The value creation through Mergers and Acquisitions on bidder. banks share prices in China

Transcription

1 The value creation through Mergers and Acquisitions on bidder banks share prices in China Master thesis, Department of finance Ni Chen ANR: MSC FINANCE Supervisor: Mina Vlachaki Second reader: Fabio Castiglionesi /2016 Date of completion: November 28th, 2016

2 Abstract This thesis investigates whether the mergers and acquisitions can have a significant and positive effect on stock returns in Chinese bidder banks during the period from January 2006 and December The main purpose is to measure the change in shareholders value for bidders using several event windows, by performing cumulative average abnormal returns (CAARs) around the announcement date. According to the results of the event study, M&A announcements do not have a statistically significant influence on CAARs over the 41-day event window. In the second step, we examine the cross-border versus the domestic M&As. According to the results, cross-border events have a positive significant return during the whole event window, while M&As do not increase the shareholder s value on the domestic side. Moreover, this thesis takes into account diversification M&As activities and focus M&As activities. According to the results, focus activities lead to a significant positive impact on cumulative abnormal returns and add value to the shareholders. In order to account for the special characteristics of the Chinese market, a big difference between big State-owned banks and the other Joint-equity banks is found. Specifically, big SOC banks can create a statistically significant impact on bidders. The evidence suggests that an association between M&As announcement and stock performance of Chinese bidder banks exists in some events, especially during the post-announcement period.

3 Table of Contents 1 Introduction Literature review and hypothesis development Selected literature and evidence European market evidence US market evidence Emerging market evidence Hypothesis development M&As activities and stock returns Geographic diversifying M&As and stock returns Activities diversifying M&As and stock returns SOC Banks M&As and stock returns Data and Methodology Data selection and Sample design Research Methodology Results Analysis results of all bidders Analysis results of geographical diversification Analysis results of Activity focus and diversifying Analysis results of SOC banks and Joint-equity banks Conclusion on Empirical Reference Appendix

4 1 Introduction Mergers and acquisitions are first overserved in the United States in the 18 th century, as an important strategy to improve the firm size and enhance market competition. Recently, due to a number deregulation and technological innovations, emerging countries experienced an increasing number of mergers and acquisitions. M&As occur in waves, due to economic conditions at the firm level (Viswanathan, 2004) and the industry level (Harford, 2005). After the Asian financial crisis and the subsequent economic reforms, a high number of M&As transactions has been carried out in the banking industry of Asian market. Especially in the Chinese banking sector, which has the biggest number of M&As in the Asian market during this period (figure 1), Figure1: Numbers of deals Banking industry M&As in Asia market during Vietnam Thailand Sri Lanka South Korea Singapore Philippines Papua N Guinea Pakistan Malaysia Indonesia India China(incl HongKong & Taiwan) Source: Thomson Reuters Over the past twenty years, with the development of Chinese economic growth, the financial institution sector has experienced several big reforms. For instance, China joined the WTO (2001), after considering the potential benefits of such action, such as the reduced cost, the decreased risk and the creation of synergies. More foreign financial institutions are allowed to come into the Chinese market. By the end of 2012, - 1 -

5 China had one of the largest banking sectors in the world in terms of total assets. Due to changes in the Chinese government policy, the corporate restructuring is more and more popular. As an important corporate restructuring strategy, the volume of mergers and acquisitions have a rapidly increase (figure 2), Figure Deal value in China Bank M&A value Aggregate deal value(mil USD) The stock price changed and the stock market seems to have an immediately reflect on the M&A transactions for the shareholder value creation (Gasper et al., 2005). The value creation from merger and acquisition may differ over different regions, due to differences in the financial systems across countries (Barth et al., 1997). A large number of recent papers studies the effect of merger and acquisition in European and US banking sectors. However, very limited theoretical or empirical work has been done on how stock markets changed via deals announcement in the Chinese banking industries (Haleblian et al., 2009). In general, both the financial figures and the stock performance can reflect the influence of mergers and acquisitions activities. The main research method evaluating the M&As events is value effect. Traditional findings from early literature suggest that there have no statistically significant for bidders wealth value during M&As activities, either in the European market (Lepetit et al, 2004; Beitel et al, 2004; Karceski et al, 2005; Hernando,

