pp. 105-114.. 立. ‧. ‧ 國. 學. er. io. sit. y. Nat. al. n. v i n C h(2004), ‘The EffectU of Ownership Structure and Barontini, R., and L. Caprio e n gPerformance: chi Family Control on Firm Value and Evidence from Continental. 9.. Europe’, Working Paper. 10. Ben-Amar, W., P. Andre (2006), ‘Separation of Ownership from Control and. Acquiring Firm Performance: The Case of Family Ownership in Canada’, Journal of Business Finance & Accounting, pp. 517-543. 11. Bigelli, M., and S. Mengoli (2004), ‘Sub-optimal Acquisition Decision under a. Majority Shareholder System’, Journal of Management and Governance, Vol. 8, pp. 373-405. 12. Bohmer, E. (2000), ‘Business Group, Bank Control, and Large Shareholders: An. Analysis of German Takeovers’, Journal of Finance Intermediation, Vol. 9, pp. 117-48. 43.

(44) 13. Bruner, R.F. (2002), ‘Does M&A Pay? A Survey of Evidence for the Decision. Maker’, Journal of Applied Finance, Vol. 12, pp. 48-68. 14. Casson, M., (1999), ‘The Economics of the Family Firm’, Scandinavian Economic. History Review, Vol. 47, pp. 10-23. 15. Chang, S. (1998), ‘Takeovers of Privately Held Targets, Methods of Payment and. Bidder Return’, Journal of Finance, Vol. 53, pp. 773-84. 16. Chami, R. (1999), ‘What’s Different about Family Business?’, Notre Dame. University and the International Monetary Fund Working Paper. 17. Claessens, S., S. Djankov, J. Fan and L.H.P. Lang (2000), ‘The Separation of. Ownership and Control in East Asian Corporations’, Journal of Financial Economics, Vol. 58, pp. 81-112.. 立. 政治大. 18. Claessens, S., S. Djankov, J. Fan and L.H.P. Lang (2002), ‘Disentangling the. ‧ 國. 學. Incentive and Entrenchment Effects of Large Shareholders’, Journal of Finance, Vol. 57, pp. 2741-71.. ‧. 19. Cronqvist, H., and M. Nilsson (2003), ‘Agency Cost of Controlling Minority. sit. y. Nat. Shareholders’, Journal of Financial and Quantitative Analysis, Vol. 38, pp. 695-719.. er. io. 20. Davis, J.F., D. Schoorman, and L. Donaldson (1997), ‘Toward a Stewardship. Theory of Management’, Academy of Management Review, Vol. 22, pp. 20-47.. al. n. v i n C h ‘the Structure ofUCorporate Ownership: Causes Demsetz, H., and K. Lehn (1985), e n gEconomy, c h i Vol. 93, pp. 1155-1177. and Consequences’, Journal of Political. 21.. 22. Dong, M., D. Hirshleifer, S. Richardson, S.H. Teoh (2002), ‘Does investor. misevaluation drive the takeover market?’, Ohio State University Working Paper. 23. Dyck, A. and L. Zingales (2004), ‘Private Benefits of Control: An International Comparison’, Journal of Finance, Vol. 59, No. 2, pp. 537-600. 24. Faccio, M., J.J. McConnell and D. Stolin (2006), ‘Return to Acquirers of Listing. and Unlisted Targets’, Journal of Financial and Quantitative Analysis, Vol. 41, No. 1, pp. 197-220. 25. Faccio, M. and L.H.P. Lang (2002), ‘The Ultimate Ownership of Western. European Corporations’, Journal of Finance Economics, Vol. 65, pp. 365-95. 44.

