A Study on the Relationship between Relative Bargaining Power and Market s Expected Synergy in Equity Exchange M&A Cases

Transcription

1 A Study on the Relationship between Relative Bargaining Power and Market s Expected Synergy in Equity Exchange M&A Cases Shean-bii Chiu * Tsung-kang Chen ** Hsien-hsing Liao *** ABSTRACT Methods to measure synergies of mergers and acquisition activities have been proposed in many literatures. However, few literatures discuss the influencing variables or conditions on the M&A activities synergy from the perspective of each participant s relative bargaining powers. Regarding synergy measurement, most studies focus on the indictors of operating performance for quite a long time after M&A activities. In this study, with an efficient market assumption, we develop an indictor Market s Expected Synergy to evaluate synergy of M&A activities within a short term after the occurrence of an M&A activity. It is constructed based upon the market reaction to the acquiring firm s stock price rather than traditional measurements mostly relying upon acquiring firm s financial performance after the M&A. Furthermore, we develop a Market s expected Net Synergy Model that can integrate the market expected synergy and merger premium in an analytical framework in terms of a plane with firm value and exchange ratio as horizontal axis and vertical axis respectively. This model can firstly make a direct linkage between expected synergy and merger premium. It not only provides a measurement tool to do time varying observation of market reaction after an M&A but also supplies an indicator for measuring the rate of return of an M&A from the perspective of the acquiring company. In addition, we introduce variables of relative bargaining powers in the explanation for market expected net synergy. We find that their explanations are quite significant in several time points during an M&A process (announcement date, completion date, and the month following completion). Therefore, the variables of relative bargaining powers provide a foundation for evaluating the potential net synergy generated by an equity exchange M&A activity. * Professor, Department of finance, National Taiwan University, Taiwan, chiushnb@mba.ntu.edu.tw ** Ph.D. Student, Department of finance, National Taiwan University, Taiwan, r97300@ntu.edu.tw *** Associate Professor, Department of finance, National Taiwan University, Taiwan, hliao@ntu.edu.tw

2 I. Introduction Methods to measure synergies of mergers and acquisition activities have been proposed in many literatures. However, few literatures discuss the influencing variables or conditions on the M&A activities synergy from the standpoint of each participant s relative bargaining powers. Regarding synergy measurement, most studies focus on the indictors of operating performance for quite a long time after the M&A activities so that it is very difficult to compare synergy with merger premium within a short term after the occurrence of an M&A activity. Nevertheless, merger premium represents the costs of the merger and synergy stands for incremental cash inflow after the merger. As a result, it is very important to establish the analytical relationship between synergy and merger premium and then can we measure the net synergy efficiently. In this paper, with an efficient market assumption, we develop a Market s expected Net Synergy Model (MNS model) that can integrate the market expected synergy 3 and merger premium in an analytical framework in terms of a plane with firm value and exchange ratio as horizontal axis and vertical axis respectively 4. Generally speaking, synergy can be divided into three parts: operation synergy, synergy from the changes of risk structure, and financial synergy (Lubatkin, 983; Singh & Montgomery, 987, Fowler & Schmidt, 989). However, these three kinds of synergy can be achieved by different M&A types including horizontal, vertical, conglomerate and diversification types. Additionally, Kitching (967) discovers that financial synergy occurs more often and operating synergy will be difficult to be appeared. In the past researches related to the proxies for synergy, most scholars used financial or market variables as the measures of synergy. Ansoff, Portner, Brandenburg, and Radosevich (97) bring up the twelve financial variables as the proxies for synergy, including sales revenue, total assets, PE ratio, EPS and so on. Kim and McConnel (977) use the creditor s return rate and debt ratio as the measures of creditor s CAR and leverage ratio. Beattie(980) measures the M&A s synergy as unsystematic risk and stock return (from the view of risk structure). Hoshino (98) selects seven variables as the proxies for performance indictors, including book value to total debt ratio, book value to total asset ratio, current ratio, debt to equity ratio, turnover ratio, net income to total debt ratio and net income to total assets ratio. Muller (985) mentions the market share as the judgment indictor to evaluate the changes of performance before and after M&A activities. 3 Normally, there are many measures for synergy. But in this paper, synergy will be defined as added value of combined firm at the time after M&A s announcement. This is because market s response for M&A s activity can be rapidly reflected on the added value and the purpose of this research is to explore the relation between the market s responses and merger premium. 4 In the past, Larson & Gonedes (969) mention an effective method (also called L-G model) to decide whether the negotiated exchange ratio is reasonable or not in equity exchange M&A activities. In the original L-G Model, they form their research framework on a plane of exchange ratio(er) and price-to-earning per share (PE). Although L-G model has been testified in several empirical tests in the U.S. and U.K. markets (Conn & Nieisen, 977; Cooke, Gregory and Pearson, 994), L-G Model has some theoretical assumptions which don t conform to the real world, especially the assumptions