6 and Hagendorff et al, 2007) or in the U.S. market (Dodd, 1980; Houston and Ryngaert, 1994; Anderson et al, 2004). Especially, M&As are reported to loss bidder values in the U.S. market, such as Piloff and Santomero,1998; Cornett et al,2003; Anderson et al,2004; Knapp et al,2005; Crouzille et al,2006. In particular, cross-border M&As activities are often find to have a negative impact on bidder shareholders values in developed markets (Gubbi et al, 2009; Amihud et al, 2002, Campa and Hernando,2004; Aw and Chatterjee,2004; Conn et al, In contrast, Goddard et al. (2012) and Wu et al. (2016) argue that cross-border M&As have a positive wealth effect in the emerging counties. Diversifying M&As activities are reported to add shareholders wealth (Cybo-Ottone and Mugria, 2000; Lepetit et al, 2004 and Ekkayokkaya et al, 2009). Otherwise, Delong (2001a, 2001b), Ismail and Davidson (2005) find higher CAARs in bank-to-bank than diversifying activities. This thesis studies the effects of M&As activities focus on Chinese banking industries and the impact of M&As announcement on the share price. It contributes to the extant literature in three ways. First, as far as we aware this study is the first time to research M&As conducted by Chinese banks. Our data include all M&As transactions in the banking sector of China A-shares stock exchange during the period of 2006 and Second, the cross-border and focus M&As activities can add shareholders value during announcement windows. Meanwhile, our outcome shows that the M&As announcement has value creation during the post-announcement period in stock market, which supports the idea that M&As announcement can create positive influence in the Asian emerging banking system. Moreover, due to the fact that many bank M&As transactions in emerging markets have been initiated by governments (Goddard et al, 2012), I use State-owned banks as a group and find that the bidder shareholders have a wealth value if they belong to SOC banks. The paper is structured as follows. Chapter 2 briefly dwells on the present previous empirical study in the different market and the hypothesis development. Chapter 3 focuses on the sample selection and present the data. In chapter 3 the event - 3 -

7 methodology used to measure the stock performance is explained. Chapter 4 shows the empirical results and the analysis. Finally, chapter 5 summarizes and concludes

8 2 Literature review and hypothesis development 2.1 Selected literature and evidence The empirical evidence on mergers and acquisitions effect on financial institution has been investigated for the European market and the US market. It is commonly known that mergers and acquisitions is a major measure to strengthen and consolidate the financial markets and improve banking value creation. However, whether M&As improve a bank s performance and how it affects the shareholders value is a debatable question in the current literature. Yet, current research shows varying results on the stock market reaction after the M&As announcements on different markets European market evidence Regarding the European market, some event studies documented that M&As activities have a positive impact on targets shareholders and no impact at all on bidders shareholders. Cybo-Ottone and Murgia (2000) examine the stock market valuation of the largest M&As in 14 European economies, using 54 large European banks during the period According to their findings, the stock abnormal returns for bidders around the announcement date are significantly positive, when using the general market index. However, neither the bidder nor the target have a positive and significant market reaction around the announcement date, when the bank sector index is used. Bietel and Schiereck (2001) study the merger activities of 98 European acquiring banks in over 24 countries across the whole world. Their results are partially consistent with those reported in Cybo-Ottone and Murgia (2000) since they document a significant positive return for the combined entity of 1.2% around the merger announcement, however the bidder banks return is insignificant. Lepetit et al. (2004) investigate the expected gains of M&As transactions in European banks between 1991 and According to their finds, there are positive and statistically significant abnormal returns for the target group in a 15- and 31-day event window around the announcement date. Additionally, the deals with geographic - 5 -

9 specialization and activity diversification are significantly positive at the 5%-level. However, the bidder group has insignificant abnormal returns in the same event windows. In order to investigate the excess returns to shareholders of the targets, the bidders and the combined entity of the bidder and the target. Beitel et al. (2004) use 98 large M&As of European bidder banks from 1985 to They find that the shareholders of targets have a significant positive cumulative abnormal returns. Regarding the combined entity of bidders and targets, they find significantly positive CARs for the whole samples. Meanwhile, there is no significant CARs to bidder shareholders in various event windows. Finally, targets can achieve more value in cross-border deals whereas bidders can get value in geographically focused transactions. Karceski et al. (2005) estimate the stock price reaction of banks around merger announcements. Their results vary, when taking into account the different bank size and strategic focus. Specifically, the target banks have positive abnormal returns and cumulative abnormal returns, while acquiring banks have insignificant CARs. Campa and Hernando (2006) analyze the magnitude of value creation with merger deals between 1998 and According to their results, targets experienced a positive excess return around the announcement. However, for acquirers, the average excess return are insignificant. Lensink and Maslennikova (2008) analyze the value gains to the acquirer in the European bank M&As wave of The sample includes 75 publicly traded banks from 19 European countries. Their found reveals that bidders CAARs value are significantly positive for [-5, 5], [-10, 10] and [-20, 20] while they are statistically insignificant for three-day event window. Additionally, bidders CAARs for crossborder and no diversifying M&As are statistically significant positive for [-10, 10] and [-20, 20]

