The Takeover Decision and Executive Compensation Incentives

Transcription

1 The Takeover Decision and Executive Compensation Incentives Isabel Feito-Ruiz University of Leon. Department of Business Administration, Campus Vegazana s/n. E Leon (Spain). Luc Renneboog Tilburg University. Department of Finance, P.O. Box 90153, 5000LE Tilburg (Netherlands) Abstract The aim of this paper is to analyze if the CEO s equity-based compensation influences the acquiring shareholder valuation around the M&A announcement in Europe. We analyze 216 M&As in the period Our results show that acquiring shareholders value the announcement of an M&A higher in firms where CEO s (or executives) receive more equitybased compensation. However, in firms where the largest shareholder has high levels of ownership concentration, acquiring shareholders value worse M&As if CEOs (or executives) earn more equity-based compensation. This result is consistent with a substitute effect between ownership and equity based compensation as internal corporate governance mechanisms. Moreover, we find that the excessive compensation negatively influences the acquiring shareholder valuation. This result supports the argument that the amount of compensation that acquiring firms pay to CEOs and the rest of executives may be influenced by agency conflicts itself. Keywords: equity-based compensation, mergers and acquisitions (M&As), shareholder protection, creditor protection, ownership concentration, excessive compensation. JEL codes: G30; G32; G34, F30.

2 1. Introduction Corporate investment decisions such as those on mergers and acquisitions (M&As) may be driven by the managers personal objectives such as maximizing personal wealth or private benefits, possibly even at the expense of maximizing shareholder value. Executive compensation contracts, especially equity-based incentives, are considered means to align managerial interests with those of shareholders. Equity-based compensation to top executive may be effective in shaping long-term corporate investment policies and encourage managers to make decisions in the interests of shareholders (Jensen and Ruback, 1983; Shleifer and Vishny, 1988). In this sense, Datta, Iskandar-Datta, and Raman (2001) find a positive relation between abnormal returns for acquiring shareholders and top executive stock option compensation for U.S. M&As. They also find a lower premium paid by managers and more risky investment when these have high levels of equity-compensation. The equity-based compensation is ineffective for high levels of managerial ownership. Therefore, the authors support the efficacy of stock option based compensation to motivate managers to take more risky projects that maximize shareholders value in the absence of effective internal control mechanism. Also for U.S. M&As, Williams and Rao (2006) document that the stock options are effective means for motivating managers to alter their risk incentive behavior. So far, the most of academic studies focused on the relation between the top executive compensation and the M&A performance analyze U.S. or U.K. M&As. To our knowledge there is not any study that analyzes the effect of the top executive equity-based payment on the bidder shareholder valuation for Continental European firms. Continental Europe differs from the U.S. and U.K. in relation to the corporate governance structures (stakeholder-oriented regimes of Continental Europe vs market-oriented regimes of Anglo-American countries). Therefore, there is no empirical evidence regarding the effectiveness of equity-based compensation -stock options and long term incentives stock plans (LTIPs stock)- under the absence of effective external control mechanism. Furthermore, corporate ownership structure is more concentrated in Continental Europe than Anglo-American countries (Barca and Becht, 2001; Faccio and Lang, 2002). In these firms the agency problem between shareholders and managers is lower. Thus, equity-based compensation may play a lower significant role in the alignment of interest of managers and shareholders when a large shareholder controls the firm than in those firms which are widely held. Block holders have more incentives to monitor managers and they can force acquirer managers to examine the acquisition strategies in order to avoid suboptimal risk investment decisions (neither risk-averse nor risk-seeking). After the corporate scandals (such as Enron, WorldCom in the U.S. or Parmalat, Ahold, Vivendi in Europe), doubts have emerged about the effectiveness of executive compensation contracts, 1

3 in particular about stock options, and the corporate governance in general. The public concern about the excess of top manager remuneration shows that executive compensation may be an agency problem itself. In U.S., new rules have been adopted in relation to executive compensation (Sarbanes-Oxley Act 2002) -i.e. more disclosure about executive compensation and limitation of stock option compensation use- in order to control excessive executive payments. In Continental Europe, the disclosure measures about executive compensation, mandatory in the annual corporate governance reports, have been implemented recently. The suboptimal executive compensation contract may provoke great shareholder value losses, since the agency problems is not controlled (Core, Holthausen and Larcker, 1999). Mergers and Acquisitions (M&As) are one of the most important investment decisions made by managers and reflect the effectiveness of corporate governance. Therefore, the analysis of the excessive executive compensation in the M&A framework is a relevant issue in the current debate over the effectiveness of corporate governance. After controlling for the board structure, network, the country corporate governance firm and transaction characteristics, we try to answer the following research questions: 1) Do all components of executive equity-compensation (LTIPs stocks and stock options) have a positive effect on the bidder shareholder valuation when an M&A is announced by European firms? 2) Are differences in the effect of executive equity-compensation on bidder shareholder valuation between concentrated and widely held firms in Europe? 3) Are European executives paid in excess and how this excess influence the acquiring shareholders M&A valuation? Our sample consists of 216 mergers and acquisitions made by European firm during the period This period is characterized by regulatory changes in relation to the executive compensation disclosure and other corporate governance practices around the world. Our results show that acquiring shareholders value the announcement of an M&A higher in firms where CEOs (or executives) receive more equity based compensation. However, in firms where the largest shareholder has high levels of ownership concentration, acquiring shareholders value M&As lower if CEOs (or executives) earn more equity based compensation. This result is consistent with a substitute effect between ownership and equity based compensation as internal corporate governance mechanisms. Moreover, we find that the excessive compensation negatively influences the acquiring shareholder valuation. This result supports the argument that the amount of compensation that acquiring firms pay to CEOs and the rest of executives may be influenced by agency conflicts itself. We contribute both merger and acquisition and executive compensation literature in this sense. The rest of the paper is organized as follows. Section II reviews the literature and formulates the hypotheses. Section III describes the sample and methodology. Section IV presents the data sources and sample characteristics. Section V presents the findings and Section VI concludes. 2