3 This model can firstly make a direct linkage between expected synergy and merger premium while the reasonable exchange ratio computed by this model can reflect market s opinion by combined valued of two firms (V ). Besides, merger premium can be compared with synergy in several time points during an M&A process within the same framework and we can further discover the market expected net synergy 5. Within the framework of synergy and merger premium, we can observe both markets and acquiring firm s opinions 6 for M&A activities through the plane of MNS model. We define a new variable the Return Rate of Market s Expected Net Synergy (RMNS) to be a proxy for the relationship between expected synergy and premium as equation (): RMNS t e Synergyt premiumo = () premium 0 According to equation (), RMNS is not only the concept of investment s return rate but also a leading indicator to determine whether an M&A activity is reasonable or not. With the assumption of efficient market, we can expect that investors will have rapid and exact reactions on the merger case based on publicly available e information (it also refers to investors expected synergy, Synergy t ) 7. We can then realize market s opinions during the three observed time periods. On the other hand, some scholars paid attention to the relationship among merger that the sum of participants earnings is immediately equal to the combination of earnings during the assessment periods (combined earnings is fixed) and no consideration of synergy. Moreover, L-G model doesn t consider synergy so that the merger premium would not directly conform to market s opinion about the future. 5 In the past literatures, most scholars (e.g. Conn & Nielsen, 977; Firth, 979; Cooke, Gregory and Pearson, 994) paid attention to the following three time points: announcement date, completion date, and the month following completion date. 6 In MNS model, the market s opinion represents for the market expected synergy and the acquiring firm s opinion stands for the merger premium in the M&A activities. 7 Theoretically, in an efficient market, all the relevant information concerning the takeover should be impounded in the share price on completion of the takeover, since that is the point at which acquisition has become a certain rather than a probable event. However, Conn &Nielsen and several event studies (e.g. Firth, 979) investigate price behavior one month after merger and we follow this to allow our results to compared with those of other results.

4 premiums and its influencing factors, including the acquiring firm s financial variables 8, operating variables and other institutional variables 9. But in these researches, they primarily focus on the acquiring firm and its performance for quite a long time after the M&A activities; however, they seldom care about the target firm s relative status to acquiring firm. Hence, we introduce the variables of relative bargaining powers in the explanation for market expected net synergy and explore the relationships among them. The variables of relative bargaining powers indicate those variables that influence an M&A s equity exchange ratio. These variables usually relate to the relative capital size, relative earning generating ability, and relative stock prices between the acquiring company and the target company. Therefore, this study use three ratios between acquiring and target companies to proxy these variables, including relative capital ratio, relative earnings per share ratio and relative stock price ratio. Through the empirical results, we find that their explanations are quite significant in several time points during an M&A process (announcement date, completion date, and the month following completion). Therefore, the variables of relative bargaining powers provide a foundation for evaluating the potential net synergy generated by an equity exchange M&A activity. The rest of the paper is divided into four sections:first, we construct a Market s Expected Net Synergy Model, including a discussion on the model s assumptions, the model s inference and the model s meanings; Second, we present the data and 8 Shawver, T.J.(000) used both simple linear regression and multiple-linear regression to discover the relationships between the independent variables and the dependent variable which is represented by merger premiums for banks. There is a positive and significant relationship between the dependent variable: merger premium, and the independent variables: the deal value, acquirers assets, target return on assets, target return on equity, and accounting method for the merger as a pooling of interests. 9 On the part of the institutional relationship between the firm s performance and the probability of CEO turnover, a negative relation between the likelihood of non-routine CEO turnover and firm performance is documented in studies by Coughlan and Schmidt (985), Warner et al. (988), Weisbach (988), Gibbons and Murphy (990), Murphy and Zimmerman (993), Blackwell et al. (994), and Kang and Shivdasani (995). In addition, Zhao and Lehn (003) discover two important findings. One is a strong inverse relation between the returns to acquiring firms and the likelihood that their CEOs are subsequently fired. The other is that no significant relation exists between the probability that bad bidders get fired and various corporate governance characteristics, including the size and structure of boards. 3