10 2.1.2 US market evidence Turning to the US banking industry, the impact of M&A transactions is not clear and it varies across studies. Some early research shows the M&As have a negative impact on bidders and that shareholders suffered wealth losses, which means there are significantly negative abnormal returns (Piloff and Santomero, 1998). Dodd (1980) investigates a sample of 151 takeovers in both cancelled and completed mergers between 1970 and According to the results reported, there is a significant negative abnormal returns of -7.22% and -5.5% at the [-1, 0] event window. Houston and Ryngaert (1994) examine the stock market s perception of bank mergers in the period They demonstrate that overall the weighted average of gains of the bidders and targets are insignificant. They further reveal that for the five-day event window the cumulative average abnormal return for bidders is -2.32%, and statistically significant. Cornett et al. (2003) conduct an event study on average abnormal returns and cumulative abnormal returns around the announcement of acquisitions for a sample of 423 banks sample during the period When taking into account the whole sample of their research, the results point out significantly negative results. Specifically, the average cumulative abnormal return is -0.74% for a three-day window [-1, 1]. In addition, they also report that both the interstate and activity diversifying acquisitions experience significant negative CARs in this window. Their figures indicate that, around the announcement date, diversify leads to significantly lower abnormal returns compared to those from the focus activities. Anderson et al. (2004) use a sample of 97 bank mergers from Following the standard event study, they estimate the wealth effects for bidders in the seven-day window [-5, 1]. The result reveals insignificant average cumulative abnormal returns around the initial announcement date. Similar results are found in Knapp et al. (2005). Their findings reflect negative returns to shareholders in large bank deals between 1987 and 1998, and moreover, there is a decrease in post-merger profits, credit quality, and fee income

11 In a more recent strand of the literature, several papers find significant and positive cumulative abnormal returns for bidders following M&As announcements in U.S. market. Jarrell et al. (1991) use 663 successful tender offers covering the period 1962 to 1985, and according to their results, the stock return around the announcement dates are small but with a significant growth over time. Olson and Pagano (2005) indicate that acquiring banks shareholders can expect to have wealth gain in long-run. DeLong and DeYoung (2007) reveal positive abnormal returns to bank merger announcements, which disappear very quickly. More importantly, they document that there is a positive relationship between both stock market reactions (short-term) and financial performance (long-term) with the number of mergers that took place in the years prior to the event announcement. To compare the acquirer returns of bank announcement in the European and the US market, Hagendorff et al. (2007) study a sample of 204 banks mergers between 1996 and According to their results, bidders have a significantly negative CAARs for all samples during a three-day window, and the CAARs for the [-10, 1] and the [-2, 2] windows are -0.18% and -0.32%, respectively. The main finding reveals that bidder banks shareholders in Europe have not any statistically significant wealth losses compared to US shareholders. In other words, bidders banks experienced higher stock returns when targeting low protection economies (most European) than bidders targeting institutions which operate under a high investor protection market (most US) Emerging market evidence However, there have been very limited empirical evidence on mergers and acquisitions studies in Asian emerging market. Crouzille et al. (2006) investigate banks M&As in eight Southeast Asian countries by using an event study to examine the relationship between bank M&As and shareholder value for emerging markets during the (Asia financial crisis period), and their final results reported significantly negative abnormal returns for both acquirers and targets during the crisis period. To assess the - 8 -

12 different stock reactions between M&As in the non-financial industries and banking industries. Ma et al. (2012) study on completed bank mergers with announced between 1998 and They find that without controlling the firm size there have a significantly positive CAARs both in non-financial and bank industries in short-term event window. For example, the CAARs in the three-day window is 0.57% for the banking sector and 1.43% for the non-financial sector. Meanwhile, the diversifying bidders are not gain significantly increase in shareholders wealth. Also, large firms have not obvious market reactions than small-medium firms. Du and Sim (2015) investigate whether the effect of M&As on bank efficiency differs for both targets and bidders in six emerging countries. By using DEA model, they documented that the effect of M&As is insignificant if combined targets and bidders. They found that there has a significantly positive effect of merger and acquisition on bank efficiency for targets banks, but no efficiency improves for bidder banks. Based on event study, Suresh et al. (2006) conduct a study to see whether there is wealth impact on public announcements of Indian companies between 2004 and The study conclude that the stock returns are negative surrounding the announcement, though not significantly different from zero on the day. Also, the post-announcement abnormal returns are lower than the pre-announcement period, which reflects erosion in the firm value. Anand and Singh (2008) analyze the bank mergers announcement impact of short-term shareholder wealth in Indian bank over the period 1999 to The result reveal significantly increase in shareholders value for both bidder and targets banks, also the CAAR of the pre-announcement period also significantly positive. Wong et al. (2009) study the effect of M&As announcements on shareholders value of the Asian firms by using the Bloomberg Database and Reuters Business Database during the period of 2000 and Their outcomes indicated that the announcement is considered positive relations between for bidder shareholders but not for targets. Moreover, they found that there has a positive abnormal returns during the post-announcement period. Ma et al. (2012) discuss abnormal returns on bidder shareholders around announcement for 10 emerging Asian markets during