4 2. Related Literature The agency conflict between managers and shareholders in publicly held corporations has been well documented. Shareholder may have problems identifying if managers undertake a project (such as a takeover transaction) with an optimal level of return and risk. Whereas shareholders can diversify away firm-specific risk, managers is frequently undiversified as part of their income and human capital depend on or are invested in their company. In this sense, managers are risk-averse and more likely to pass up value-enhancing risky projects (Smith and Stulz, 1985). To address this agency problem, compensation contracts (arm s-length contracting between shareholders and managers) can embed the right incentives to drive managers towards valueenhancing risky projects. However, not all components of the compensation package have a uniform effect on the risk incentives. Cash compensation, such as a bonus, does not provide the right incentives for managers to increase firm risk given this mechanism s short-term horizon (Lambert and Larcker, 1987). Long term equity-based incentive plans (LTIPs or restricted stock) and stock options may act as a device to incentivize managers to take projects whose payoff is long-term and more risky (Sudarsanam and Huang, 2007). Previous studies focusing on U.S. M&A decisions document that stock options are an effective means to motivate managers to alter their risk incentive behavior and maximize shareholder value in the absence of effective internal control mechanism (Datta et al, 2001; Williams and Rao, 2006) Compensation contract incentives in widely-held firms. There is no empirical evidence that the stock options are also effective means to align shareholder and manager interests when there are not efficient external control mechanisms. In Continental Europe, corporate governance structures are different from the U.S. or U.K. Therefore, the effectiveness of equity-compensation on acquirer shareholder valuation may vary among countries. In this sense, we establish that: Hypothesis 1: Equity-based compensation (stock options and restricted stock) positively influences the European bidder shareholder value Compensation contract incentives in the presence of concentrated ownership. In Continental Europe, ownership is significantly more concentrated than U.S. or U.K. firms (La Porta, Lopez-de-Silanes and Shleifer, 1999; Barca and Becht, 2001; Faccio and Lang, 2002). The median US firm does not have any shareholder owning a share stake of 5% (the disclosure threshold) or more, whereas in the median German firm the largest shareholder has majority control. Most of major shareholder activism is happening behind the scenes (Becht et al, 2009; Cziraki et al, 2010; Renneboog and Cziraki, 2011; McCahery et al., 2012) and they may force management to carefully evaluate the acquisition decision in order to avoid suboptimal risky projects (Shleifer and Vishny, 1986). Thus, higher shareholder control may induce less frequent use of equity-based compensation (Bryan, Nash and Patel, 2006; Fernandes 3

5 et al., 2009). Therefore, the presence of a large shareholder in the bidder firm may neutralize the effect of the equity-based compensation on the bidder shareholder valuation. Still, there is a dark side of the concentrated ownership. Majority shareholders could collude with managers to undertake an M&A that provide them private benefits or reject other valueenhancing projects. Agency problems between large shareholders and minority shareholders may emerge in these firms (Fama and Jensen, 1983). In Continental Europe minority shareholders are less protected and the costs of extract private benefits are lower than the U.S. or U.K. (La Porta et al., 1998; Djankov et al., 2008; Martynova and Renneboog, 2008). Thus, equity based executive compensation may have a negative effect on the bidder shareholder valuation: Hypothesis 2: At a takeover transaction, the equity-based compensation (stock options and restricted stock) of the bidder s management will have no effect on the bidder s shareholder value in the presence of a large monitoring shareholder in the bidder firm Suboptimal CEO compensation contract. While equity-based compensation seeks to minimize the agency costs between managers and shareholders, excessive equity-based compensation may lead to non-value-maximizing behavior. The convex payoff structure of stock option compensation may incentivize managers to engage in risk-seeking behavior and overinvest in non value-enhancing projects (Elson, 2003). LTIPs stocks may be a substitute for stock options to align managers and shareholder interests. These LTIPs stocks may reduce the managerial incentives to take risky valueenhancing projects. In addition, public concern about the excess of top manager remuneration shows that executive compensation may be an agency problem itself (Bechuck and Fried, 2003; Weishbach, 2007). A higher executive compensation may lead to poorer future performance in firms with weaker governance structures given that managers act discretionally in the executive compensation decision and pay themselves high compensation (Core et al., 1999; Bechuck and Fried, 2003; Weishbach, 2007).Therefore, we may expect that excess remuneration exists in these firms. Therefore, a negative effect of excessive compensation on the bidder return could be expected. However, if there is no relation or it is positive this effect will not be related to agency problems. Hypothesis 3: Excessive CEO (executive) compensation is negatively correlated to the bidder shareholder value at a takeover transaction Other corporate governance mechanisms. The board structure, director networks, and the country corporate governance regulation may also influence the valuation of takeovers by acquiring shareholders, which is why we control for the following types of variables: 4