5 methodology; Third, we empirically examine the effectiveness of our model and explore the relationships between RMNS and the variables of relative bargaining power. In the last section we conclude this study. II. Market Expected Net Synergy Model In this section we are going to introduce the construction of the Market s expected Net Synergy Model. We will first present the assumptions of this model and then provides a detailed derivation of it. Besides, we also make a discussion on the implication of each quadrants classified by the MNS model in M&A activities. II-. Model Assumptions:. This model only applies in the equity exchange M&A cases.. Since this research focus only in equity exchange M&A cases, we assume that the total debt value after combination is equal to the sum of participants debt values and it will keeps unchanged during our research period. It can be expressed as equation (). D t = D = D D () 3. The goal is to pursue maximization of wealth both for corporate management and stockholders. 4. The acquiring firm and the target firm are both traded in the exchanges or OTC. Figure represents a graphical determination of the exchange ratio in which subscripts,, and refer to acquiring, acquired and combined firms, respectively, and: ER: exchange ratio, number of acquiring firm s shares exchanged for each share of acquired firm s equity. AER: actual exchange ratio S : acquiring firm s pre-combination shares outstanding S : acquired firm s pre-combination shares outstanding E, P : the total earnings and stock price per share of acquiring firm 4

6 E, P : the total earnings and stock price per share of acquired firm P, PE : expected price per share and PE ratio of the combined entity V : the market value of the combined entity ER : the maximum exchange ratio which is acceptable to acquiring firm s stockholders. ER: the minimum exchange ratio which is acceptable to the acquired firm s stockholders. Furthermore, we can observe the Return Rate of Market s Expected Net Synergy (RMNS) through the relationship between market expected net synergy and participants expected synergy 0. First, let definev = Equity value + Debt value = ( PE E ) + D, E is a random variable with time-varying including the consideration of market s expected synergy so that equity value can reflect market s attitude for this M&A event. D stands for combine firm s total debt and is a constant according assumption. Second, make inferences in the following: When P = P (acquiring firm s wealth position is unchanged): P = PE ER V = EPS D V = P S D E P S S E + S ER P = S V + S D ER (3) When P P = PE ER = V = P (acquired firm s wealth position is unchanged): ER EPS P D V = S P D E S S E + S ER P ER = S V D + ER S (4) ER, V are random variables and form the new plane of MNS model. Therefore, when participants wealth position are unchanged: 0 Participants expected synergy refer to the merger premium willing to pay by acquiring firm or acquired firm.. 5

7 P V = P = D = D P = P ER = ER P + ( P + ( P S S + P ER S + P S ) = V ) = 0 D + ( P S P + P ( ) S P ) (5) According to the all above, we illustrate the new plane of the MNS model as figure. 0 V is combined firm s value on announcement (assume no synergy) V AER is participants anticipated value including synergy. 0 V AER V is participants expected synergy; that is merger premium V T is combined firm s value at the time after announcement(including market expected synergy ) ER Ⅱ ER AER P P Ⅲ Ⅳ Ⅰ ER V V AER V T Figure. Expected synergy of the market and the participants in MNS model The plane constructed by MNS model is similar to the one generated by L-G model (Larson & Gonedes, 969). However, different from L-G model, MNS incorporates the concept of synergy, relaxes the limitation of one-period wealth and further establishes an analytical framework that can integrate the market s expected synergy and merger premium. In addition, we create a new measure RMNS (the return rate of market expected net synergy) and then it can be taken as an indictor for assessing market s reaction toward an M&A. All these advantages are not found in the L-G model. A detailed introduction of L-G model pleases refer to footnote 4. 6