13 They reported positive cumulative abnormal returns in 3 different event windows: [0, 1], [-1, 1], and [-2, 2] window. At the same time, the information leakages have a benefit impact on shareholders value. Sehgal et al. (2012) analyze the abnormal returns used completed M&A data in BRICKS countries from 2005 to Their results reflected that there is a significantly positive in pre-announcement returns for sample countries. Their results also reflected the leakage of information in the system which has an impact on financial regulators. 2.2 Hypothesis development Since 2001, and in order to comply with the world trade organization (WTO) commitments, China implemented economic reforms, which improved significantly its macroeconomic indicators and the financial sector. Although the number of mergers and acquisitions have a steady increase due to the subsequent opening up of the Chinese banking sector, there is no clear evidence about the relationship between M&As announcement and stock returns in China M&As activities and stock returns According to the literature discussed in the previous sub-section, when considering the short term, M&As activities create zero or negative average abnormal returns for bidder banks and positive abnormal returns for target banks on the announcement. As stated by Piloff and Santomero (1997), there is no relationship between M&As deals announcement and wealth creation. Houston et al. (2001) analyzed 64 large banks acquisitions between 1985 and 1996, and they conclude that the average target achieves a significant value creation, while gains for acquirers are negative. Similar results are found in Crouzille et al. (2006), who advocate that a significant negative market reaction to the M&A deals occurred during in emerging markets. By comparing and analyzing performance before and after the M&As announcement of listing enterprises in China. Chi et.al. (2008) used 1,148 M&As deals in Chinese listed firms and find positive abnormal returns for the acquiring firms

14 Due to pervious event studies, it is important to analyze bidders out of combined M&As entities during event studies. Based on the hubris hypothesis, the net gains to a merger is zero. As discussed previously, most empirical studies find that the acquiring bank s shareholders value is not significantly different than zero for the European and the US banking industry. Firstly, I would like to test if this is the case for China, as well. Hypothesis 1: M&A announcements are not expected to create an impact on stock returns of bidders at conventional significance levels. Hypothesis 1 is tested by examining if the bidders CAARs are not significantly significant during the announcement windows. Hypothesis 1 will be rejected if the stock returns experience significantly positive or negative CAARs Geographic diversifying M&As and stock returns By the fifth mergers and acquisitions wave sweeping across the world, the cross-border M&As are an effective way to boost international expansion. International M&As offer the opportunity to the firms of an emerging economy to obtain resources that may not be available in their domestic market, and thereby enhance their capabilities of becoming competitive in the post-reform period (Gubbi et. al. 2009). On average, recent studies documented a negative abnormal returns from geographical diversification in developed markets. Amihud et al. (2002) find that cross-border M&As activities achieve significantly negative CARs of acquiring banks. Campa and Hernando (2004), Aw and Chatterjee (2004) & Conn et al. (2005) report that cross-border bidders gain lower abnormal returns than domestic bidders. Lensink and Maslennikova (2008) analyze value gains to acquirers based on a sample of 75 banks from 19 European countries, they find that stock returns from diversifying cross-border deals are positive and insignificant. However, Goddard et al. (2012) find that in emerging markets geographical diversification creates shareholder value for acquirers. Wu et al. (2016) document that Chinese firms overseas mergers and acquisitions achieve a significantly positive wealth effect than domestic acquirers, therefore, cross-border deals can be

15 more successful in the Chinese stock market. Wang and Peng (2016) evaluate the financial accounting performance and their findings reveal that the financial performance improves in the first year and declines in the long term. Overall, the crossborder M&As do not significantly improve the operating performance of Chinese firms. The international diversification hypothesis state that cross-border M&As exploit the international market, and they provide the most valid measure to acquire strategic assets including marketing, human& natural resources and technologies. Meanwhile, firms can reduce the operating cost and shareholders can gain substantial returns on the announcement. Combing all, I assume that cross-border M&As will receive positive stock abnormal returns and higher than domestic bidders during the event window. Therefore, Hypothesis 2a: Cross-border M&As activities are expected to create a positive impact on stock returns of bidders at conventional significance levels. Hypothesis 2b: Domestic M&As activities are expected to create a positive impact on stock returns of bidders at conventional significance levels. Hypothesis 2a and 2b are tested by analyzing whether the bidders CAARs are significantly positive in different event windows. It will be rejected if the cumulative abnormal returns are not significantly positive or CAARs are significantly negative Activities diversifying M&As and stock returns Pervious researches on the European market and the US market always take geographic and activity diversification as an important factor to measure the stock performance. Besides conducting the cross-border M&As returns, it is necessary to analyze the activities diversifying returns. There have two different opinions in this question. Delong (2001a, 2001b) argues that there is a significantly negative relationship between market reaction and activity diversification around the M&As announcements. The

16 paper further argues that mergers with both activities enhance the stockholders value by 3%, whereas diversifying mergers do not create value. In contrast, Cybo-Ottone and Mugria (2000) advocate higher gains in product-diversifying mergers. Lepetit et al. (2004) state that market reaction towards cross-product diversification is positive and significant, especially when cross-product diversification and geographic specialization are combined together. Ismail and Davidson (2005) find higher abnormal returns in bank-to-bank deals compared to cross-activity deals. When they take into account evidence of abnormal returns in the diversifying and cross-border deals, they report that diversifying M&As do not increase values for any of the event windows. Considering the existing academic literatures, I hypothesize activities diversification as follows: Hypothesis 3a: Activities focus M&As transactions are expected to create value on stock returns at conventional significance levels. Hypothesis 3b: Activities diversifying M&As transactions are expected to create value on stock returns at conventional significance levels. Hypothesis 3a and 3b are examined by conducting whether the bidders CAARs are significantly positive. It will be rejected if the CAARs are not significantly or if they are negative. If those hypotheses are rejected, while the null hypothesis will valid that activities diversifying are not expected to create values to stock returns, also the crossborder and focus deals are not gain values for bidders SOC Banks M&As and stock returns Commercial banks in China are classified into five types of banks: large State-owned commercial banks, joint-equity commercial banks, urban commercial banks, rural commercial banks and rural banks. Based on table 1, we can observe that big five Stateowned commercial (SOC) banks and Joint-equity banks are the two major components of banks M&As