6 - Board structure: the board of directors is an internal control mechanism to promote and protect shareholder interests. We consider the board size, number of executives, independent board, and CEO/Chairman duality. Higher board size has been shown to be associated with less effective boards and more managerial power (Yermarck, 1996). Given that in European countries exist two-tied board systems (executive and supervisory board), the proportion of executives in the board may influence the acquiring shareholder valuation. Boone, Field, Karpoff and Raheja (2007) consider that the effect of the board size on the acquiring shareholder valuation is not clear because of the fact that the board size may be endogenous. It could be determined by the characteristics of the firm. - To supervise the management, the board of directors should comprise a number of independent non-executives members. Previous studies show mixed results in relation to the effect of board independence on the merger performance. Franks, Mayer, and Renneboog (2001) and Sudarsanam and Mahate (2006) show a positive effect, but Weir and Laing (2000), Adams and Ferreira (2007), Harris and Raviv (2008) disagree. Duchin, Matsusaka and Ozbas (2010) find that outside directors are more effective when the costs of information are lower. We lack a clear prediction on the effect of this variable on M&A performance. - CEO/chairman duality is associated with higher CEO compensation (Core et al., 1999) and less effective corporate governance, which induces a negative effect in our models. Professional networks. A CEO s professional network may positively or negatively influence his decisions on takeovers and other types of corporate restructurings. First, being on the board of other firms may yield valuable information about those firms, sectors, and industries (or even interesting views on the evolution of macro-economic variables) such that his takeover decisions will be more informed which could then be reflected in the expected returns (Renneboog and Zhao, 2011). Second, the fact that executive directors have been offered non-executive directorships in other firms may reflect that they have been successful managers in the past and may signal talent (Fama and Jensen, 1983). Third, the counterargument to the above positive relation between the number of nonexecutive directorships and takeover returns is that busy CEOs divide their attention over the firm they actively manage and firms they supervise or about which they develop the strategy. Acquirer returns are also higher in transactions with a seconddegree connection where one acquirer director and one target director serve on the same third board (Cai and Sevilir, 2012). Consequently, it may be that such busy CEOs make poorer takeover decisions. Weak corporate governance may arise if many directors of a specific firm are busy (Fich and Shivdasani, 2006). Ahn, Jiraporn and Kim (2010) document a non-linear relationship between the multiple directorships and the acquiring shareholder valuation. Up to a 5

7 threshold, executive business increases the M&A valuation, but higher levels of business links decrease the M&A valuation by acquiring shareholders. Country Corporate Governance. - Quality of the legal and institutional environment in the bidder country may play an important role: a lower quality may induce more asymmetric information and agency problems (La Porta et al., 1998). Therefore, we expect a positive relation between high quality of the legal and institutional environment and bidder shareholder valuation around M&A announcement (Martynova and Renneboog, 2008). We consider the three index used by Martynova and Renneboog (2008) which are: shareholder protection; minority shareholder protection; and creditor protection to proxy the quality of the legal and institutional environment. The shareholder protection measures the degree of shareholder orientation of a national regulation. The index increases with the number and quality of legal provisions that provide shareholders with effective power to appoint and dismiss the board of directors and to control most of the important corporate decisions on, for instance, equity issues or anti-takeover measures. A higher index score represents a higher likelihood that management acts in the interest of shareholders. The minority shareholder protection depends on the regulatory provisions that increase the relative power of the minority shareholders in the presence of strong majority shareholders. In firms with concentrated ownership, dominant shareholders may extract private benefits of control by influencing managerial decisions for his own benefit (see e.g. Durnev and Kim, 2005). The higher index score the higher minority shareholders' interests protection. The creditor protection depends on the regulatory provisions that allow creditors to force repayment more easily, to take possession of the collateral, or even to gain control over the firm in case of financial distress. The higher index the higher corporate governance with respect to creditor protection Firm and transaction characteristics. It is important to control for several bidder and deal characteristics that have been shown to influence the expected returns in takeovers. Lang, Stulz and Walking (1989) show that firms with a high growth opportunities (large market-to-book ratio) obtain high abnormal returns around the time of the acquisition announcement. Dong, Hirshleifer, Richardson and Teoh (2006) do not agree on this issue as they demonstrate that a negative relation prevails, which leads them to regard the ratio as a proxy for overvaluation rather than growth opportunities. Jensen s (1986) free cash flow hypothesis predicts that unrestrained managers of firms with high free cash flow have more resources at their disposal to engage in empire building acquisitions at the detriment of shareholder value. High free cash flow is therefore related to lower bidder 6

