US MNCs' shareholder wealth effects of mergers and acquisitions in developing and developed countries, evidence from China and the UK

(1)

US MNCs' shareholder wealth effects of mergers and acquisitions in

developing and developed countries, evidence from China and the UK

Master Thesis Shihang Ge

University of Groningen: S2342898 Uppsala University: 891207-P480

Email: nicolege89@gmail.com Supervisor: Dr. W. (Wim) Westerman

Assessor: Dr. H. (Halit) Gonenc January 10th, 2014

MSc International Financial Management Faculty of Economics and Business

University of Groningen

MSc Business and Economics Faculty of Social Sciences

(2)

Abstract

This paper investigates the US MNCs' shareholder wealth effects of mergers and acquisitions in developing and developed countries. I choose the firms in the US as acquirers, the firms in the UK and China as targets. The sample consists of 107 acquisitions during the years 2009-2013. I find evidence that bidder’s cumulative announcement returns are higher when they acquire the firms in China. But I do not find evidence that M&As decrease bidders' shareholder wealth. I also illustrate several elements to explain why the differences exist between acquiring Chinese firms and British firms. Amongst these factors, the relative deal size, bidders' book to market ratio and

payment methods play an important role in explaining the difference between acquiring firms in

the UK and China.

Key Words: M&As, Cumulative abnormal returns, the UK, China

(3)

1. Introduction

An extensive empirical literature has explored the multinational companies’ (MNC's) shareholder wealth effects of mergers and acquisitions (M&As). Numerous empirical studies have reported the value creation for the acquiring firms' shareholders after the M&As, and a lot of studies have already investigated the abnormal return gained from MNCs' M&As in advancing economics. According to previous studies, shareholders will get excessive returns after M&As (e.g. Farrell and Shapiro, 1990; Corhay and Rad, 2000; Boateng et al., 2008). For example, Farrell and Shapiro (1990) believe that firms who choose to acquire mostly treat M&A as a profitable alternative investment. Successful takeover activities can improve efficiencies in finance, management and operations. On the contrary, a plenty of scholars also hold the opposite view, they claim that acquisition could destroy value (e.g. Berkovitch and Narayanan; 1993; Berger and Ofek, 1995). For instance, Berkovitch and Narayanan (1993) illustrate three major motives for M&A: synergy, agency, and hubris. They hold the idea that hubris and agency can destroy the shareholders’ wealth. However, among these documents, there are few articles focusing on the comparison of the short-term announcement effects of M&As in advanced and advancing economics. Pantzalis (2001) claims that MNCs that operate both in advancing economics and advanced economics will obtain a significant greater value than MNCs that only have businesses in advanced economics.

(4)

Economic growth in China has soared significantly in recent decades. A large amount of inward investment has rushed into this market. China is an attractive market for MNC investors because of the sufficient resources and unexplored market. Therefore, I choose Chinese firms which were merged or acquired by US companies as target firms. Furthermore, I study the manufacturing industry. The manufacturing industry has a broad range of sectors, varying from high-technology to mature industries such as automobiles, petroleum, chemicals. These industries witness an intensive pressure of cross-border M&As. In this paper, I also involve some variables to capture the differences of bidders' shareholders value creation between acquiring the UK firms and the Chinese firms. In previous paper, variables are classified into firm-specific, industry-specific and country-specific level to explain the cumulative abnormal return (e.g. Aybar and Ficici, 2009). Furthermore, this paper contributes to enrich the academic literature on the bidders’ shareholder wealth in cross-border acquisitions. It also provides evidence to support the internalization theory, the transaction cost theory and the positive multinational network hypothesis (e.g. Buckley, 1988; Teece, 1981; Williamson, 1975; Doukas and Travlos, 1988; Kogut, 1983). Additionally, this paper also gives some specific suggestions for managers who determine to go abroad in the process of taking M&As.

This paper is structured as follows. Section 2 briefly reviews the literature on M&As' value creations, and gradually develops two hypothesis. Section 3 and 4 describe the sample selection, the methodology and results. Section 5 provides the results discussion, mainly the important variables. At last, the conclusion, suggestions for managers and possibilities for future research will be covered in the section 6.

2. Literature review and hypothesis development

Nowadays, M&As and announcement effects are discussed by many scholars. Some scholars hold the idea that M&As could bring positive abnormal return. They propose the efficiency theory (e.g. Trautwein, 1990), the market power theory (e.g. Caves, 1981) to support the M&As' contributions to shareholder wealth effects. On the contrary, other scholars argue that M&As could destroy value. They apply theories, such as the hubris theory (e.g. Roll, 1986), and the

(5)

2.1.1 M&As create value for shareholder wealth - theories

Trautwein (1990) analyzes acquisition mode, entry mode and integration mode which are dominated by the efficiency theory. In his opinion, the main goal of acquisition is to achieve synergies. Financial synergies can be achieved by lowering the cost of equity. When firms use investment portfolio to decrease the risk, the information problem occur, insiders always have more information than outsiders. Barry and Brown (1985) state that information problem can make it difficult for managers to estimate future payoff. When they turn to information intermediaries such as ranking agencies or auditing companies, firm’s cost of equity will increase since firms need to pay additional compensation to investors. Firms with more information can reduce the cost. When firms merge with others, they become insiders of targets and financial synergies can be achieved. Operation synergies can be achieved by lowering the cost of transferring assets and knowledge resulted from M&As. Managerial synergies are realized by superior planning and monitoring. M&As can increase the efficiency of the company. However, the efficiency theory is criticized by some authors, for example, Rumelt (1982) finds no evidence that financial synergies can be achieved by M&As and Porter (1996) points out that managerial synergies and operation synergies do seldom realize.

The market power is an ability to influence price, market shares, quality and the nature of the product in the market. Market power can lead to high and risk-free profits by diversification. Caves (1981) develops the market power theory, and held the idea that M&As can help companies to build a market power by reciprocal buying. However, Eckbo (1992) finds no evidence of market power theory after analyzing M&As in both USA and Canada.

2.1.2 M&As destroy value for shareholder wealth - theories

(6)

Management entrench theory is derived from agency theory. Managers can make manager-specific investments by investing the firm’s resources at the cost of shareholder’s wealth, which minimize the risk of replacement. When they make more manage-specific investments, they will have higher wages and higher personal value. Then it is hard for shareholders to replace managers because of higher cost (Shleifer and Vishny, 1989). According to this theory, managers have incentives to make decisions of acquisition by themselves. To reduce the risk, managers will pursuit to acquire other companies to increase the personal value, though many acquisitions do not have value.

2.1.3 M&As create or destroy value for shareholder wealth-empirical studies

Some scholars hold the idea that M&As can create value for both bidders and targets. Ariff and Finn (1989) study the announcement effects of earnings, dividends and capitalization changes in Singapore equity market. They report that the announcements play a positive role in affecting the share prices. Kiymaz (2004) investigates 70 US targets and 207 US bidders in cross border acquisitions from 1989 to 1999. He finds that US targets gain positive and significant wealth, whereas US bidders gain positive but insignificant value. The work of Kumar and Panneerselvam (2009) demonstrates that both bidders and targets benefit on M&As' announcements in the short term. Moreover, bidders gain more than targets. Selvam et al. (2010) examine 17 acquiring firms from manufacturing industry in India during the period from 2000 to 2002. They sustain that both acquired and acquiring firms are winners, while acquirers gain less returns. The announcement effects are impacted by market volatility, liquidity, cash position and financial leverage of the firms.