17 Table 1 BANK Percentage (%) SOC Joint-equity Urban Rural Rural commercial commercial commercial commercial 68% 28% 4% Compared to other financial markets, the Chinese market is different. Specifically, it has higher macro efficiency than micro efficiency. As a result, the government has more than 50% of the voting rights in 31.4% of the listed firms (Yuan, 1999). It is worth noticing that most of the Chinese firms which participate in M&As are listed firms, especially in the banking industry. Su et al. (2008) state that the principal principal problems increase due to the fact that the government may be opposite to value maximization for individual shareholders. Because of political motivations, the authors analyze the relationship between government ownership and stock returns and their findings document a negative impact on bidders CAARs in a 2-day window during cross-border M&As. Taking this into account, this thesis also analyzes the SOC bidder banks stock returns during M&As, I suggest the following hypothesis: Hypothesis 4: Large SOC banks M&As activities are expected to create a negative impact on stock returns of bidders at conventional significance levels. Hypothesis 4 is examined by conducting whether the bidders CAARs in event windows are significantly negative. It will be rejected if the CAARs are not significant or CAARs are positive. If I fail to reject the null hypothesis, then the conclusion will be that SOC banks M&As activities create values to stock returns

18 3 Data and Methodology 3.1 Data selection and Sample design The date on the characteristics of deals was collected by the bank s annual report or announcement during the period between 2006 and All the necessary stock databases are obtained from the China stock market and accounting research database (CSMAR), which include close price, adjusted the close price and market index. In this study, bidders are limited to be China listed bank and sample data include cross-border, domestic, activities diversifying and focus M&As in China. Meanwhile, only samples in accordance with several criteria can be chosen: first, cross-border M&As transaction should more than RMB 10 million Yuan. Second, the acquiring bank holds less than 50% of the target firm shares before the announcement. Third, M&As agreement should be solely by the parent bank and make public announcements. Forth, excluding the small target relative to the acquirer (< 1% of MV). Furthermore, samples which have two or more deals and the announcement will be delated to avoid the heterogeneity on CAARs. There are two different shares in China, one is called A-shares which shares of RMB currency and traded on the Shanghai and Shenzhen stock exchange, while the other is H-shares which traded on the Hong Kong stock exchange. In this article, we only use A-shares and take Shanghai stock exchange composite index (SHCOMP: IND) as the market benchmark, due to not all the Chinese domestic investors float their shares simultaneously on both of these two stock exchanges markets. Since the market cannot realize which deals will be completed at the time of the initial merger announcement, we analyze both the completed and uncompleted merger agreements. Finally, there are 38 transactions and 11 banks engaged in this study, and the basic information is shown in Table 2:

19 Table 2 BANK CODE Domestic M&As Cross-border M&As Activities focus M&As Activities Diversifying M&As SOC/ Joint-equity Joint-equity Joint-equity Joint-equity SOC SOC SOC Joint-equity Joint-equity Urban-com SOC SOC TOTAL Research Methodology In terms of methodology, two major methods are used in the literature. Papers who follow the first method compare pre- and post-m&as accounting figures, such as ROA, ROE, and cost of capital etc. On the other hand, a parallel strand of articles uses the event study to find if the stock prices change in a special market around the announcement date (Yener and David, 2004). Considering the difference between accounting performance and event studies, if stock markets are efficient, then information about the potential synergies are fully impounded into prices at the announcement date and M&As are clustered due to stock market valuations (Shleifer and Vishny, 2003). Our target is to understand whether M&As have any effect on shareholders value of Chinese bidder banks. Inspired by Ahmad and Ian (2005), in this article, I will adopt the event study as the methodology to measure whether mergers and acquisitions can create values on improving shareholders value during the event window. In order to evaluate the stock reaction to the mergers and acquisitions, the objective of this study is to find whether there have a positive CAARs to shareholders of bidders around the announcement. Due to capture,

20 the leakage of information and the reaction of the market may affect the abnormal return, I will take a longer window in this study. The parameters for the market model are estimated over the period -230 to -30 days before the announcement. Daily abnormal returns for each bank are calculated over a 41 day event window, and regarding the event windows, we will focus on 15 main windows: [-20,0], [-10,0], [-5,0], [-2,0], [- 1,0], [-20,20], [-10,10], [-5,5], [-2,2], [-1,1], [0,1], [0,2], [0,5], [0,10], [0,20]. The choice of those windows is based on the method of Cybo-Ottone and Murgia (2000) and the standard market model is based on CAPM. Following Brown & Warner (1985), an OLS regression is performed, in order to estimate the relationship between market index and stock prices for each M&As activity, using the following equation: Ri, t = αi + βirm,t + εit Where: Ri,t =the return of stock i at time t, Rm,t =return of market, at time t, α and β are ordinary least squares (OLS) parameters estimated through the market model regression coefficient and εit is a statistically error. All stock returns used in this paper are daily log-returns, which are calculated in the following way: Ri, t = In (Pi, t+1/pi, t) In order to assess the stock price reaction to announcements of M&A, the abnormal returns are calculated as the difference between the actual returns and the expected returns: ARi,t = Ri,t E(Ri,t) Where: ARi, t is the abnormal return of stock i at time t (in days). Ri,t is the realized return of stock i at time t (in days), E(Ri,t) is the expected return of stock i at time t (in days)