8 returns. High debt levels reduce the future free cash flow and limit managerial discretion, so that leverage could be an effective bonding mechanism (Masulis, Wang and Xie, 2007; Marynova and Renneboog, 2009b) which would lead to more positive bidder returns. - Information leakage, insider trading, market anticipation or trading on rumours related to the takeover deal can be reflected in the price runup. A high runup could thus reflect that part of the information generated by the takeover announcement is already incorporated in the stock prices which could induce negative announcement returns (Martynova and Renneboog, 2008, 2009). - The Target firm status may also matter because takeover negotiations may be less complex and hence lead to lower acquisition costs relative to listed firms. Furthermore, takeovers of unlisted targets are often friendly and without bidder competition. Unlisted firms may also seek an acquired because of liquidity problems which enables the acquirer to offer a lower price (Officer, 2007). That is to say, adverse selection forces the price to drop (Akerlof, 1970). Chang (1998) shows greater gains for the bidder when the target firm is unlisted. The target managers will become shareholders of the new firm with incentives to monitor the bidder managers if the deal is paid with stocks and the risk of the deal will be sharing between bidder and target owners and it will reduce the negative effect of stock payments (Chang, 1998; Officer, Poulsen and Stegemoller, 2009). - Moeller, Schlingemann, and Stulz (2004) find that bidder size is negatively correlated with the bidder return around the M&A announcement. They state that this effect is related to managerial hubris hypothesis (Roll, 1986). We expect that when large firm are subject to larger more agency problems, worse acquisitions are made than their smaller counterparts. A related issue is the relative deal size: a larger target firm may be more transparent and fewer adverse selection problems in its valuation may arise (Asquith, Bruner and Mullins, 1983). However, the larger target size may generate higher integration costs (Agrawal et al., 1992), which affects bidder expected returns negatively. - Morck, Shleifer and Vishny (1990), Lang and Stulz (1994) and Berger and Ofek (1995) document that the takeover diversification (when bidder and target are in unrelated industries) diminishes the acquirer s wealth due to managers tendency to overpay. On contrast, Jensen and Ruback (1983) and Campa and Kedia (2002) associate the diversification with wealth creation in M&As transactions, which makes the expected relation ambiguous. - The method of payment (and the sources of financing) may also influence the announcement returns. Managers prefer to offer equity if they believe that their shares are overvalued (Myers and Majluf, 1984). Consequently, an acquisition announcement 7

9 paid with equity will emit a negative signal and thus trigger a negative price reaction (Travlos, 1987; Sudarsanam and Mahate, 2003; Moeller et al., 2004; Martynova and Renneboog, 2009). - Earlier studies found mixed results regarding the effect of cross-border M&As on the acquiring-firm return, and a positive effect if domestic M&As. The positive crossborder M&A valuation is associated with benefits from the access to international capital markets and corporate governance transferences (Francis et al., 2008; Martynova and Renneboog, 2008; Feito-Ruiz and Menéndez-Requejo, 2011, among others). On the contrary, the negative cross-border M&A valuation is related to more agency conflicts and asymmetric information problems in the acquisition of foreign target firms (Moeller and Schlingemann, 2005). The empirical evidence shows that there is a relationship between the cross-border M&A frequency and the differences between the acquiring and target firms legal and institutional environments (Bris and Cabolis, 2008; Martynova and Renneboog, 2008). A weaker legal and institutional environment in the target firm might be a source of gains for acquiring firms, given that this provides target firms with access to cheaper financial funds, through the internal capital market generated by the acquiring firm (Francis et al., 2008). It also allows a better corporate governance system to be imposed on the target firm, generating value for both firms and reducing managerial opportunism (Hagendorff et al., 2007; Martynova and Renneboog, 2008; Feito-Ruiz and Menéndez-Requejo, 2011). On the contrary, other studies show that a stronger legal and institutional environment in the target country may reduce the gains for acquiring firms, if they have to pay a higher price for the target firm in order to compensate it for having to adopt a worse corporate governance system (Stark and Wei, 2004; Kuipers et al., 2009). In contrast, acquiring firms may voluntarily adopt the better corporate governance regulation of the target country (Martynova and Renneboog, 2008).. - The takeover attitude (hostile takeover) raises the price offered in an M&A. The bargained up price induces a negative valuation of the bidder s stock (Goergen and Renneboog, 2004; Campa and Hernando, 2004). The same occurs in the case of multiple bidders because competition causes the winning bidder to succumb to the winner s curse and to pay a higher price for the target firm (Moeller et al., 2004). A tender offer is not unlike a hostile bid because the target board is bypassed when the offer is made, which we expect to have a negative impact on the bidder s announcement returns (Moeller et al., 2004). - A full acquisition limits the use of acquisitions to transfer wealth from the target s minority shareholders to themselves for example by using pyramidal control chains (La 8