(7)

But Bashir et al. (2011) show opposite results, they claim that the acquisition activities contribute to the bidders' shareholders instead of the targets' shareholder. Even more, the target firms experience a loss of value during takeover activities.

We can see that both views are supported by theories. But the theories stating that M&As create value for shareholders face quite some criticisms. Although various studies have been done in this field, the research results are contradictory. Based on the literature review above, we can see that most scholars hold the view that acquisition can bring abnormal return to target shareholders but not bidder shareholders. However, still some scholars believed that the bidders can earn benefits during the acquisitions. In order to test whether the shareholders' wealth effects are positive for the bidders, I provide the first hypothesis below:

Hypothesis 1: US Companies’ shares have no announcement effect after these companies undertake M&A activities in the UK and China.

2.2.1 Cross-border M&As and announcement effect - theories

Cross-board acquisition has been analyzed by many scholars. Most of them thought cross-border M&As would have higher abnormal return than domestic M&As. Some theories were proposed, such as the internalization theory (e.g. Buckley, 1988), the transaction cost theory (e.g. Teece, 1981; Williamson, 1975), the positive multinational network hypothesis (e.g. Doukas and Travlos, 1988; Kogut, 1983) and Pantzalis’s (2001) theory, which is similar to multinational network hypothesis but deals with the comparison of developing and developed countries. I will discuss these theories in detail.

The internalization advantage has been discussed by many scholars. However, was no specific model until the internalization theory turned up. Firms can take the advantage of market imperfections in host-countries to lower the cost. Buckley (1988) says that firms which go abroad are driven by three factors: least cost location for each activity, future growth opportunities and ownership advantage. Furthermore, ownership advantage can lead to lower agency problem and lower transaction cost. Francoeur (2006) investigates the long-run performance of domestic M&As and cross-border M&As for Canadian bidders, and finds evidence to support the internalization theory.

(8)

the first one to propose the transaction cost theory. Companies will face problems of high transaction cost because of opportunism, bounded rationality and small numbers condition. The transaction cost can be reduced by establishing a high hierarchic structure in the firm; managers would be monitored and were less likely to act as opportunism in a hierarchy structured firm. Buckley (1988) suggests that one of the advantages for firms to go abroad is the ownership advantage. According to his theory, a rational profit-maximizing MNC tends to use wholly owned subsidiaries in an imperfection market. As a result, cross-border M&As would decrease the transaction costs. Teece (1981) focuses on the international technology transfer cost. He thinks that transaction costs will be lower due to the arms-length contracting. Beamish and Banks (1987) test the transaction cost theory by analyzing joint ventures in different countries, and find the evidence that transaction costs would decrease after joint ventures.

The other related theory is from Doukas and Travlos (1988)'s positive multinational network

hypothesis which states that firms will not benefit from acquiring others without

country-diversification. They study the announcements effects on international acquisitions, and provide evidence for the positive multinational network hypothesis. Kogut (1983) proposes that the primary advantage for country-diversification is flexibility to transfer resources across borders through a globally maximizing network. It can be found that positive multinational network

hypothesis focusing on the advantages of flexibility of a multinational system, is different from

internalization theory or transaction cost theory which pay attentions to the market imperfection. Prather and Min (1998) analyze 240 international joint ventures in both developed and developing countries, and confirm that announcement effects of international joint ventures will be negatively related to the level of development of foreign partners’ country.

Similar to the positive multinational network hypothesis, Pantzalis (2001) focuses on the comparisons of developing and developed countries. He acknowledges the importance of the geographic matter of multinational corporations, and utilizes geographic scope, which is also known as the location factors, to capture the degree of segmentation or integration between the US and other countries where the international companies operates. His results indicate that MNCs operating both in advancing and advanced economics will obtain a significant greater value than MNCs that only have businesses in advanced economics.

(9)

they will be more efficient because of flexibility.

2.2.2 Cross-border M&As and announcement effect - empirical studies

With the rapid economic development of some Asian countries such as China, scholars start to pay close attentions to the Asian market. However, their findings cannot fully support the theories above. Wong and Cheung (2009) examine the market reaction to the M&As in Asian market from 2000 to 2007. They assert that M&As' transactions destroy target firms' shareholder wealth. More importantly, the Asian market reacts negatively to acquisition activities, it does not produce value for targets' shareholders. The impacts of M&As are further discussed by Liang (2009). He selects US companies listed in NYSE and Chinese companies listed on the Shanghai and Shenzhen stock exchange in order to analyze the acquiring firms' announcement effects of takeovers. He declares that the announcement effects are not significant for the US companies; however, they are significant for Chinese companies. Luedi (2008) believes that acquiring Chinese firms will face uncertainty over the source of capital and political interference. He does not find a negative effect on acquiring Chinese firms, but he gets a mixed result of abnormal returns.

For a US company, acquiring a firm in the UK is a form of country diversification. However, the degree of diversification is much lower than that when acquiring a Chinese firm. The UK and the US show more similarities. First, they are both developed countries; second, the UK and the US are liberal market economies (Hall and Soskice, 2001). It is shown that the UK and the US had few differences. According to positive multinational network hypothesis, the level of diversification between the UK and the US is low, thus acquisition cannot increase value much. With the foundation of theories above, I can provide the second hypothesis.

Hypothesis 2: Shareholders wealth effects of MNCs' M&As in China will be higher than the effects in the UK.

3. Sample selection

(10)

acquirers which takeover the Chinese or UK firms. Ordinary least square regressions and logistic regressions will also be run to test the effects of the variables other than country that influence M&As' returns.

The data used to analyze the changes of announcement effects after US companies' cross-border acquisitions in China and the UK, is derived from Zephyr, which contains very detailed information. The sample selections has five conditions:

1. The time period is covered from the beginning of January 2009 to the end of October 2013. 2. The acquirers should be the US listed companies in manufacture industry. The UK SIC 2007 industry classification is used to specify the manufacture industry.

3. The target should be the listed UK and Chinese companies. 4. The deal values are set to be at least 1 million Euro

5. The ownership of final stock for the acquirers should be more than 50% after the acquisition. A total number of 110 acquisitions meet the above requirements. Then I collect data on stock price from Nasdaq, and find 3 bidders became listed companies after the acquisition. In the end, Zephyr provides a sample of 107 acquisitions, including 75 acquisitions in the UK and 32 acquisitions in China. Some bidders acquire different Chinese or UK firms in the sample period. In the sample, 63 US bidders take acquisition actions to the UK firms, 28 US bidders take acquisition actions to Chinese firms.

4. Methodology and results

The majority of the documents to investigate shareholders' wealth effects were analyzed on the basis of “event studies”. Learning mainly from the work of Aybar and Ficici (2009), I implement various items of previous research to explain the key elements in this research. The shareholders wealth is mainly reflected in the stock price. The stock price can be evaluated by using the market model with the assumption that a linear relationship exists between the return of the security and the return of the market portfolio.

Where is the return of stock i at time t, and the represents the return of market at time t.

(11)

random error term of the stock i at time t.

First of all, I collected the data of share prices for each acquiring companies on Nasdaq. Then I use the prices from 255 days before announcement date to 10 days after announcement date for each company. Secondly, the S&P Dow Jones total market index is used to find the Rmt.

Equation (1) is estimated by using a 245-day estimation period from t= -10 to t= -255, where t=0 is the event day. For example, at the beginning, I calculate the daily return on market from 2008 (255 days before 2009) to 2013. Second, I calculate the daily stock return of firm j from the day -255 to the day -10; third, I use to match . Then I use ordinary least squares to calculate the and of company j.