21 Further, the daily abnormal returns are averaged during the event windows. AARi,t = ( 1 ) n ( N i=1 ARi,t) Where: N is the number of banks in the sample and ARi is the sum of each event abnormal return in period t. In order to make the individual sample reflect the general situation, the cumulative average abnormal returns are calculated by summing the average abnormal returns: t2 CARi, [t1, t2 ]= ( t=t1 AARi,t) CAAR = ( 1 N ) N i=1 CAR i In the end, the final step is to conduct a t-test to examine whether cumulative abnormal returns are statistically significant from zero. The t-test for CAR (cumulative abnormal return) was tested using the Dodd and Warner (1983) method. t = n CAR/n i= σcar Where: CAR is cumulative abnormal return for period [t1, t2] σcar is the standard deviation of CAR based on the estimation window Comparing the t-test results using the critical values at different significant levels, we can conclude about the hypothesis in the following way: H0: µ(car) µ(car) =0 VS H1: 0 σ(car) σ(car) The significant level for 90%, 95% and 99% are P <0.1, P <0.05 and P<

22 respectively. If the P-value falls inside the specific interval, it means significantly different from zero and the null hypothesis is rejected. In this article, all the conclusion will based on significant at the 90% level and P <

23 4 Results 4.1 Analysis results of all bidders Table 3 demonstrates the outcomes of cumulative abnormal returns with various event windows for the whole Chinese bidder banks in the sample. These results are similar to the ones presented in the literature for the US market. In 14 event windows, the CAAR figures are not significantly different from zero. Based on the 14 event windows of different length, we can conclude that the influence of the M&As on the stock return is not significant at any conventional level. For example, the CAAR is -0.46% for the three-day event window [-1, 1], 0.76% for the 11-day window and 1.83% for the 42- day, but they are all insignificant. As an economic interpretation, these results reflect the fact that shareholders of bidders do not have significant gain or loss compared to the market return. This is partially similar to Bietel et al. (2004), who find no significant CAARs to the bidder shareholders in any of the announcement windows. This no value add to shareholders of bidder banks is further consistent with several papers which investigate the European market, such as Bietel and Schiereck (2001), Karceski et al. (2005), Campa and Hernando (2006) and Dimitris and Shuai (2015). Also, these no significant figures support the Hubris Hypothesis. Only in the longer pre-announcement window, the CAAR figure is statistically significant and positive at the 90%-level which imply that the stock return of bidders higher than the market index return. Thus, in my opinion, this significant positive window may reflect the information leakages. But the significant level is lower

24 Table 3: Cumulative average abnormal returns all bidders EVENT WINDOW CAAR t-statistics [-20, 0] [-10, 0] [-5, 0] [-2, 0] [-1, 0] [-1, 1] [-2, 2] [-5, 5] [-10, 10] [-20, 20] [0, 1] [0, 2] [0, 5] [0, 10] [0, 20] 2.40%* % % % % % % % % % % % % % % t-statistics are obtained by using robust standard errors *** p<0.01, ** p<0.05, * p<0.1 Figure 3 present the shape of bidder CAAR changes during the whole announcement period. The cumulative average abnormal returns are plotted between 20 days before announcement and 20 days after the event date. We can find that there is a slightly decline performance of CAAR in the pre-announcement period. Meanwhile, we can find the lowest level of the CAAR is occurred in the day before the announcement date. However, the return reflects an obvious volatility around the announcement date in the five-day period. To conclude, we can accept the hypothesis 1, according to which there is no value add to shareholders during the whole event period

25 Figure 3: Cumulative abnormal returns on all bidders (%) Bidders CAARs over time 3 2,5 2 1,5 1 0, , Analysis results of geographical diversification Table 4 presents the results of bidder shareholder cumulative abnormal returns in domestic and cross-border M&As activities. In the cross-border deals side, Panel A shows the number of days in the event window for which cross-border bidders have statistically significant CAARs. The trend of CAARs shows the rising returns during the event period. For the pre-announcement period, it peaks 20 days before the announcement date at almost 3.40%. Meanwhile, there is a continued trend in CAARs after the announcement. However, the t-statistic indicates that there are no significantly positive abnormal returns in the traditional three-day window similar results are obtained for the three-day window and the fiveday window, as well. Panel B shows that domestic M&A deals do not create a significant impact on shareholders value. More importantly, we find in 20 days after the announcement the CCAR is -4.83%, this result is consistent with European domestic studies, such as