10 Porta et al., 1999; Martynova and Renneboog, 2011). Therefore, a positive effect on the bidder shareholder is expected. 3. Data sources and sample description To test the above hypotheses, we collect information on M&As announced by European listed firms and involving both listed and unlisted target firms from around the world. Our sample period starts in 2002 when the M&A market was slowly recovering (subsequent to the equity crisis of March 2000) and ends at the end of 2007 when the first financial crisis stuck which slowed down the market for corporate control. We build our dataset from the Thomson One Banker Mergers & Acquisitions Database, DataStream. We retain transactions that: (1) are completed; (2) involve a change in the control; and (3) for which bidder share prices are available in Datastream. We gather executive compensation information from BoardEx (compiled by the firm Management Diagnostics Limited, which provides detailed information on executive compensation outside of the US for publicity listed firms and includes biographic information. 1 We find ownership data in the Amadeus Database from Bureau Van Dijk. We also check the M&A information through Lexis Nexis. We collect the cash and equity-based compensation data from all executive and non-executive board members from BoardEx. Cash compensation includes base salary and bonus, while equity-based compensation 2 includes the value of restricted stock, stock options, and other elements of long-term incentives plans (LTIPs) granted in a given year. Our final sample of M&A transactions consists of 216 cases involving firms from 26 countries (see Table 1). As expected, the largest European market for corporate control is that of the UK. The median CEO of our bidding companies earns a cash compensation (salary and bonus) of about USD 779,000 (Panel A of Table 2). Although UK top managers earn on average more than their continental counterparts, this is not the case for the sample because the continental bidders are on average larger than the bidding UK firms. The CEOs of UK bidding firms receive a yearly remuneration in cash of USD 783,000) whereas the continental CEOs earn USD 954,000 in cash; still the former receive larger bonuses. About half of the sample firms do not pay out equity compensation, but when they do, this type of remuneration is higher than the cash compensation. On the average value of the allocated stock optionsand LTIPs amounts to 1 For some Continental European firms, the compensation data are not (yet) available. The European Commission has recommended listed companies to report details on individual compensation packages since 2004, but this recommendation was only later translated into national regulations in a number of countries. Still, even when no national regulation exists for compensation to be disclosed at the individual level, listed companies have often opted for full compensation transparency. 2 BoardEx distinguishes between 1) the intrinsic value of options, which is calculated by multiplying the number of options awarded in the period by the difference between stock and exercise price, and 2) the estimated value of options awarded, which is a theoretical value that captures the potential value of the option during the vesting period by means of the Black Sholes model. In the remainder of the paper, we work with the estimated value. For the value of LTIPs, BoardEx displays the maximum value obtainable under the long term incentive plan. We call the top manager CEO even when the firm calls him/her Managing Director or Chairman Executive. 9

11 about USD 1.5 million; higher in Continental Europe (USD 2.1million) than in the UK (USD 1.2 million). A similar picture is exhibited in Panel B of Table 2 for the other executive directors. Table 1. Geographical distribution of M&A sample Sample includes 216 M&A deals announced by European listed bidders over the period 2002 to All Country Acquiring firm Target firm Australia 2 Austria 1 Belgium 1 1 Canada 8 China 2 Czech Republic 1 Denmark 4 Egypt 2 Finland 7 2 France 18 9 Germany 3 9 Hong Kong, China 1 Ireland-Rep 9 6 Israel 1 Italy 2 4 Netherlands 9 6 Norway 5 4 Russian Fed. 1 Singapore 1 South Africa 1 Spain 4 7 Sweden 10 3 Switzerland 3 3 UK United States 31 Utd Arab Emirate 1 All

12 Table 2. CEO and Executive Compensation: Descriptive Statistics Summary statistics for the sample of M&A announcements by European listed firms over the period 2002 to The target firms comprise both listed and non-listed firms from around the world. The table shows the mean and median values), the standard deviation, and the maximum values. We winsorize the financial ratios at the bottom and top 1% level. Variable description is given in the appendix. The remuneration data are the most recent ones before the M&A announcement. We also consider subsamples based on bidders from Continental Europe and the UK. The difference in means is based on a t-test. ***,**, and * indicate statistical significance at the 1%, 5% and 10% levels, respectively. Data source: Own calculation based on BoardEx. Full Sample (N=216) Bidder Firms From Continental Europe (N=62) Bidder Firms From UK and Ireland (N=154) Variable Mean Median Std. dev. Max Mean Median Std. dev. Max Mean Median Std. dev. Max Mean Dif. (p-value) Panel A: CEO Compensation Salary (000 $) (0.000)*** Bonus (000 $) (0.046)** Cash LITPs (000 $) (0.857) All Cash Compensation (000 $) (0.104) Options (000 $) (0.006)*** Stocks (000 $) (0.541) LTIPs (000 $) (0.825) Equity-based Compensation (000 $) (0.099)* Pension (000 $ (0.279) Other (000 $) (0.845) Total Compensation (000 $) (0.023)** Equity Compensation (%) (0.108) Panel B: Executive Compensation (Average) Salary (000 $) (0.041)** Bonus (000 $) (0.628) Cash LITPs (000 $) (0.651) All Cash Compensation (000 $) (0.084) Options (000 $) (0.158) Stocks (000 $) (0.347) LTIPs (000 $) (0.000)*** Equity Compensation (000 $) (0.002)*** Pension (000 $) (0.015)** Others (000 $) (0.001)*** Total Compensation (000 $) (0.142) Equity Compensation (%) (0.001)*** 11