The abnormal return (AR) and the cumulative abnormal return (CAR) are introduced to estimate the stock price changes around the announcement in a specific even window, say, 10 days before the announcement of acquisitions and 10 days afterwards (-10, +10). Abnormal return is the difference between the actual return of a security and the expected return. The formula is written up below.

ARit is the estimated abnormal return for security j over day r. and are calculated by the

formula (1), and are calculated of -10 days to +10 days around the announcement day. The average abnormal return of acquiring firms is obtained by aggregating all abnormal returns on each event day t and dividing by the number of firms with return data on day t.

ARt = (AR1t + AR2t + … +ARjt)/j (3)

(12)

Table 1: AR Category Statistics

Variable Observations Mean Std. Dev. positive negative

AR0 103 -0.0046 0.0232 50 53 AR1 76 0.0041 0.0424 37 39 AR2 56 -0.0007 0.0208 31 25 AR3 66 0.0036 0.0254 36 30 AR4 72 -0.0020 0.0198 35 37 AR5 58 0.0046 0.0228 33 25 AR6 77 0.0038 0.0192 41 36 AR7 102 0.0010 0.0281 53 49 AR8 77 -0.0063 0.0257 34 43 AR9 57 -0.0001 0.0255 30 27 AR10 75 0.0002 0.0202 36 39 AR-1 75 0.0019 0.0297 41 34 AR-2 58 -0.0008 0.0262 34 24 AR-3 75 -0.0002 0.0197 37 38 AR-4 74 0.0009 0.0297 39 35 AR-5 63 0.0004 0.0216 33 26 AR-6 78 -0.0026 0.0217 40 38 AR-7 102 0.0061 0.0429 56 46 AR-8 78 0.0042 0.0209 47 31 AR-9 56 -0.0001 0.0381 28 28 AR-10 76 0.0008 0.0213 39 37 All 1554 0.0006 0.0259 810 744

The cumulative abnormal return is the sum of all abnormal returns. The cumulative abnormal return (CAR) of the stock i is written below. Since the event window is from 10 days before ( ) to 10 days after ( ) the announcement.

(13)

Chinese firms after the M&As. It will help to testify the hypothesis 1 and 2. It is generally accepted that the error term is a random variable with mean equal to zero. To test whether or abnormal returns resulted because of an event, we need to test the hypothesis that the

cross-section mean of at the event day is different from 0.

The Panel A in table 2 shows the statistics of CARs for full sample. The period (-10,+10) shows a positive average of cumulative abnormal return, but other two event windows show negative average of cumulative abnormal returns. No events windows have significant results. However, the medians are all above zero. The results show that in a wider event window, the US companies have insignificant positive announcement effects after they take acquisition activities to companies in the UK and China; for a narrower windows, the US firms have negative announcement effects. I find weak evidence to support the first hypothesis – US companies’ shares have no announcement effect after these companies undertake acquisition activities in the UK and China.

The panel B in table 2 provides the summary statistics of CARs for firms which targets are Chinese firms. In three event windows, (-10, +10) and (-2, +1) have positive averages of cumulative abnormal returns. The longer period (-10, +10) shows bidders can get significant positive cumulative abnormal return, the mean of CAR is significant at 5% level. Furthermore, we can see that CARs are smaller in narrow windows than wider windows, especially CAR (-1, 0) shows a negative result. If we look back to the UK subsample, there is no such trend. Aybar and Ficici (2009) also got similar results, in their work, if firms take acquisition activities to Asian firms, the CARs are much smaller for narrow windows than wider ones; but targets from Latin America show opposite results.

The panel C in table 2 shows the summary of statistics of CARs for firms which targets are UK companies, we can see none of the time period show positive cumulative abnormal returns. The period (-1,0) shows the highest average cumulative abnormal return, which is -0.0026. The averages of CARs show insignificant results. But the medians are all above zero, which means the number of positive cumulative abnormal return is larger than negative ones.

(14)

Table2 Cumulative Abnormal Returns for windows Panel A: Surrounding the Event Day Full sample Time Interval

Mean median ST.P T-Statistic T-Value Positive: Negative Total

Sample

Positive Market Reaction

CAR-10,+10 0.0098 0.0136 0.112 0.901 0.36 62:45 107 57.94%

CAR-2,+1 -0.0005 0.0053 0.053 -0.112 0.91 59:48 107 55.14%

CAR-1,0 -0.003 0.0029 0.031 -0.993 0.32 59:48 107 55.14%

Panel B: Surrounding the Event Day Chinese Firms as targets Time Interval

Sample

CAR-10,+10 0.0448** 0.0381 0.112 2.216 0.03 21:11 32 65.63%

CAR-2,+1 0.0107 0.0073 0.063 0.934 0.35 19:13 32 59.38%

CAR-1,0 -0.004 0.0023 0.038 -0.599 0.55 17:15 32 53.12%

Panel C: Surrounding the Event Day UK Firms as targets Time Interval

Sample

CAR-10,+10 -0.005 0.0081 0.109 -0.398 0.69 41:34 75 54.70%

CAR-2,+1 -0.0054 0.0053 0.047 -0.974 0.33 40:35 75 53.30%

(15)

5. Variable analysis and discussions

5.1 Variables descriptions

Aybar and Ficici (2009) utilized firm-specific, industry-specific and country-specific variables to explain the differences in cumulative abnormal return. In the analysis, industry-specific variables include diversification. Firm-specific variables include firm size, relative deal size, bidder’s

financial leverage, bidders’ book to market ratio, payment method, other mergers and operational experience.

(1) Diversification

A focus strategy refers to when a firm is acquiring others in the same or similar industry, while diversification strategy refers to a firm acquiring others in other industries. Doukas and Travlos (1988) propose the positive-multinational-network hypothesis: country as well as industry diversified acquisitions were expected to obtain highest shareholders' wealth effects. They also pointed that in an integrated environment, a focus strategy would not perform a valuable function for investors.

However, other scholars found negative relationship between firms’ diversification strategy and the wealth of shareholders. If merging firms were drawn from similar industries, they would have operational synergies, which could increase value (Sudarsanam et al., 1996). Lang and Stulz (1993) find strong evidence that diversified firms perform worse than focus firms. According to their views, it is not wise to acquire firms in other industries. Ravenscraft and Scherer (1987) study 471 cases between 1950 and 1977 and state that diversified lines of business are associated with profit decrease. It seems that more studies show that diversification will destroy value (e.g Berger and Ofek, 1995; Denis et al., 2002). Firms that operate in an unfamiliar environment suffer from the increased risk and uncertainty, and decreased profits and portfolio assets, resulting in harms to the shareholders’ confidence and negative abnormal returns.

Both the risk of diversification and positive-multinational-network effects are found, and the conclusion of diversification effects is still conflicted. It is unpredicted in terms of the result of diversification or focus strategies.