8 II-. Model Implications: In this study, the primary purpose of MNS model is to decide the RMNS and then we can measure the reaction degree of market on the M&A activities; The other objective is to find ER and ER that will make participants of mergers could maintain at least their original firm s values. Therefore, we can judge that participants wealth positions are expropriated or not at the related three time points in M&A process from figure. The actual quadrant status (I, II, III, IV) of the merger is determined by the relation of the AER to the exchange ratios implied by the ex post combined firm value. The following criteria define the realized quadrant of any given merger: If ER>AER>ER, then Quadrant I. It implies that the wealth of acquiring company and acquired company are both not expropriated. If AER>ER & ER, then QuadrantⅡ It implies that the wealth of acquiring company is expropriated but the acquired company is not. If ER>AER>ER, then Quadrant Ⅲ. It implies that the wealth of acquiring company and acquired company are both expropriated. If AER<ER & ER, then Quadrant Ⅳ. It implies that the wealth of acquired company is expropriated but the acquiring company is not. According to the above criteria, we can summarize that it is a good M&A deal for acquiring company when the negotiated exchange ratio (AER) locates in quadrant In MNS model, each participant s wealth position would maintain at least original level if AER falls behind Quadrant I. However, how to decide which Quadrant AER falls behind? In this study, we introduce the concept of methods mentioned by Larson and Gonedes (969) as follows: First, computing the combined firm s value (V ) at any observed time points. Second, let V into formulas of ER and ER and get the range for judgments. Besides, in order to avoid pre-announcement effects, the P and P variables were calculated using an average of monthly high and low share prices during the period three to five months prior to the month of merger announcement. 7

9 I or IV and it is a bad M&A deal for acquiring company when AER locates in quadrant II or III. III. Data and Methodology I. Data The data required for this study are described as follows: The sample of 5 U.S. equity exchange mergers for the period 997 to 00 is obtained from Bloomberg database. It includes merger type, exchange ratio, merger premium, announcement date, completion date, and some details about the mergers. The variables of relative bargaining powers include relative price ratio, relative capital ratio, and relative EPS ratio 3 for merger participants (acquiring firm and acquired firm). These related financial data are obtained from COMPUSTAT and CRSP. II. Methodology In this study, two major inquiries are investigated. We state the methodologies for each inquiry respectively as follows. Inquiry. Can MNS model effectively differentiate good M&A deals and bad M&A deals? To examine this inquiry, we will compute the cumulative abnormal returns (CAR) of each group (quadrant) for the following year from announcement or completion date. And under this inquiry, we will expect that good deals will have fairly positive cumulative abnormal returns and bad deals will have negative cumulative abnormal returns (no consider the deal size). That is to say, we expect that the groups both of quadrant I and quadrant IV should have positive cumulative abnormal returns and the groups both of quadrant II and quadrant III should have negative cumulative abnormal returns for acquiring company. 3 Bargaining power of each participant plays an important role in the negotiation process. However, it also primarily depends on each wealth status and each future growth of merger participants. As a result, we employ these three relative ratio as the proxies for each participant s bargaining power. 8

10 Inquiry. Can the variables of relative bargaining power explain the market s expected net synergy? To examine this inquiry, a regression analysis is used for the cases with positive merger premiums 4. P S EPS RMNSt = α + β ( ) + β ( ) + β 3 ( ) + β 4 AER + β 5 premium + β 6 Quadrant t + ε t P S EPS Where RMNS t is the return rate of market s expected net synergy at time t; (P /P ), (S /S ), and (EPS /EPS ) are all the proxies for relative participants bargaining power. And the other variables are taken as control variables, AER is the negotiated exchange ratio; premium is the merger premium of each case, and Quadrant t is the decided area of the MNS model. In order to proceed the above regression model, we have to calculate the return rate of market s expected net synergy 5 by the definition of equation (). In addition, we will make regressions on the three related time points of M&A process (announcement date, completion date, and the month following completion) and observe the regression results on the three related time points. Before proceeding the investigation of the above two inquiries, we firstly have to classify these M&A cases into the four quadrants defined by MNS model. With the information of the quadrant that each sample case belongs, we can differentiate good M&A deals and bad M&A deals. 4 In this study, we focus on the cases with positive premium. In negative premium cases, the acquiring firm will be beneficiary when gain the compensation from the acquired firm. It reveals that the acquired firm is bad in essence so that it is no meaning to exploring the relative wealth status; that is to say, acquired firm has no bargaining power on the negotiating process! 5 In this study, we introduce the concept of efficient market assumption and we can reasonably think that investors will have some expectations on what happened. Application on mergers announcement, investors will not only consider merger premium but also expected synergy from the mergers. 9