26 Campa and Hernando (2006) report the abnormal returns are significantly negative, Cybo-Ottone and Murgia (2000) state no statistically significant values to bidders. Above all, the announcement of international activities has gain significant wealth for bidder shareholders that started from 20 days before the announcement and sustained 20 days after the announcement. The maximum wealth for shareholders is obtained during 41-day event window. In all sub-period windows, cross-border deals perform better than national deals. Table 4: Results of geographic diversifying Panel A: cross-border M&A EVENT WINDOW CAAR t-statistics [-20, 0] [-10, 0] [-5, 0] [-2, 0] [-1, 0] [-1, 1] [-2, 2] [-5, 5] [-10, 10] [-20, 20] [0, 1] [0, 2] [0, 5] [0, 10] [0, 20] 3.40%* % % % % % % %* %** %** % % %** %** %* 2.07 t-statistics are obtained by using robust standard errors *** p<0.01, ** p<0.05, * p<

27 Panel B: Domestic M&A EVENT WINDOW CAAR t-statistics [-20, 0] 1.25% 0.46 [-10, 0] 0.76% 0.51 [-5, 0] -1.52% [-2, 0] -1.37% [-1, 0] -0.86% [-1, 1] -1.45% [-2, 2] -1.60% [-5, 5] -1.65% [-10, 10] -0.43% [-20, 20] -2.91% [0,1] -1.25% [0,2] -0.89% [0,5] -0.67% [0,10] -1.85% [0,20] -4.84%* t-statistics are obtained by using robust standard errors *** p<0.01, ** p<0.05, * p<0.1 Figure 4 depicts the shape of CAARs in the entire 41 days event window for both crossborder and domestic bidders. It is obvious that the CAARs have a big difference between cross-border and domestic bidders in the post-announcement period. For the cross-border bidders, the blue curve drops before announcement. The CAAR is less than 1% just one day before the announcement, but still positive. After the announcement, it reverts to increase. In contrast, the graph of domestic side (green line) reflects a negative trend during event windows. There is a small gain before the announcement and followed by a fall. Above all, we can infer that geographic diversifying activities generated substantial positive returns for shareholders, which support hypothesis 2a and consistency to Goddard et al. (2012) & Wu et al. (2016) reported

28 Figure 4: Cumulative abnormal returns on geographic diversifying (%) Cross-border vs Domestic bidders Domestic CAAR Cross-border CAAR 4.3 Analysis results of Activity focus and diversifying The results in Table 5 report the outcomes for activities focus and diversifying M&A by various event windows for bidders. The CAAR for the 3-day and 5-day event window [-1, 1] and [-2, 2] are positive but not significantly different from zero. However, these results are statistically significant at a 5% significant level for 11-, 21- and 41-day event windows. Specifically, in the windows [-5, 5], [-10, 10] and [-20, 20], the CAARs are 3.77%, 6.48% and 9.24%, respectively. The positive figures indicate that bidder stock returns have a better performance than the market. M&A have a positive effect on the stock value if they take focus activities. This results is consistent with Cornett et al. (2003), who report that the product-focus has a significantly positive impact on the value creation of US-bank mergers and acquisitions, thus hypothesis 3a is accepted. Turning to Table 6, the results show that diversifying M&A activities are not having a statistically positive impact on abnormal returns, which is partially similar with DeLong (2001), according to which stock returns are not affected by the geographical or

29 product diversification deals, meanwhile the focused mergers significantly enhance values for combined entity(bidders and targets). However, Beitel and Schiereck (2001) document that geographically focused deals create more value for bidder shareholders than cross-border deals, product diversifying deals have a wealth created, and our results do not support this opinion. Due to the negative and insignificant CAARs in the activity diversifying results, we cannot accept the hypothesis 3b. Table 5: Results of activities focus bidders EVENT WINDOW CAAR t-statistics [-20, 0] [-10, 0] [-5, 0] [-2, 0] [-1, 0] [-1, 1] [-2, 2] [-5, 5] [-10, 10] [-20, 20] [0,1] [0,2] [0,5] [0,10] [0,20] 6.40%** %* % % % % % %** %** %** % % % %** % 1.78 t-statistics are obtained by using robust standard errors *** p<0.01, ** p<0.05, * p<

30 Table 6: Results of activity diversifying bidders EVENT WINDOW CAAR t-statistics [-20, 0] [-10, 0] [-5,0] [-2,0] [-1,0] [-1, 1] [-2, 2] [-5, 5] [-10, 10] [-20, 20] [0,1] [0,2] [0,5] [0,10] [0,20] 0.45% % % % % % % % % % % % % % % t-statistics are obtained by using robust standard errors *** p<0.01, ** p<0.05, * p<0.1 The CAARs of the activity focus and diversifying are present in figure 5. As the 41- day event window, the figure plotted the average abnormal returns before twenty days before the announcement and 20 days after it. On average, the abnormal returns of bank-to-bank M&As (green line) decreases fall to -1% before the announcement. However, its recovery one day before the announcement and have an increasing trend during the post-announcement period. There is a different trend happened in activities diversifying bidders, the blue line reflects the stock abnormal returns decline 10 days before the announcement and recovery 5 days before the announcement to 1% before another decrease in 5 days after the announcement date. Also, the decline stock returns are not significant. Thus we can conclude that, unlike the activity diversification hypothesis, our results do not support the positive relationship between activities diversifying and shareholders value created