(2) Bidders' firm size

(16)

method. Moeller et al. (2004) take 12,023 acquisitions made by public firms from 1980 to 2001 in US. He concluded that small firms received better returns compared to large firms. Unless acquiring public firms with share offers, large firms lost since they paid more premiums than small firms, leading to negative synergy returns. Managers in big companies typically have less ownership of the firm and they may be more prone to hubris and they overpay (Roll, 1986). Besides, Moeller et al. (2004) provide evidences that managers of large firms paid more for acquisitions, thus decreasing the abnormal return. Meanwhile, in line with the agency theory, Offenberg (2008) suggests that managers in large companies are more likely to pursuit their own interest rather than shareholders. He finds that CEOs of larger companies are more likely to merge companies without creating value. Also, Mitchell and Lehn (1990) show a negative relationship between firm sizes and returns by using a sample from 1982 to 1988, a period that witnessed the advent of hostile takeovers for large corporations. I predict that there is a negative relationship between acquires’ size and cumulative abnormal returns.

(3) Relative deal size

Relative deal size equals to the deal size divided by bidders' firm size. Moeller et al., (2004) examine a sample of 12,023 acquisitions by public firms from 1980 to 2001 and report a positive relationship between relative deal size and announcement returns in the sample of small bidders, while relative deal size is negatively related to announcement returns in the sample of big bidders. Meanwhile, the deal size can be the reflection of targets' size. In such cases, Morck et al., (1990) show that managers are more likely to overpay for large targets since they are more likely to get self-benefit from the large targets’ acquisitions, and large relative deal size could significantly influence the managers’ decisions and lead to agency problems, which decrease shareholders’ wealth. Additionally, Boston Consulting Group Research (2007) shows that deal value will destroy value when deal value exceeded 1 billion dollars.

(17)

(4) Bidders' financial leverage

Dhaliwal et al., (1991) claim that financial leverage could represent the risks of a firm, and risks would negatively influence the abnormal earnings, resulting in the negative relationship between abnormal earnings and financial leverage. Meanwhile, Harrison and Oler (2008) believe that a high level of leverage could lead to financial constraints for bidders, thus leading to higher risks. However, some scholars also argue that that as the bidder’s debt increases, bidders' management is more closely monitored by its creditors. As a consequence, firms have less cash flow to spend, making bad acquisitions less likely to happen (Song and Walkling, 2000). Masulis et al. (2007) also point out that higher financial leverage is positively related to a firm’s wealth. Managers under the pressure of high financial leverage had more incentives to improve firm performance.

Although the financial leverage is associated with higher risks, other factors such as opportunity and profit derived from the risk premium should also be taken into consideration. In this paper, I predict that financial leverage is positively related to abnormal return.

(5) Bidders’ book to market ratio

Learned from Barber and Lyon (1997), I also include the bidder’s ratio of book value to market value. In their work, both firm size and bidder’s book to market ratio presented significant results. Griffin and Lemmon (2002) find that high book-to-market ratio is consistent with high risk, and the higher the risk, the lower the earnings. Fama and French (1995) also show similar results by using different country samples from the years 1963 to 1992. They adopt the theory of rational pricing, arguing that a high book value to market value ratio signaled poor earnings, while a low book to market ratio indicated strong earnings. However, in some cases higher risks also increase earnings. Dichev (1998) finds that banks with higher book to market ratio earn higher returns because of risk premium. So, I predict that bidders’ book to market ratio can influence the abnormal return both positively and negatively. It depends on managers’ ability to overcome the risks.

(6) Payment method

(18)

signaling theory plays a key role in the differences between cash offers and stock offers. CEOs prefer cash offers if a firm is undervalued, while stock offers always indicated that bidders are overvalued. This finding is consistent with the theory of Myers and Majluf (1984) who insist that additional equity offer would decrease the current stock price because of signal effects. After examining 101 M&As in Netherlands between the years 1954 to 1997, Van Frederikslust et al., (2005) endorse that acquired firms benefited, but acquiring firms obtained indifferent positive results. Compared to stock payment method that created negative value, cash payment generated a positive wealth gains for bidders.

However, scholars also examined the downsides of the cash payment. For example, Wansley et al., (1983) focus on the tax effect of cash payments. By using a sample of 200, they find that taxes could reduce the abnormal return for bidders. Stock payment does not have such disadvantage. Chang (1998) studies privately held companies, and reports that acquiring firms enjoyed a positive abnormal return in stock offers, but zero abnormal returns in cash offers.

Most of the studies show a positive association between cash payments and abnormal return, and a negative association between stock offers and abnormal returns. So I predict that acquisition with cash payments can increase the abnormal return, while takeover by stock will decrease the wealth of shareholders.

(7) Other mergers

Other mergers is a dummy variable, if a firm has other acquisition activities during the period of 255 days before announcement day to 10 days before announcement day, dummy=1; 0 otherwise. This item can affect a firm’s beta, since beta is calculated by using the stock price from 255 days before announcement day to 10 days before announcement day. Takeovers during this period can significantly impact the stock price and thus affecting the beta. According to the formula - , we can see beta is negatively related to the abnormal return. M&As can temporarily increase the beta of firm because of higher risk (Elgers and Clark, 1980). So I predict that the other acquisition during the period of 255 days before announcement day to 10 days before announcement day will decrease the abnormal return.

(8) Operational experience

(19)

UK or Chinese for the first time, dummy=0; otherwise=1. Doukas and Travlos (1988) find that firms with operating experience would have negative abnormal returns, but firms would have positive abnormal returns if they operate in a target country for the first time. The markets would be more integrated if firms already have operating experience in target countries. It is consistent with the positive multinational network hypothesis (Doukas and Travlos, 1988), that firms could not benefit from M&As in an integrated world. So I predict that operating experience can decrease the abnormal return.

The panel A in table 3 shows the sources and descriptions of the variables. The panel B shows the statistic descriptions of the variables under three different samples. We can see that the bidders who take acquisitions in China have higher book to market ratio and financial leverage. While the bidders who take acquisitions in the UK have larger firm size and relative deal size. Moreover, bidders prefer to take the diversification strategy, and they usually have more operational experiences in the UK than they do in China. In addition, bidders who take M&As in China, are more likely to take the cash payments.

Table 3 panel A: Variables Description

Variables Description Expected

sign

Source

CAR Cumulative abnormal return Nasdaq

DCAR Dichotomous variable Nasdaq

Country Dummy=1 if target firm is a UK firm,0 otherwise - Zephyr Diversification Dummy variable, if acquirers use a focus strategy, the dummy=0; if firms acquire a

company in different industry, the dummy=1. I define this dummy variable by using US 2 digit industry SIC code.

+/- Zephyr

Firm size Represent bidder firms' size; the natural log of acquirer firm total asset. - Compustat item 4 Relative deal

size

Deal value divided by firm size. +/- Zephyr, Compustat item 4 Financial

leverage

Bidders’ financial leverage; the acquirers’ ratio of total asset to total equity. + Compusutat item 4, Compustat item 60 Bidders' book

to market ratio

Acquirers’ book price divided by market price at the end of M&A year. +/- Orbis

Payment method

Dummy variable, dummy=1 if only cash is used to settle transaction; dummy=0 otherwise.

+ Zephyr Other merge Dummy variable, if acquirer has other acquisition activities in the period from 255

days to 10 days from this announce date, dummy=1; dummy=0 otherwise.

- Zephyr Operational

experience

Dummy variable, if a firm takes acquisition activities to firms that in UK or Chinese for the first time, dummy=0; otherwise=1.

(20)

Panel B: Statistic description of variables

This table reports the mean, median, maximum, minimum, stand deviation, skewness and kurtosis values of variables for the full sample, the Chinese subsample and the UK subsample. Besides, the numbers of observations in each sample are shown.