11 IV. Empirical Results Empirical results in this section are divided into three sub-sections. First, we present the classification results of the sample M&A cases by the MNS model. Based on these results, the empirical examination results of the MNS effectiveness are shown in the next sub-section. In the third part of the section, we demonstrate the empirical exploration results of the second inquiry, that is the explaining ability of the variables of relative bargaining powers for market s expected net synergy. IV- Classification Results of the sample M&A cases Table shows the result that how many cases are reasonable or not out of 5 equity exchange M&A cases in U.S. As table illustrates, 48 of the 5 proposed mergers produced sufficient combined value to maintain each participant s on quadrant I status at the time of merger announcement. Alternatively, the announcement effects produced wealth losses for one or both firms in 03(70+5+8) of the mergers. For the merger period examined (i.e., from announcement to the month following completion), at least 40% do not locate on the rationality quadrant of the MNS model. In addition, figure provides how many merger cases on quadrant I, II, III, and IV for the announcement and completion time periods. The incidence of wealth loss was much greater for acquiring firms than acquired. Quadrants II and III represent realized wealth declines for the acquiring firms. In the periods of merger announcement and completion, 95 and 04 acquiring firms respectively, out of 5, incurred decreases in stockholder wealth. On the other hand, stockholders in only 33 and 45 acquired firms in the two time periods failed to maintain an equivalent wealth position (quadrants III and IV). Similarly, rarely did the acquiring firm s stockholder gain while the acquired firm s stockholders lost. Quadrant IV contains only 8 and 5 merger cases for the two time periods respectively. 0

12 Table. Numbers of mergers by ex post quadrant and changes in quadrant status from month of announcement to month following merger completion. From Period of To Period of Month Following Announcement Announcement Completion Completion Ⅰ Ⅱ Ⅲ Ⅳ Ⅰ Ⅱ Ⅲ Ⅳ Ⅰ Ⅱ Ⅲ Ⅳ Ⅰ Ⅱ Ⅲ Ⅳ Ⅰ Completion Ⅱ Ⅲ Ⅳ Month Following Completion Ⅰ 38 Ⅱ 65 Ⅲ 4 Ⅳ 6 ER 70a, 64c Ⅱ ER 5a, 40c Ⅲ 48a, 4c Ⅰ 8a, 5c Ⅳ ER Figure. Number of mergers by quadrants of realized combined value in month of announcement(a) and completion(c) for 5 mergers sampled V

13 In table, stock market is apparently capable of reassessing the merger prospects during the period from the announcement to the month following completion. But we can t exactly know that investors are optimistic or pessimistic at announcement date. It is mainly because that information is time-varying and new information may significantly change investors expectations 6. However, from the date to a month following completion, we still find a trend skewing toward quadrant I. That is to say, most investors tend to think that the mergers are reasonable. IV- Effectiveness of the MNS model in differentiating good and bad M&A deals After categorizing merger cases into different quadrants by MNS model, we perform further tests for MNS model s effectiveness by using CAR 7 during the period from announcement or completion to the following year. It will show as in Figure 3, 4. CAR 0.3 Performance From Announcement Date Quadrant I Quadrant II Quadrant III Quadrant IV Time Figure 3. Acquiring company s performance for the period from the prior one month to the following one year 6 Due to the evidence of semi-strong efficient market, publicly available information will exactly and rapidly react on the changes of stock prices. Applying on the three time periods (announcement, completion, one month after completion), there will be different information set, that is to say, investors expected synergy will vary with time. 7 Benchmarks: All stocks traded in NYSE/AMEX/NASADQ exchanges

14 CAR 0.5 Performance After Completion Date Quadrant I Quadrant III Time Quadrant II Quadrant IV Figure 4. Acquiring company s performance for the period from completion to the following one year From figure 3, we discover that there are some evidences supportive of the MNS model. Acquiring firms in quadrants I or IV will perform well in the following one year from the announcement date; in contrast, acquiring firms in quadrants II or III will perform badly in the following one year. As a result, the hypothesis III will match up with the empirical results judged by the MNS model. We therefore can say that MNS model is useful for judging mergers are good deals or not. However, there is little different result in figure 4. As figure 4 illustrates, acquiring firms in quadrant I will keep constant CAR (about 5%) in the following one year from completion date, in quadrant II will keep decreasing trend, in quadrant IV will have large CAR, but in quadrant III unexpectedly perform from bad state to good state in the future. It is a confusing problem! Consequently, in order to explore what causes, we will divide three groups for quadrant III by deal size. The result will shows as figure 5. 3