13 Table 3. Board and CEO Characteristics: Descriptive Statistics Summary statistics for the sample of M&A announcements by European listed firms over the period 2002 to The target firms comprise both listed and non-listed firms from around the world. The table shows the mean and median value, the standard deviation, and minimum and maximum values. We winsorize financial ratios at the bottom and top 1% levels. Board characteristics incorporates Board size (total number of executive and non-executive directors); Executives (%) is the proportion of executives on the board; Independent board (5) is the percentage of non-executive directors; CEO/chairman duality is 1 if the chairman and CEO is the same person and 0 otherwise. The Network variables consist of the average of the CEO s and executives board positions (in quoted firms); Busy CEO (Executives) takes (take) the value 1 if the CEO (executives) has (have) more than 1 board in other firms. CEO characteristics incorporate: CEO tenure is the number of years that a CEO has held his position as CEO; CEO male gives the percentage of male CEOs; CEO nomination/audit/remuneration committee equals 1 if the CEO is a member of nomination/audit/remuneration committee. We also consider subsamples based on bidders from Continental Europe and the UK. The difference in means is based on a t-test. Data source: Own calculations based on BoardEx. Full Sample Bidder Firms From Continental Europe Bidder Firms from UK and Ireland Std. Variable Obs. Mean Median Std. dev. Min Max Obs. Mean Median Std. dev. Min Max Obs. Mean Median Min dev. Board characteristics Board size (0.000)*** Executives (%) (0.000)*** Independent board (%) (0.274) CEO/Chairman duality (%) (0.000)*** Network CEO s external directorships (0.000)*** Busy CEO (%) (0.000)*** Executives external directorships (0.000)*** Busy Executives (%) (0.112) CEO Characteristics CEO tenure (years) (0.963) CEO age (0.024) CEO male (%) CEO founder (%) (0.039)** Same CEO one year after M&A (%) (0.876) CEO in nomination committee (%) (0.064)* CEO in audit committee (%) (0.001)*** CEO in remuneration committee (%) (0.007)*** ***, **, and * indicate statistical significance at the 1%, 5% and 10% levels, respectively. Max Mean Dif. (p-value) 12

14 Table 3 presents of an overview of the board and CEO characteristics for the Continental European and UK firms. The board size averages to 9 directors; 11.6 for Continental Europe (where, in some countries such as Germany, the non-executive directors operate in a supervisory board, separated from the management or executive board) and 8 for the UK. The difference in board composition between Continental Europe and the UK is striking: the percentage of executive directors is lower for Continental Europe than for the UK (21.4% versus 46.8%, respectively). CEO/Chairman duality occurs more frequently in Continental Europe (with 78.7%) than in the UK (5.9%). The difference is statistically significant. CEOs and other executive directors are busier in Continental Europe than the UK; as they hold more as non-executive directorships in Continental Europe (4 on average for CEOs) versus only 1.8 in the UK. The average number of position for executives is also higher in Continental Europe (with 3.5) than the UK (1.7). The proportion of CEOs who founded their company amounts to 4.8% for Continental Europe firms and to 14.9% for the UK firms. CEOs are in more nomination and remuneration committees in Continental Europe than in the UK. In Table 4, we present the ownership distribution of the M&As by type of largest shareholder in the bidder firm (Panel A). The first four columns represent the number of the deals in which the majority shareholder is a family firm or individual, financial firm, non-financial firm, venture capital or a foundation and held less than 10% of ownership, among 10-20%, among 20-60% and among %. Column (5) and (6) indicate the total number of the deals for each type of shareholder and the percentage that it represents over the total deals. Column (7) and (8) show the mean of ownership held by each type of majority shareholder and its standard deviation. In Panel A, we observe that financial firms hold blocks in 59.72% of the bidding firms; 18.5% has an individual or a family; 17.6% has a non-financial company; 3.2% has a venture capital; 0.90% has a foundation. The mean of the ownership hold by the largest shareholder in the bidder firm is 23.3% on average. The percentage of ownership held by a family or an individual is 25.2%,17.60% if it is a non financial firm, 17.42% if it is a financial firm, 3.20% if it is a venture capital and 0.90% if it is a foundation. In panel B we compare the ownership distribution of the M&As by type of largest shareholder for Continental Europe and UK bidder firms. We observe that there are significant differences between the mean of the percentage of ownership when the largest shareholder is a family or an individual. In Continental Europe the percentage hold by a family or an individual is 48.45%, while in the UK it is 18.41%. These results are in line with the fact that there are more concentrated ownership structures in firms from Continental Europe than UK. 13