Statistic description of variables in the Full sample

Mean Median Maximum Minimum Std. Dev. Skewness Kurtosis Observations

Diversification 0.560 1 1 0 0.498 -0.244 1.059 107

Firm size 21.43 21.562 25.973 17.242 2.066 0.099 2.714 107 Relative deal size 0.069 0.030 0.541 0.0001 0.101 2.541 10.131 107

Leverage 2.424 1.901 5.87 -2.54 3.136 4.077 14.589 107

Bidders book to market ratio 0.689 0.563 4.85 -0.281 0.599 3.707 13.984 107

Payment method 0.392 0 1 0 0.490 0.440 1.193 107

Other merger 0.224 0 1 0 0.419 1.321 2.747 107

Operational experience 0.429 0 1 0 0.497 0.283 1.080 107 Statistic description of variables in the Chinese subsample

Diversification 0.406 0 1 0 0.498 0.3817 1.145 32

Leverage 2.889 1.776 5.87 -2.54 5.581 2.230 11.248 32

Bidders book to market ratio 0.803 0.540 4.85 -0.281 0.946 2.808 11.884 32

Payment method 0.437 0 1 0 0.504 0.251 1.063 32

Other merger 0.218 0 1 0 0.42 1.360 2.851 32

Operational experience 0.25 0 1 0 0.439 1.154 2.333 32 Statistic description of variables in the UK subsample

Diversification 0.626 1 1 0 0.486 -0.523 1.274 75

Leverage 2.225 1.961 4.553 1.053 0.952 2.002 8.120 75

Bidders book to market ratio 0.640 0.569 1.796 0.154 0.363 1.012 3.467 75

Payment method 0.373 0 1 0 0.486 0.523 1.274 75

Other merger 0.226 0 1 0 0.421 1.305 2.704 75

(21)

5.2 Variable analysis

An empirical study is carried out to analyze the determinants of the announcement effect in cross-border acquisitions by American firms. I conduct several tests to figure out if the influences of the factors are statistically significant on the CAR because of the acquisition.

P-value in brackets p<0.01***, p<0.05** and p<0.1*

This table presentsunivariate results obtained by regressing the different variables one by one in full samples, Chinese subsample and UK subsample. Three event windows are used: (-10, 10), (-2, 1), (-1, 0).

5.2.1 Univariate analysis

The univariate analysis is used to test the effects of each variable alone. The table 4 shows the results of the univariate analysis. Nine variables are tested separately by the three event windows, and six of them have significant coefficient. The negative coefficient of firms’ diversification strategy is only significant at 10 % in the event window (-1, 0) in the Chinese subsample. The results of relative deal size show significantly negative signs in the UK subsample. The UK

Table 4: Univariate analysis

Full sample Chinese subsample UK subsample

Car(-10,+10)

Car(-2,+1) Car(-1,0) Car(-10,+10) Car(-2,+1) Car(-1,0) Car(-10,+10) Car(-2,+1) Car(-1,0) Diversification -0.0128 (0.5632) -0.0042 (0.6848) -0.0039 (0.5322) 0.0143 (0.7311) -0.028 (0.2237) -0.023* (0.0951) -0.0115 (0.6644) 0.0107 (0.3525) 0.0039 (0.5783) Firm size -0.0069 (0.1959) -0.0004 (0.8465) 0.0011 (0.4567) -0.02 (0.1773) 0 (0.9967) 0.0016 (0.5823) 0.0003 (0.956) 0.0003 (0.8943) 0.0007 (0.6761) Relative deal size 0.1584

(22)

subsample shows a persistent significant negative association between bidders’ book to market

ratio and CARs. The positive coefficients of payment method are shown in the full sample and

the UK samples, and they are significant at 5% level in event window (-1, 0). The negative sign of operational experience is only significant at 10% in the event window (-1, 0) of Chinese subsample. Additionally, country is negatively related with bidders’ abnormal return at 5% significant level, which means that acquiring UK firms destroys the value of bidders. All the coefficients of firm size, leverage and other mergers are not significant in the univariate analysis, which means these variables do not affect CARs.

Table 5 provides an overview of the correlation of variables used in the empirical analysis. We can see that the correlations between cumulative abnormal return are high. It is common in other scholars’ papers as well. Correlations between independent variables do not show the problem of multicollinearity, except firm size and relative deal size, which exceed 0.5 in all subsamples. I include either firm size or relative deal size separately in the cross-sectional regression analysis.

Table 5: Panel A Correlation-Full Sample CAR-10,10 CAR-1,0 CAR-2,1

(23)

To explain the cross-sectional variation in the cumulative abnormal returns, I implement the following multivariate model by using an ordinary least square regression:

+ + +

To check the robustness of the multivariate regression results, I also employed a binary logistic regression analysis. In the logistic regression model, I define the dependent variable DCAR as a dichotomous variable, designated to a value of 1 if it is positive and 0 otherwise. Then I use the same set of independent variables as in the models above.

5.2.3 Discussion of cross-section analysis

(24)

Table 6: Cross-sectional regressions: cumulative abnormal returns

Cross-sectional regressions: cumulative abnormal returns - Full sample

Panel A: Ordinary Least Squares Panel B: Binary Logistic regression

Dependent variable Car(-10,+10)1 Car(-10,+10)2 Car(-2, +1)1 Car(-2, +1)2 Car(-1,0)1 Car(-1,0)2 DCar(-10,+10)1 DCar(-10,+10)2 DCar(-2, +1)1 DCar(-2, +1)2 DCar(-1,0)1 DCar(-1,0)2 C 0.220 * (0.100) 0.014 (0.697) -0.028 (0.659) 0.024 (0.167) -0.054 (0.154) -0.006 (0.567) 1.179 (0.629) 0.357 (0.588) 0.409 (0.867) 1.360** (0.053) -0.969 (0.690) 0.107 (0.872) Country -0.041 (0.111) -0.048*** (0.060) -0.014 (0.265) -0.011 (0.347) 0.004 (0.537) 0.006 (0.380) -0.412 (0.382) -0.433 (0.357) -0.222 (0.634) -0.132 (0.784) 0.185 (0.690) 0.249 (0.594) Diversification 0.004 (0.862) 0.000 (0.996) -0.002 (0.879) -0.001 (0.917) -0.005 (0.452) -0.004 (0.521) 0.102 (0.816) 0.077 (0.858) 0.028 (0.949) -0.065 (0.884) -0.207 (0.633) -0.221 (0.610) Firm size -0.009 (0.153) 0.002 (0.488) 0.002 (0.249) -0.040 (0.738) 0.010 (0.932) 0.038 (0.754)

Relative deal size 0.194 *

(0.096) -0.081 (0.1510 -0.055 * (0.095) 0.272 (0.898) -6.787 ** (0.013) -3.035 (0.173) Leverage 0.001 (0.876) 0.000 (0.988) 0.000 (0.915) 0.000 (0.868) 0.000 (0.758) 0.000 (0.659) -0.001 (0.987) -0.004 (0.947) 0.082 (0.349) 0.085 (0.361) 0.108 (0.246) 0.111 (0.233)

Book to market value -0.008

(25)

Cross-sectional regressions: cumulative abnormal returns - Full sample (Continued)