15 CAR 0. Performance After Completion Date~Quadrant III Time below 0.5 billions 0.5~.0 billions above.0 billions Figure 5. Completion effect for Quadrant III From figure 5, we can see that small deal will have positive CAR in the following one year from the completion date. This is because that small deals influences slightly on the combined firm whether the merger case is successful or not. Hence, if we exclude small deals from sample, acquiring firms in quadrant III will perform expectedly bad in the future. Therefore, we can conclude that the MNS model is exactly effective to judge which merger case is reasonable or not. IV-3 Relationship between the return rate of market s expected net synergy and the relative bargaining powers variables of participants (regression analysis from 997 to 00on the positive premium cases) Table shows the results of testing whether the variables of relative bargaining power have significant impacts on the return rate of market s expected net synergy on announcement date. The tests include three parts, one is for the whole sample with positive premium, the other is the sub-sample with smaller deal size (which is below 4

16 0. billions), and another is the sub-sample with larger deal size (which is above.0 billions). In the same way, we perform similar tests on completion date and the following one month after completion. These results will show in table 3 and 4 respectively. In other words, we perform the regression model during the three important time periods. One is for the period of the announcement, the other is for the period of the completion, and another is for the period of the following month after the completion. The purpose of observing during the three periods is to explore what relative bargaining powers variables still significantly influence market s reaction in the future. From tables, 3 and 4, the results show that in all three periods, the relative price ratio and the relative capital ratio have positive and significantly influence on the return rate of market s expected net synergy in the whole sample. A positive relationship between the relative price ratio and the return rate of market s expected net synergy means that participants stock prices are considered by the market. It also implied that the higher the relative price ratio is, the higher the negotiated ability of acquiring firm is. In this situation, acquiring firm hardly encounter harm from unreasonable or higher premium. Therefore, market will give the positive reaction on the mergers. The similar explanation can be applied to the relative capital ratio. That is, the higher relative capital ratio means strong bargaining power of the acquiring firm, too. Moreover, these two significant variables also influence the return rate of market s expected net synergy in the same direction during the three periods. Further, we select two sub-samples for the smaller and the larger deal size out of the whole sample. On the part of the smaller deal size sub-sample, the relative price ratio and capital ratio are significantly positive-related to the return rate of market s expected net synergy; the relative EPS ratio and premium are significantly negative-related to the return rate of market s expected net synergy. It implies that the higher the acquired firm s EPS is in smaller deals, the higher the return rate of market s expected net synergy is. That is to say, the merger case is valuable when 5

17 acquired firm operated well even though it is a small deal. As for premium, the lower the compensation for the acquired firm is and then the higher return rate of market s expected net synergy is. On the other part of the larger deal size sub-sample, the relative capital ratio is significantly positive-related to the return rate of market s expected net synergy; the quadrant variable is significantly negative-related to the return rate of market s expected net synergy. It reveals that relative capital ratio plays a very important role in participants negotiating process when it will be a larger deal size. This is primarily because the relative capital ratio is a good proxy for relative negotiating ability. And when the merger case locates in quadrant I, the return rate of market s expected net synergy will be higher. Besides, there is something to notice that the relative price ratio is not significant in larger deal. It may be that the relative capital ratio represents the bargaining power better than the relative price ratio in larger deals. Moreover, according to the tables, 3 and 4, we discover that their explanations for market expected net synergy are quite significantly in several time points during an M&A process (announcement date, completion date, and the month following completion) 8 and the explanation power of regression model increase from announcement date to completion date 9. It revels that the uncertainty of the M&A activities will decrease by approaching the completion date and then the variables of relative bargaining power can explain more for the return rate of market expected net synergy 0. 8 It is meaningless to compare the return rate of market s expected synergy in different time points of M&A process. This is because information set is not fixed. Due to information randomly enters into the market, we can t directly compare the return of the announcement to the return of the completion. 9 For total cases with positive premium, the adjusted R-square changes from 0.64 to 0.54 during the period between announcement date and completion date. 0 It reflects on the decreasing of uncertainty, namely the error term (unexplainable parts). In addition, 6