(45) 26. Faleye, O. and M. Huson (2002), ‘Understanding Acquisition Return: The role of. Corporate Governance’, University of Alberta Working Paper. 27. Fama, E., and M. Jensen (1985), ‘Organizational Forms and Investment. Decisions’, Journal of Financial Economics, Vol. 14, pp. 101-119. 28. Firth, M. (1991), ‘Corporate Takeovers, Stockholder Return and Executive. Reward’, Managerial and Decision Economics, Vol. 12, pp. 421-28. 29. Gomez-Mejia, L., M. Nunez-Nickel, and I. Gutierrez (2001), ‘The Role of Family. Ties in Agency Contracts ’, Academy of Management Journal, Vol. 44, pp. 81-95. 30. James, H., (1999), ‘Owner as Manager, Extended Horizons and the Family Firm’,. International Journal of the Economics of Business, Vol. 6, pp. 41-56.. 政治大. 31. Jensen, M.C (1986), ‘Agency Costs of Free Cash Flow, Corporate Finance and. 立. Takeovers’, American Economic Review, Vol. 76, pp. 323-29.. ‧ 國. 學. 32. Jensen, M.C., W.H. Meckling (1976), ‘Theory of the Firm: Managerial Behavior,. ‧. Agency Costs and Ownership structure’, Journal of Financial Economics, Vol. 3, pp. 305-360.. y. Nat. 33. Johnson, S., R. La Porta, F. Lopez-de-Silanes, and A. Shleifer (2000), ‘Tunneling’,. io. sit. American Economics Review, Vol. 90, pp. 20-27.. n. al. er. 34. La Porta, R., F. Lopez-De-Silanes and A. Shleifer (1999), ‘Corporate Ownership. i n U. v. around the World’, Journal of Finance, Vol. 54, pp. 471-517.. Ch. engchi. 35. La Porta, R., F. Lopez-De-Silanes and A. Shleifer (2002), ‘Investor Protection and. Corporate Valuation’, Journal of Finance, Vol. 57, pp. 1147-70. 36. Lee, J.Z., and Y.H. Liao (2004), ‘Board of Director Characteristics and Earnings. Management-evidence from the Effect of Family-controlling on Taiwan Corporations’, Taiwan Accounting Review, Vol. 5, No. 1, pp. 1-39. 37. Lemmons, M.L., and K.V. Lins (2003), ‘Ownership Structure, Corporate. Governance, and Firm Value: Evidence from the East Asian Financial Crisis’, Journal of Finance, Vol. 58, pp. 1445-68. 38. Lewellen, W., C. Loderer and A. Rosenfeld (1985), ‘Merger Decisions and. Executive Stock Ownership in Acquisition Firms’, Journal of Accounting and Economics, Vol. 7, pp. 209-231. 45.

(46) 39. Loderer, C., K. Martin (1990), ‘Corporate Acquisitions by Listed Firms: The. Experience of a Comprehensive Sample’, Financial Management, Vol. 19, pp. 17-33. 40. Moeller, S.B., F.P. Schlingemann, and R.M. Stulz (2004), ‘Firm Size and the Gains. from Acquisitions’, Journal of Financial Economics, Vol. 73, pp. 201-228. 41. Morck, R., A. Schleifer, and R.W. Vishny (1990), ‘Do managerial objectives drive. bad acquisition?’, Journal of Finance, pp. 31-47. 42. Morck, R., A. Schleifer, and R.W. Vishny (1988), ‘Management Ownership and. Market Valuation: An Empirical Analysis’, Journal of Financial Economics, Vol. 20, pp. 293-315.. 治政大 Columbia University Working Paper. 立. 43. Peres-Gonzalez, F. (2001), ‘Does Inherited Control Hurt Firm Performance?’,. ‧ 國. 學. 44. Rau, P.R., and T. Vermaelen (1998), ‘Glamour, Value & the Post-acquisition. Performance of Acquiring Firms’, Journal of Financial Economics, Vol. 49, pp. 223-253.. ‧. 45. Schwert, G.W. (2000), ‘Hostility in Takeovers: In the Eye of Beholder?’, Journal of. sit. y. Nat. Finance, Vol. 55, pp. 2599-2640.. er. io. 46. Shleifer, A. and R.W. Vishny (1997), ‘A Survey of Corporate Governance’, Journal. of Finance, Vol. 52, pp.737-83.. al. n. v i n C h Trap: How Companies Sirower, M. (1997), The Synergy Lose the Acquisition U i e h n Game (New York: The Free Press). g c. 47.. 48. Stein, J. (1988), ‘Takeover Threats and Managerial Myopia’, Journal of Political. Economy, Vol. 96, pp. 61-80. 49. Stein, J. (1989), ‘Efficient Capital Market, Inefficient Firms: A Model of Myopic. Corporate Behavior’, Quarterly Journal of Economy, Vol. 103, pp. 655-669. 50. Travlos, N.G. (1987), ‘Corporate Takeover Bids, Method of Payment, and Bidding. Firms’ Stock Return’, Journal of Finance, Vol. 42, pp. 943-963. 51. Zhang, G. (1998), ‘Ownership Concentration, Risk Aversion and the Effect of. Financial Structure on Investment Decision’, European Economic Review, Vol. 42, pp. 1751-78. 46.

(47)