Dependent variable Car(-10,+10)1 Car(-10,+10)2 Car(-2, +1)1 Car(-2, +1)2 Car(-1,0)1 Car(-1,0)2 DCar(-10,+10)1 DCar(-10,+10)2 DCar(-2, +1)1 DCar(-2, +1)2 DCar(-1,0)1 DCar(-1,0)2 Payment method 0.029 (0.210) 0.029 (0.206) 0.009 (0.428) 0.008 (0.452) 0.016** (0.017) 0.016** (0.017) 0.764* (0.075) 0.757 * (0.077) 0.337 (0.422) 0.276 (0.528) 0.457 (0.274) 0.428 (0.310) Other mergers 0.048 * (0.096) 0.041 (0.145) -0.011 (0.424) -0.010 (0.460) -0.003 (0.696) -0.002 (0.836) -0.345 (0.508) -0.385 (0.447) -0.280 (0.588) -0.373 (0.463) -0.363 (0.485) -0.369 (0.462) Operational experience -0.005 (0.852) -0.004 (0.875) -0.015 (0.230) -0.018 (0.154) -0.008 (0.292) -0.009 (0.232) 0.015 (0.975) -0.019 (0.968) -0.305 (0.526) -0.750 (0.131) 0.060 (0.901) -0.099 (0.837) R-squared 0.089 0.096 0.051 0.066 0.085 0.098 Adjusted R-squared 0.015 0.022 -0.027 -0.010 0.010 0.025 McFadden R-squared 0.037 0.036 0.037 0.092 0.030 0.043 P-value in brackets p<0.01***, p<0.05** and p<0.1*

(26)

Cross-sectional regressions: cumulative abnormal returns-Chinese firms as target

Dependent variable Car(-10,+10)1 Car(-10,+10)2 Car(-2, +1)1 Car(-2, +1)2 Car(-1,0)1 Car(-1,0)2 DCar(-10,+10)1 DCar(-10,+10)2 DCar(-2, +1)1 DCari(-2, +1)2 DCar(-1,0)1 DCar(-1,0)2 C 0.657 *** (0.005) -0.071 (0.206) -0.122 (0.401) 0.007 (0.845) -0.165 ** (0.032) -0.005 (0.776 ) 5.244 (0.294) -0.201 (0.845) -3.521 (0.497) 0.851 (0.424) 2.156 (0.655) 0.327 (0.762) Diversification 0.005 (0.905) 0.019 (0.679) -0.018 (0.516) -0.018 (0.514) -0.010 (0.502) -0.012 (0.447) -0.190 (0.839) -0.047 (0.959) -0.467 (0.595) -0.569 (0.525) -0.690 (0.433) -0.695 (0.438) Firm size -0.033 *** (0.004) 0.007 (0.309) 0.008 ** (0.038) -0.253 (0.293) 0.187 (0.462) -0.131 (0.583)

Relative deal size 0.540 *

(0.073) 0.153 (0.387) -0.030 (0.755) 2.482 (0.648) -5.207 (0.369) -7.351 (0.261) Leverage 0.002 (0.676) 0.001 (0.785) 0.000 (0.907) 0.001 (0.819) 0.000 (0.726) 0.001 (0.659) -0.009 (0.897) -0.010 (0.890) 0.094 (0.368) 0.095 (0.371) 0.097 (0.316) 0.093 (0.357)

Book to market ratio 0.008

(27)

Cross-sectional regressions: cumulative abnormal returns-Chinese firms as target (continued)

Dependent variable Car(-10,+10)1 Car(-10,+10)2 Car(-2, +1)1 Car(-2, +1)2 Car(-1,0)1 Car(-1,0)2 DCar(-10,+10)1 DCar(-10,+10)2 DCar(-2, +1)1 DCari(-2, +1)2 DCar(-1,0)1 DCar(-1,0)2 Operational experience 0.064 (0.378) 0.025 (0.750) -0.039 (0.421) -0.025 (0.599) -0.044 * (0.079) -0.033 (0.211) -0.503 (0.759) -0.732 (0.646) -2.113 (0.184) -1.934 (0.203) 0.166 (0.913) -0.184 (0.903) R-squared 0.376 0.227 0.139 0.128 0.373 0.251 Adjusted R-squared 0.194 0.001 -0.113 -0.126 0.190 0.032 McFadden R-squared 0.162 0.138 0.152 0.159 0.153 0.181 P-value in brackets p<0.01***, p<0.05** and p<0.1*

(28)

Cross-sectional regressions: cumulative abnormal returns-UK firms as target

Dependent variable Car(-10,+10)1 Car(-10,+10)2 Car(-2, +1)1 Car(-2, +1)2 Car(-1,0)1 Car(-1,0)2 DCari(-10,+10)1 DCari(-10,+10)2 DCari(-2, +1)1 DCari(-2, +1)2 DCari(-1,0)1 DCari(-1,0)2 C -0.012 (0.945) 0.039 (0.454) -0.004 (0.952) 0.049 ** (0.022) 0.002 (0.970) 0.015 (0.253) 0.199 (0.949) 0.079 (0.934) 2.614 (0.432) 2.777 ** (0.014) -0.805 (0.797) 0.454 (0.641) Diversification -0.018 (0.526) -0.013 (0.635) 0.005 (0.694) 0.004 (0.736) 0.000 (0.955) -0.001 (0.929) 0.186 (0.725) 0.184 (0.716) 0.391 (0.479) 0.196 (0.723) -0.023 (0.965) -0.014 (0.979) Firm size 0.004 (0.684) 0.002 (0.673) 0.000 (0.948) -0.005 (0.975) -0.062 (0.712) 0.048 (0.765)

Relative deal size 0.097

(0.451) -0.155*** (0.004) -0.069 ** (0.037) 0.169 (0.943) -8.184 ** (0.020) -2.262 (0.357) Leverage -0.003 (0.826) 0.000 (0.983) -0.006 (0.367) -0.006 (0.306) -0.001 (0.702) -0.002 (0.600) 0.088 (0.748) 0.085 (0.736) -0.048 (0.864) -0.129 (0.665) 0.256 (0.379) 0.278 (0.299)

Book to market ratio -0.069 *

(29)

Cross-sectional regressions: cumulative abnormal returns-UK firms as target (Continued)

Dependent variable Car(-10,+10)1 Car(-10,+10)2 Car(-2, +1)1 Car(-2, +1)2 Car(-1,0)1 Car(-1,0)2 DCari(-10,+10)1 DCari(-10,+10)2 DCari(-2, +1)1 DCari(-2, +1)2 DCari(-1,0)1 DCari(-1,0)2 Operational experience -0.036 (0.225) -0.024 (0.406) -0.012 (0.338) -0.022 * (0.064) 0.000 (0.958) -0.005 (0.458) -0.014 (0.980) -0.007 (0.989) -0.157 (0.774) -0.803 (0.166) 0.057 (0.916) -0.049 (0.927) R-squared 0.080 0.085 0.115 0.218 0.112 0.169 Adjusted R-squared -0.016 -0.010 0.022 0.137 0.020 0.082 McFadden R-squared 0.161 0.155 0.071 0.149 0.029 0.037 P-value in brackets p<0.01***, p<0.05** and p<0.1*

(30)

Country

We can see that the country dummy variable is negatively related to bidders’ cumulative return and significant at 1% level in the event window (-10, +10) by using linear regression. Though the results of binary logistic test fail to show the significant results, the results of linear regression is consistent with univariate analysis and the hypothesis 2, which means that acquiring UK firms lead to negative announcement effect.