18 Table. Multiple-linear regression of the return rate of market s expected net synergy against the variables of relative bargaining powers for the period of announcement. (Positive premium) P S EPS RMNSt = α + β ( ) + β ( ) + β 3 ( ) + β 4 AER + β 5 premium + β 6 Quadrant t + ε t P S EPS RMNS for the Announcement Period (N=4) Regression model R-square Adj R-square 0.64 ANOVA F-value 6.345*** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value.9679*** *** Sub-sample test deal size is below 0. billions (N=74) Regression model R-square Adj R-square ANOVA F-value *** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value 5.039*** 3.07*** *** * Sub-sample test deal size is above.0 billions (N=68) Regression model R-square Adj R-square ANOVA F-value.876** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value ** *** *** Significant under 99% confidence ** Significant under 95% confidence * Significant under 90% confidence 7

19 Table 3. Multiple-linear regression of the return rate of market s expected net synergy against the variables of relative bargaining powers for the period of completion. (Positive premium) RMNS for the Completion Period Regression model R-square Adj R-square 0.54 ANOVA F-value *** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value.740*** *** Sub-sample test deal size is below 0. billions (N=74) Regression model R-square Adj R-square ANOVA F-value *** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value 5.6*** 5.663*** -5.54*** ** Sub-sample test deal size is above.0 billions (N=68) Regression model R-square Adj R-square 0.6 ANOVA F-value 4.798*** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value *** ** *** Significant under 99% confidence ** Significant under 95% confidence * Significant under 90% confidence 8

20 Table 4. Multiple-linear regression of the return rate of market s expected net synergy against the variables of relative bargaining powers for the period of the following month after completion. (Positive premium) RMNS for the following month after Completion Period Regression model R-square Adj R-square ANOVA F-value *** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value 3.443*** *** Sub-sample test deal size is below 0. billions (N=74) Regression model R-square Adj R-square ANOVA F-value *** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value *** 4.59*** *** ** Sub-sample test deal size is above.0 billions (N=68) Regression model R-square Adj R-square ANOVA F-value 4.90*** Regression coefficients results Coefficients β β β 3 β 4 β 5 β 6 Estimates t-value *** ** *** Significant under 99% confidence ** Significant under 95% confidence * Significant under 90% confidence 9

21 In summary, our inquiries are supported by empirical results. In inquiry, the four quadrants decided by the MNS model exactly perform differently by the trend of the cumulative abnormal returns since the date of the announcement or completion. Good deals (quadrant I or IV) for acquiring company will have positive CAR and bad deals (quadrant II or III) for acquiring company will have negative CAR. In inquiry, some relative bargaining powers variables, which are the relative price ratio and the relative capital ratio, significantly influence the return rate of market s expected net synergy in the whole sample. Moreover, their explanations are quite significant in all three observed time points in M&A process and increase with approaching the completion date. In particular, the relationships between the return rate of market s expected net synergy and all the variables of relative bargaining power are the same in all three time points. Therefore, the variables of relative bargaining powers provide a foundation for evaluating the potential net synergy generated by an equity exchange M&A activity. V. Limitation In summary of the above, there are some theoretical limitations of the MNS model. First, since not every merger participant is traded in exchanges, the information of the market values of participants may be restrictive. Second, the equity exchange type is part of mergers so that the MNS is limited in this merger type. Third, the MNS model also neglects the possible change in the risk-return profile of the merger participants. If a merger results in reduction of unsystematic risk that is unobtainable for existing stockholders, their risk-return position may actually improve even if return declines. As a result, the criteria of at least maintaining price per share in the MNS model may substantially restrict the rationality quadrant. In practice, not every merger participant is traded in exchanges; in other words, not every merger participant s value can be described as stock value. Please refer to An Empirical Test of the Larson-Gonedes Exchange Ratio Determination Model, 977 0

22 VI. Conclusions The main contributions of this study are threefold. It introduces the concept of expected synergy, investigates the relationship between merger premium and expected synergy in MNS model, and empirically examines the influence of the relative bargaining powers related variables on the return rate of market s expected net synergy for merger cases. We develop the MNS model that much better match the real world (semi-strong form efficient market) without the constraint of the single period wealth of traditional model. In addition, the concept of efficient market also provides explanations for time-varying information set and changes of the return rate of market s expected net synergy during the three time periods (announcement, completion, one month after completion). With regard to the relationship between the relative bargaining powers and the market expected net synergy, empirical results shows that the former explain the latter quite significantly in several time points during an M&A process (announcement date, completion date, and the month following completion). Therefore, the variables of relative bargaining powers provide a foundation for evaluating the potential net synergy generated by an equity exchange M&A activity. Further study could be extended to include relative institutional factors ratios of participants to provide more explanations for the market expected net synergy.