15 Table 4. Bidder Ownership Structure: Descriptive Statistics. The table shows the number of firms in which the largest shareholder (panel A) or strategic shareholders (panel B) own(s) has/have a specific ownership stake. The table also presents the mean and standard deviation of the ownership held by the majority shareholder (panel A) and the strategic shareholders (panel B). We distinguish the following types of the largest shareholders (panel A): family or individual; institutional investor; financial firm; non-financial firm; venture capital company; and foundation. For panel B, we add to the above types of shareholders also employees, government and foreign investors. Data source: Own calculation based on Amadeus Bureau Van Dijk and Datastream. Panel A: Majority Shareholder Ownership Structure Type of Majority Shareholder Below 10% 10-20% 20-60% Percentage of Ownership % All % Mean (%) Stand. Desv. (%) Family/Individual Financial Firm Non-Financial Firm Venture Capitalist Foundation All Panel B: Majority Shareholder Ownership Structure Continental Europe vs UK Percentage of Ownership in Continental Europe Type of Majority Shareholder All % Below 10% 10-20% 20-60% % Mean (%) Stand. Desv. (%) Below 10% 10-20% Percentage of Ownership in UK 20-60% % All % Mean (%) Stand. Desv. (%) Mean Dif. (p-value) Family/Individual (0.0004)*** Financial Firm (0.358) Non-Financial Firm (0.525) Venture Capitalist (0.134) Foundation All

16 Table 5 presents the firm and M&A characteristics of the sample. The market to book ratio of acquiring firms is on average 1.6, the cash flow is 0.1, and the leverage 0.2. On average there is not price run-up.the majority of the deals are announced by firm with high levels of size (66.2%). The relative size of the target firm is 0.2. The majority of the target firms are unlisted (75% of the deals). Acquiring firms paid for the majority of M&As with cash (69%). The majority of the M&As are focus (70.4%), domestic (54.3%), friendly (99.5%) and full acquisitions (97.7%). Table 5. M&A Firm Characteristics: Descriptive Statistics. The table shows the mean and median value, the standard deviation, minimum and maximum values of the firm and transaction characteristics. Firm characteristics consist of Growth opportunities (book value of the total assets minus book value of equity plus market value of equity divided by book value of total assets), Cash flow (EBITDA over total assets of the bidder firm), Leverage (total debt divided by total assets at the end of the previous year (WS item /WS item 02999)), Runup (cumulative abnormal returns (CARs) of the bidder firm over the window (-60, -20) days preceding the takeover), Large Bidder (is 1 if the bidder is within the highest quartile of the market cap at the end of the semester prior to the transaction announcement), Relative deal size (the log of the transaction value divided by the market value of bidder before the transaction). Deal characteristics comprise the method of payment (stock payment, cash payment, mixed payment, which equal 1 when the payment consists for 100% of stock, 100% of cash or of a mix of stock and cash; Unlisted target equals 1 if target is not listed; Unrelated industry equals 1 when the bidder and target are in different industries (difference in the two first digits of the SIC codes); Cross-border equals 1 if the bidder and target firm country are not in the same country; Takeover attitude (friendly) equals 1 if the target reaction to the deal announcement is not hostile; Multiple bidders equals 1 when there are many bidders; Full acquisition equals 1 if bidder firm acquires the hundred percent of the target s shares; Tender offer takes the value of 1 in case of a tender offer. Data source: Own calculation based on Thomson One Banker M&As, and Datastream. Variable Observation Mean Median Std. dev. Min Max Firms characteristics Market-to-Book Cash flow Leverage Price runup (%) Large bidder (%) Relative size Deal characteristics Stock payment (%) Cash payment (%) Mixed payment (%) Unlisted target (%) Diversified takeover (%) Cross-border (%) Takeover attitude (friendly) (%) Multiple bidders (%) Full acquisition (%) Tender offer (%) Bidder shareholder valuation and equity-based compensation In the first part of our analysis, we estimate the acquiring shareholders returns around the M&A announcement which is followed by modeling of the determinants of the acquiring shareholder valuation. 15