31 Figure 5: Cumulative abnormal returns on activity diversifying (%) Focus on banks vs Diversification bidders Focus on banks bidders Diversification bidders 4.4 Analysis results of SOC banks and Joint-equity banks Table 7 shows the results for SOC banks M&As activities. There is a significant positive reaction, which is in support of the findings in Goddard et al. (2012), according to which Government-instigated M&As transactions appear to create more value for acquirer shareholders than privately instigated transactions. The average abnormal return of the State-owned commercial banks is positive for 11 event windows, only 4 windows are negative but insignificant. The maximum CAARs, 5.26%, appears in the [-20, 20] window, i.e., the longest event window. While it is not supported Crouzille, Lepetit, and Bautista (2008) state that market reacted negatively to the M&As in government led banks. For the Joint-equity commercial banks, we can see no statistically significant in any event window. We can conclude that there is a value creation for shareholders in State-owned commercial banks, which is not supported by hypothesis

32 Table 7: Results of State-owned commercial banks bidders EVENT WINDOW CAAR t-statistics [-20, 0] 2.94%* 2.02 [-10, 0] 1.20%* 1.96 [-5, 0] -0.01% [-2, 0] -0.18% [-1, 0] -0.20% [-1, 1] -0.03% [-2, 2] 0.70% 0.83 [-5, 5] 1.72%* 1.72 [-10, 10] 3.78%** 2.67 [-20, 20] 5.26%** 2.36 [0, 1] 0.01% 0.21 [0,2] 0.85% 0.95 [0,5] 1.70%* 1.99 [0,10] 2.54%** 2.48 [0,20] 2.28%* 1.78 t-statistics are obtained by using robust standard errors *** p<0.01, ** p<0.05, * p<0.1 Table 8: Results of the Joint-equity banks bidders EVENT WINDOW CAAR t-statistics [-20, 0] 2.36% 0.74 [-10, 0] 2.12% 0.89 [-5,0] 0.90% 0.36 [-2,0] 0.31% 0.20 [-1,0] 0.87% 0.85 [-1, 1] 0.85% 0.56 [-2, 2] -0.24% [-5, 5] 0.55% 0.17 [-10, 10] 0.90% 0.32 [-20, 20] -0.95% [0, 1] 0.42% 0.33 [0, 2] -0.10% [0, 5] 0.10% 0.06 [0, 10] -0.78% 0.42 [0, 20] -2.87% t-statistics are obtained by using robust standard errors *** p<0.01, ** p<0.05, * p<

33 The graphs in figure 6 plot the CAARs of the State-owned commercial banks and Jointequity banks. On average, the abnormal return decreases before the announcement date. After announcement date, they meet each other at 0.42 (t=1). However, after this point, this two line experienced by opposite trend. The State-owned banks (green line) enjoy an increase CAARs during the post-announcement period. Otherwise, the blue line shows stock returns decline before the announcement to approximately 0.3% and there is a slight recovery two days before announcement date, and continued to decline in the subsequent days. Figure 6: Cumulative abnormal returns on SOC and Joint-equity banks (%) SOCB vs Joint-euqity bidders SOCB Join-equity

34 4.5 Conclusion on Empirical Chapter 4 analyze the results of M&As announcement on bidder shareholders abnormal returns in various event windows and different M&As over the banking industry. In general, for all bidder banks, the outcomes reflect that the merger and acquisitions have no significant impact on shareholders value during the whole announcement period. However, if we take pre-announcement as an example, we can find that all the CAARs in a relatively longer event window are positive and most of them are statistically significant which means shareholder can achieve values before announcement date (t=0). For all bidders as an example, the CAAR is positive (2.70%) in the 20-day event window and significant at the 10%-level. I think this phenomenon may explain by leakages in information, such as media announcement before the public announcement by the firm. Also, it is very common in emerging market (Sehgal et al. 2012). This will influence short-term investors to make their investment decisions, due to investors may overestimate banks future efficiency and it will create synergies after M&As activities. Therefore, they enjoyed holding more stocks when they heard something before announcement date. However, by more information publish in stock markets such as the rules and conditions of merger proposals and financial status of firms engaged in M&A activities, some stock investors change their minds. As the figures show, the CAARs have a decrease trend before event date (t=0). Like other academic papers reported in emerging market, cross-border M&As have a value effect on bidder firms and activities focus also have a significant increase of bidder stock returns. At the same time, all the activities focus deals are the cross-border M&A, thus we can indicate that bidder shareholders experienced value-enhancing by the focus activities. Within China, due to the different characteristic of banks, the figure shows that the investors can have wealth if they are the State-owned bank s shareholders during M&As announcement. In other words, it reflects that the market does not reject the M&As executed by State-owned commercial banks, it may reveal that there has a relationship between political connections and bidders share price