23 Reference. Anonymous, 99, Stock Price Premiums: A Market Phenomenon, Mergers and Acquisitions Philadelphia Volume: 6 Issue: Ansoff, H. Igor., Brandenburg, Richard G., Portner, Fred E., and Radosevich, Raymond, 97, Acquisition Behavior of U.S. Manufacturing Firms, , Vanclerbilt University Press. 3. Beattie, D. L., 980, Conglomerate Diversification and Performance: A Survey and Time Series Analysis, Applied Economics, pp Blackwell, David W., James A. Brickley and Michael S. Weisbach, 994, Accounting Information and Internal Evaluation: Evidence from Texan Banks, Journal of Accounting and Economics 7, pp Brealey and Myers, 004, Principles of Corporate Finance, 7 th edition, McGraw-Hill /Irwin Company, pp Conn, R. L. and J. F. Nieisen, 977, An Empirical Test of the Larson-Gonedes Exchange Ratio Determination Model, The Journal of Finance, 3(3), pp Cooke Terry, Gregory Alan and Pearson Bernard, 994, A UK Empirical Test of the Larson-Gonedes Exchange Ratio Model, Accounting and Business Research, Vol. 4, No. 94, pp Coughlan, Anne T. and Ronald M. Schmidt, 985, Executive Compensation, Management Turnover, and Firm Performance: An Empirical Investigation, Journal of Accounting and Economics 7, pp Firth, M., 979, The Relationship between Stock Market Returns and Rates of Inflation, The Journal of Finance 34, pp Fowler L.K, and R.D. Schmidt, 988, Tender Offers, Acquisition, and Subsequent Performance in Manufacturing Firm, Academy of Management Journal, 3(4), pp.4~57.

24 . Gibbons, Robert S. and Kevin J. Murphy, 990, Relative Performance Evaluation for Chief Executive Officers, Industrial and Labor Relations Review 43, 30S-5S.. Hoshino Yasuo, 98, The Performance of Corporate Mergers in Japan, Journal of Business Finance and Accounting, 9, pp Kang, Jun-Koo and Anil Shivdasani, 995, Firm Performance, Corporate Governance, and Top Executive Turnover in Japan, Journal of Financial Economics 38, pp Kim, E. Han, and McConnell, John, J., 977, Corporate Mergers and the Co-Insurance of Corporate Debt. The Journal of Finance, pp Kitching, J, 967, Why Merger Miscarry?, Harvard Business Review, 45, pp.84~ Larson, K.D. and N.J. Gonedes, 969, Business Combinations: An Exchange Ratio Determination Model, The Accounting Review, pp Lubatkin Michael., 983, Mergers and the Performance of the Acquiring Firm., Academy of Management Review, 8, pp Mueller, Dennis C., 985, Mergers and Market Share., The Review of Economics and Statistics, 67, pp Murphy, Kevin J. and Jerold L. Zimmerman, 993, Financial Performance Surrounding CEO Turnover, Journal of Accounting and Economics 6, pp Penn, Thomas A. 98, Premiums: What Do They Really Measure? Mergers and Acquisitions Philadelphia Volume: 6 Issue: 3, Shawver, T.J., 000, Determinants of bank merger premiums. Bank, Accounting & Finance, 5(6), Shay, Scott A., 98, Setting the 'Right' Premium in an Efficient Market, Mergers and Acquisitions Philadelphia, Volume: 6 Issue:

25 3. Singh, Habir and Cynththia A. Montgomery, 987, Corporate Acquisition Strategies and Economic Performance, Strategic Management Journal, 8(4), pp.337~ Warner, Jerold B., Ross L. Watts, and Karen H. Wruck, 988, Stock Prices and Top Management Changes, Journal of Financial Economics 0, pp Weisbach, Michael S., 988, Outside Directors and CEO Turnover, Journal of Financial Economics 0, pp Zhao Mengxin and Lehn Kenneth, 003, CEO Turnover After Acquisitions: Do Bad Bidders Get Fired?, Working paper. 4