Relative deal size

Previous studies showed different results as to the relationship between relative deal size and cumulative announcement return. In this paper, we can see that in the ordinary least squares regression, relative deal size shows significant results in all of the samples. In the Chinese subsample, relative deal size shows a significant positive result in the linear regression. It suggests that the ability of competitiveness increases when targets are relative large firms in China, which make investors more confidence and overcome the bad effects of overpay (Gorton et al, 2009).The logistic test and the univariate test fail to confirm the positive sign. I can only state that there is a weak positive relation between relative deal size and CARs when acquiring Chinese firms. In the UK subsample, in line with the univariate test, there is a nearly persistent significantly negative coefficient in relative deal size. The logistic regression also supports that there is a negative relation between relative deal size and bidders’ CARs. Larger relative deal size could create more agency problems and affect managers’ decisions, which stimulate managers to overpay for the targets, thus decreasing the shareholders’ value. In the sample, a bigger mean of relative deal size is shown in the UK subsample than in the China subsample. Probably the American managers are more likely to overpay for the UK targets. I think for the firms in the US, merging relative large firms in the UK will increase the agency problem, managers are more likely to overpay (Morck et al. 1990). In the full sample, both negative and positive results are provided, but the logistic test supports the negative relationship between relative deal size and CARs. In total, the number of firms in the UK is nearly twice of the firms in China. This can be a factor which influences the results of relative deal size in the full sample.

Book to market value

(31)

previous univariate test. There is a persistent negative relation between book to market ratio and CARs in the UK subsample by using ordinary least squares, and all of the coefficients are significant, which is consistent with previous literature that higher risks lead to lower earnings. Higher book to market value ratio represent higher risk but not risk premium. Therefore, the risks in the UK decrease the bidders’ abnormal return. (e.g. Griffin and Lemmon, 2002). According to the management entrench theory, managers face the risk of replacement, in order to avoid replacement, they will do manager specific investments which decrease the profit of shareholders (Shleifer and Vishny, 1989). Higher risk may encourage managers to do more manager-specific investments. The results of logistic regression support the negative signs in the UK subsample. It suggests that higher book to market ratio destroy the value when acquiring firms in the UK, and higher bidders’ book to market ratio will create risk rather than risk premium. The results of the Chinese subsample and the full sample have insignificant coefficients, which mean that the book to market ratio has few impacts when taking acquisition in China.

Payment method

Most scholars hold the idea that paying by cash increase abnormal return (e.g. Travlos, 1987), there is a negative signaling effect if firm pay by stock. I investigate the impact of payment method to cumulative announcement return. Paying by cash has a significant positive effect on CARs in the full sample and the UK subsample by using linear regression and logistic tests, which is consistent with the univariate analysis. Moreover, the logistic test in the China subsample also provided significant result in the window (-10, +10), which also gives some evidence that paying by cash has advantages when acquiring Chinese firms. In short, the results suggest that cash offers increase the value of bidders’ abnormal return.

Firm size, Operational Experience, Other mergers and Diversification

(32)

Taking the univariate regression into consideration, we can notice that the results of firm size are insignificant. Though ordinary least square regression presents two significant results in the Chinese subsample, neither the univariate regression nor the logistic test can support the results. To sum up, the current evidence means that firm size does not have much influence on the bidders’ CARs.

Referring to the operational experience, the previous univariate analysis only provides one significant negative coefficient in the China subsample in the event window (-1, +1). Affected by other variables, operational experience shows one significant result in the event window (-1, +1) of the China subsample and one significant result in the event window (-2, +1) of UK subsample by using ordinary least squares. The signs are negative, meaning that firms with operational experience suffer from value destroy during the M&As. However, I notice that all the significant results in the unvariate and ordinary least square regression are only significant at 10% level, and only a few coefficients show significant results in these two tests. Besides, all the results are insignificant by using the logistic test, which cannot confirm the two significant findings in the linear tests. In the end, there are only weak evidence to support the prediction.

Concerned about other mergers, I predicted that it would have a negative effect on the CARs. For all the samples, the significant result is only showed once in the full sample by using ordinary least squares regression. The negative sign is significant at 10% level in the event window (-10, +10). The previous univariate analysis and the logistic test do not confirm the significant result in the full sample. Therefore, I only find a weak evidence to support the prediction.

At last, the results of diversification are insignificant in cross-sectional analysis in all the samples. In the previous univariate analysis, the bidders’ diversification strategy has a significant result only in the Chinese subsample. In the cross-sectional regression, no coefficients of diversification are significant at any level. It is probably that the diversification is effected by other variables in the cross-sectional regression. As a result, bidders’ diversification or focus strategies do not have large impacts on CARs.

(33)

The number 1 behind event window means the coefficient is estimated by including firm size in the regression model, while the number 2 behind event window means the coefficient is estimated by including relative deal size in the regression model. Besides, the coefficients and significant levels are also shown. p<0.01***, p<0.05** and p<0.1*

Table 7: Summary of Significant results

Univariate analysis Ordinary least square Binary logistic regression Country (-10,+10)Full -0.049** (-10,+10)2Full -0.048***

Diversification (-1,0)China -0.023*

Firm size (-10,+10)1China -0.033***

(-1,0)1China 0.008**

Relative deal size (-10,+10)2Full 0.194* (-2,+1)2Full -6.787** (-1,0)2Full -0.055*

(-10,+10)2China 0.54*

(-2,+1)UK -0.113** (-2,+1)2UK -0.155*** (-2,+1)2UK -8.184** (-1,0)UK -0.059* (-1,0)2UK -0.069**

Leverage

Book to market ratio (-10,+10)UK -0.063* (-10,+10)1UK -0.069* (-10,+10)2UK -0.066*

(-2,+1)UK -0.036** (-2,+1)1UK -0.035** (-2,+1)1UK -1.697** (-2,+1)2UK -0.041*** (-2,+1)2UK -2.199*** (-1,0)UK -0.018** (-1,0)1UK -0.016*

(-1,0)2UK -0.019**

Payment method (-1,0)Full 0.015** (-1,0)1Full 0.016** (-10,+10)1Full 0.764* (-1,0)2Full 0.016** (-10,+10)2Full 0.757* (-10,+10)1China 2.25** (-10,+10)2China 2.109* (-1,0)1China 1.533* (-1,0)UK 0.015** (-1,0)1UK 0.014* (-1,0)2UK 0.014**

Other merger (-10,+10)1Full 0.048*

(34)

6. Conclusion

This paper has examined the bidders’ announcement effects of 107 cross-border acquisitions during the years 2009 to 2013. I choose the US firms as acquirers, UK and Chinese firms as targets. The event window study is used to analyze cumulative abnormal returns in three groups: full - sample, Chinese - subsample and UK - subsample. To begin with, I came up with the first hypothesis to examine whether the M&As can increase or destroy value for bidders' shareholders. The results show that there is no evidence to support the first hypothesis because none of results in the full sample’s event window appear to be significant. The second hypothesis is concerned that the US companies who acquired Chinese firms will obtain a higher abnormal return than those who acquired UK firms. The results of event window study show that when acquiring UK firms, bidders have negative return, while bidders gain positive abnormal returns when acquiring Chinese firms. In the Chinese subsample, the (-10,+10) event window shows a significant positive abnormal return. It is consistent with internalization theory, the transaction cost theory and the positive multinational network hypothesis (e.g. Buckley, 1988; Teece, 1981; Williamson, 1975; Doukas and Travlos, 1988; Kogut, 1983). The findings of second hypothesis also support the empirical studies by Liang (2009).