17 4.1. Stock price reaction. To examine the stock price effect of the M&A announcement, we calculate the cumulative abnormal returns (CARs) around the announcement date, by means of the market model of which the parameters are estimated in the period (-200, -21) days before the announcement date using a local index. We use the parametric Dodd and Warner (1983) and non-parametric Corrado (1989) test to establish the significance of the CARs and the difference. The table 6, we partition the sample into transactions of bidding firms of which the CEO is remunerated by means of high and low levels of equity based compensation. The CARs for the acquiring shareholders is positive, but there are no differences between high and low equity compensation around the M&A announcement. However, the CARs for acquiring shareholders are higher for the subsample of firms that pay low equity based compensation. Whereas Datta et al. (2001) find that U.S. acquiring managers earning high equity-based compensation gets higher return around the M&A and in the postacquisition period, we conclude that in Europe acquiring shareholders may perceive that high equity-based compensation for managers is excessive and they behave opportunistically when they decide about their compensation. Table 6. Bidder Cumulative Average Abnormal Returns (CARs) We show the cumulative abnormal returns for 216 M&A transactions announced by European listed firms on listed and unlisted target firms (all around the world). To test the significance of the returns, we use the (parametric) Dodd and Warner T-test (1983) and the (non-parametric) Corrado test (1989). ***, **, * represent statistical significance at the 1%, 5% and 10% level. Event Window All (N=216) Test Dodd & Warner (Corrado Tests) High CEO EBC (N=108) Test Dodd & Warner (Corrado Test) Low CEO EBC (N=108) Test Dodd & Warner (Corrado Test) Difference Test (p-value) %*** 6.90 (1.11) 0.62%*** 3.61 (1.20) 0.81%*** 6.14 (0.45) (0.677) (-1,+1) 1.38%*** 6.82 (0.85) 1.16%*** 3.33 (0.52) 1.59%*** 6.31 (0.72) (0.517) (-2,+2) 1.62%*** 5.81 (1.04) 1.51%*** 3.11 (-0.48) 1.73%*** 5.11 (1.99) (0.788) (-8,+8) 0.67% 1.29 (0.95) -0.44% (0.80) 1.78%*** 3.15 (0.60) (0.053)* (-20,+20) 0.46% 0.23 (-0.31) -1.33% (-0.78) 2.25%* 1.97 (0.31) (0.046)** (-20,0) 0.31% 0.55 (0.58) -0.34% (0.28) 0.96% 1.21 (0.58) (0.298) (0,+5) 1.60%*** 5.51 (0.03) 1.50%** 2.94 (-0.89) 1.69%*** 4.85 (0.92) (0.827) (0,+8) 1.11%*** 3.20 (0.62) 0.62% 0.63 (0.09) 1.60%*** 3.90 (0.82) (0.295) (0,+20) 0.87% 1.27 (-0.77) -0.37% (-1.11) 2.10%** 2.88 (-0.04) (0.055)* 4.2. Determinants of bidder returns. To examine the effects of the equity-based compensation on the bidder shareholder returns we carry out a multivariate analysis using an OLS regression with robust standard errors. Besides, we consider transaction and firm characteristics and other corporate governance mechanism as controls. The specification of the model is as follows: CAR i, j EquityCompensation BidderOwnership i 3EquityCompensation * BidderOwnership i t 0 t 1 YearDummies i i, j (1) i 2 4BoardCharacteristics i 5Network i 6CEOCharacteristics i 7CorporateGovernance i 8FirmCharacteristics 9DealCharacteristics i IndustryDummies CountryDummies The window over which we define the dependent variable (CAR i ) for the bidder is (-2, +2). 16 k k m m

18 The explanatory variables include: -Equity-based compensation (EBC) which is the natural logarithm of 1 plus the total CEO equity compensation (stock option and restricted stock) divided by total CEO compensation earned out over the fiscal year prior to the announcement. We will also test the stock option compensation (natural logarithm of 1 plus stock option divided by total CEO compensation) and the LTIPs (restricted stock) (natural logarithm of 1 plus LTIPs divided by total CEO compensation) separately. Hypothesis 1 points out that we expect a positive effect on the bidder shareholder valuation. We will also interact equity-based compensation with bidder ownership in order to investigate whether the presence of the higher shareholder control may neutralize the effect of equity-based compensation on the bidder shareholder value around the M&A announcement. - Salary and Bonus are defined as the natural logarithm of 1 plus the total CEO salary/bonus divided by total CEO compensation earned over the fiscal year prior to the announcement. - Bidder ownership defined as the percentage of control right held by the ultimate owner. Following Martynova and Renneboog (2009) we focus on control rights rather than on ownership as we want to control for dual class shares, pyramidal ownership structures, multiple control chains, and cross-holdings. Strong control may represent tight monitoring of managerial decisions such that the management is less inclined to undertake takeover transactions that are not value-enhancing. Recent research suggests that not only the size of ownership but also the type of owner could be relevant. We also consider the identity of the ultimate owners partition them into these categories: (i) Family, for individuals or a families; (ii) Financial firm for bank, financial entities, mutual funds or pension funds; (iii) Non-financial firm, for private or public firms (not active in the financial industry); (iv) Venture capital, for venture capital firms; (v) Foundation, for trusts and foundations. While families and non-financial blockholders may actively monitor managerial decision making and are thus more concerned about the firm value creation in acquisitions (Feito-Ruiz and Menendez-Requejo, 2010), it is not certain whether all institutional investors try to influence the firm s strategic policies (Hartzell and Staks, 2003). While some institutions can be regarded as activists (Ferreira and Matos (2008) demonstrate that foreign institutional investors often engage more with the firm they invest in) others take a passive stance. We control for board characteristics, networks, corporate governance, and firm and deal characteristics (which are defined in the Appendix). We also include industry, country, and year fixed effects. 17