Next, I use univariate analysis, ordinary least square regressions and logistic tests for the further study. I find that country, relative deal value, bidders’ book to market value and payment method are important determinants for bidders’ cumulative abnormal return. Especially, the signs of

relative deal value and bidders’ book to market value show some differences when acquiring

Chinese firms and the UK firms, which help to explain the different cumulative announcement effects when acquiring Chinese firms and the UK firms. Based on the result analysis, I can make some conclusions about these crucial variables. To start with the relative deal size, there is a weak positive relationship between relative deal size and CARs when acquiring Chinese firms, but a strong negative relationship exist when acquiring the UK firms. Secondly, I find that a higher

bidders' book to market value ratio decreases the bidders’ announcement effects when acquiring

the UK targets. Thirdly, cash offers bring positive abnormal returns to acquirers in all of the sample. Finally, I find that the diversification strategy, firm size, financial leverage, other

mergers and operational experience are not important factors for the US firms to acquire the

firms either in the UK or China.

(35)

announcement effects, while acquiring firms in the developed countries may destroy the value of bidders' shareholders. Through this paper, we can combine more specific factors together in order to provide suggestions for firms who determine to take M&As abroad. Primarily, managers should decide which targets to merge, and the target size or the deal size will definitely affect the bidders' returns. It is suggested that when taking acquisitions in the developed countries, managers should avoid the behavior of overpay for big targets. Also based on the results of this paper, the target size or deal size is not an issue when taking acquisitions in the developing countries like China. But in my opinion, bidders should be careful about choosing the targets' size or deal size in China. The size of a company does not stand for its inside capability, and the amount of money that bidders pay for targets does not means the real value creation capabilities of the targets. In China, some firms are bigger because of historical or political reasons, and their technologies are not updated. Besides, the bigger firms usually include a stronger bureaucratic atmosphere. This makes it hard for managers to know the inside structure of the targets or to control the employees there. In terms of the payment method, cash offering is an optimal choice for managers to take acquisitions in either advanced or advancing countries. If managers determine to use stock payment, it does no matter which countries that they are planed to step in. Because paying by stock will lead to negative returns in any cases.

(36)

topic investigated by many scholars(e.g. Xin and Pearce, 1996). The environment, culture and tradition in China are different from the condition in the US, which can causes many problems like cultural conflicts, leading to the value destruction in the acquisition. The managers' knowledge of target countries can directly influence the whole process of the acquisition. Managers can choose to merge in the country where they have some acquisition experiences, or the country where they have known well, so that they can avoid some unnecessary loss.

Some limitations still remain in this paper. First of all, the sample data is comparatively small due to the limitation of data. Especially the Chinese subsample, only 32 acquisitions are included. This restriction undermines the explanatory power of some variables and increases the difficulty to achieve significant results. We can see that the UK subsample show more significant results than the Chinese subsample. If a larger Chinese subsample included, a more significant results we will get. Secondly, I test the events only in three event windows in the study. Some scholars include a wide range of even windows in the research. For example, Aybar and Ficici (2009) analyze seven event window periods, and some results are varied from different windows. Thirdly, I only measure the short-term announcement effects. Long-term abnormal returns are also helpful to evaluate the value creation of M&As. For instance, Loughran and Vijh (1997) find that during a five-year period following the acquisition, firms earned significantly positive excess returns of 61.7%. Long-term analysis may give a different result from short-term analysis.

(37)

References

Agrawal, A., Jaffe, J. F., and Mandelker, G. N. (1992). The Post-Merger Performance of Acquiring Firms: A Re-examination of an Anomaly. The Journal of Finance, 47(4), 1605-1621.

Al-Sharkas, A. A., Hassan, M. K., and Lawrence, S. (2008). The impact of mergers and acquisitions on the efficiency of the US banking industry: further evidence. Journal of

Business Finance and Accounting, 35(1-2), 50-70.

Asquith, P., Bruner, R. F., and Mullins Jr, D. W. (1983). The gains to bidding firms from merger. Journal of Financial Economics, 11(1), 121-139.

Asquith, P., and Kim, E. (1982). The impact of merger bids on the participating firms' security holders. The Journal of Finance, 37(5), 1209-1228.

Ariff, M., and Finn, F. J. (1989). Announcement effects and market efficiency in a thin market: An empirical application to the Singapore equity market. Asia Pacific Journal of

Management, 6(2), 243-265.

Aybar, B., and Ficici, A. (2009). Cross-border acquisitions and firm value: An analysis of emerging-market multinationals. Journal of International Business Studies, 40(8), 1317-1338.

Barber, B. M., and Lyon, J. D. (1997). Detecting long-run abnormal stock returns: The empirical power and specification of test statistics. Journal of financial economics, 43(3), 341-372. Barry, C., and Brown, S. J. (1985). Differential information and security market

equilibrium. Journal of Financial and Quantitative Analysis, 20(4), 407-422.

Bashir, A., Sajid, M.R. and Sheikh, S.F. (2011). The Impact of Mergers and Acquisitions on Shareholders Wealth: Evidence from Pakistan. Middle-East Journal of Scientific Research, 8-1, 261-264.

Beamish, P. W., and Banks, J. C. (1987). Equity joint ventures and the theory of the multinational enterprise. Journal of International Business Studies, 1-16.

Berger, P. G., and Ofek, E. (1995). Diversification's effect on firm value. Journal of financial

economics, 37(1), 39-65.

Berkovitch, E., and Narayanan, M. P. (1993). Motives for takeovers: An empirical investigation. Journal of Financial and Quantitative Analysis, 28(3), 347-362.

(38)

of strategic motives and performance. Thunderbird International Business Review, 50(4), 259-270.

Bruner, R. F. (2002). Does M&A pay? A survey of evidence for the decision-maker. Journal of

Applied Finance, 12(1), 48-68.

Buckley, P. (1988).The limits of explanation: Testing the internalization theory of the multinational enterprise [J]. Journal of international business studies, 19(2): 181-193. Caves, R. E. (1981). Intra-industry trade and market structure in the industrial countries. Oxford

Economic Papers, 33(2), 203-223.

Chang, S. (1998). Takeover of Privately Held Targets, Methods of Payment and Bidder Returns.

The Journal of Finance, 2, 773-784.

Corhay, A., and Rad, A. T. (2000). International acquisitions and shareholder wealth Evidence from the Netherlands. International Review of Financial Analysis, 9(2), 163-174.

Denis, D. J., Denis, D. K., and Yost, K. (2002). Global diversification, industrial diversification, and firm value. The Journal of Finance, 57(5), 1951-1979.

Dhaliwal, D. S., Lee, K. J., and Fargher, N. L. (1991). The association between unexpected earnings and abnormal security returns in the presence of financial leverage. Contemporary

Accounting Research, 8(1), 20-41.

Dichev, I. D. (1998). Is the risk of bankruptcy a systematic risk? The Journal of Finance, 53(3), 1131-1147.

Doukas, J., and Travlos, N. G. (1988). The effect of corporate multinationalism on shareholders' wealth: Evidence from international acquisitions. The Journal of Finance, 43(5), 1161-1175. Eckbo, B. E. (1992). Mergers and the value of antitrust deterrence. The Journal of Finance, 47(3),

1005-1029.

Elgers, P. T., and Clark, J. J. (1980). Merger types and shareholder returns: Additional evidence. Financial Management, 66-72.

Fama, E. F., and French, K. R. (1995). Size and book -to-market factors in earnings and returns. The Journal of Finance, 50(1), 131-155.

Farrell, J., and Shapiro, C. (1990). Horizontal mergers: an equilibrium analysis. The American

Economic Review, 107-126.