Country-level determinants of completion of cross-border mergers and acquisitions and risk-taking

(1)

Country-level determinants of completion of cross-border mergers and acquisitions and risk-taking

Master Thesis June 9, 2017

Student name: Jinkang Li Student number: S3016714 Study Programme: MSc IFM Supervisor: Dr. Halit Gonenc Co-assessor: Prof. Niels Hermes

Abstract: The purpose of this thesis is to examine country-level determinants of the likelihood of completion of cross-border mergers and acquisitions (M&As), and also the effects of completion on an acquirer’s subsequent risk-takings. Based on a worldwide sample of 1903 cross-border M&As during the period of 2006-2010, this thesis documents potential determinants affecting the likelihood of completion of cross-border M&As. The results show that a cross-border M&A is more likely to succeed in a home country with a high level of economic freedom but less likely to be completed when there is significant cultural distance between the two countries involved in the M&A. Moreover, the effect of the completion of a cross-border M&A has a significant impact on an acquirer’s subsequent risk-takings with the shareholder protection in home country. The evidence also shows that the cultural dimension of individualism has a positive and significant effect on risk-taking.

Field Keywords: Cross-border M&As, completion and withdrawal, economic freedom,

country culture, corporate risk-taking

(2)

Acknowledgement

In completing this master thesis, I am very thankful to my supervisor, Dr. Halit Gonenc for his professional instructions and constant supports. Writing this thesis has greatly broadened my international perspective and scientific understanding of this research subject. Dr. Halit Gonenc has provided me with a deeper understanding of how to think as an IFM student by always remembering “international”, “financial”, and “management” in heart.

I also want to thank Prof. Niels Hermes for assessing my thesis.

Moreover, I really appreciate the firm supports from my girlfriend and my parents. Without

their encouragements and help, it is would be difficult for me to successfully complete my thesis.

(3)

1. Introduction

In recent decades, the wave of globalization has provided huge opportunities for multinational enterprises to expand their businesses. In 2015, the total turnover from global Mergers and Acquisitions (M&As) exceeded $4 trillion, with cross-border M&As accounting for more than $1 trillion of these deals (Deloitte, 2015). According to World Investment Report 2016, although the majority of outward foreign direct investment (FDI) was created by developed economies, emerging markets have grown significantly and contributed to more than a quarter of global FDI in 2016. For example, China, which invested $128 billion, became the third largest investor (UNCTAD, 2016). Nevertheless, apart from great achievements, recent data show that a large number of cross-border acquisitions made by Chinese acquirers were cancelled in 2016 due to the regulation of Chinese Government (Financial Times, 2017).

Withdrawal from a cross-border M&A will result in substantial costs, such as penalties and the damage of corporate image, to acquirers (Zhou et al., 2016). Based on institutional theory (North, 1990; Buckley et al., 2007; Du and Boateng, 2015), withdrawal from deals shows that cross-border acquirers are subject to specific country-level institutional environments.

Moreover, as an international investment activity, the success or failure of cross-border M&As may have further impacts on an acquirer’s future risk-taking.

In the M&As literature, most studies focus on the effect of cross-border M&As on an acquirer’s financial performance (e.g. Aybar and Ficici, 2009; Lebedev et al., 2015), but scholars overlook reasons for completion or withdrawal of cross-border M&As. The current studies about the completion or withdrawal from M&As mostly capture the features of acquirers from developed countries (Dikova et al., 2010), BRIC (Brazil, Russia, India, and China) countries (Zhou et al., 2016; Zhang et al., 2011) or specific industries (Muehlfeld et al., 2012).

The impacts of completion/withdrawal on acquirers’ risk-taking are still ambiguous. Hence, this thesis aims to fill this gap and answer the following questions: what are the effects of country-level factors on the likelihood of completion of cross-border M&As, and how is the acquirers’ risk-taking affected by the completion of cross-border deals and country-level factors?

With regard to the factors influencing the likelihood of completion of cross-border M&As,

(4)

this thesis expects a positive effect from the economic freedom in an acquirer’s home country and a negative effect from the cultural distance between the two countries involved in a deal.

Moreover, the prior experience of completion of cross-border M&As will moderate the relationship between country-level factors and the likelihood of completion of cross-border M&As. With regard to the model of risk-taking, this thesis expects that the completion of cross- border M&As has a positive effect on risk-taking. This thesis also expects a positive effect of shareholder protection and the cultural dimension of individualism on risk-taking, but a negative effect from the cultural dimension of uncertainty avoidance. Moreover, the shareholder protection may moderate the relationship between the completion and risk-taking.

This thesis analyzes a sample of cross-border M&As from all around the world between 2006 to 2010. Binary probit regression and OLS regression estimations are applied to test the impacts of country-level factors on the likelihood of the completion of deals and the effect of completion on corporate risk-taking.

The findings of this study show that economic freedom in the acquirer’s home country can positively increase the likelihood of completion of cross-border M&As. Moreover, the cultural distance between home and host country negatively affects the completion. However, the experience does not show a significant moderating effect on cultural distance. With regard to the risk-taking model, the completion of focal cross-border M&As does not provide a significant effect on acquirers’ risk-taking, which is against the expectation. However, the results suggest that the shareholder protection in the acquirer’s home country significantly moderate the relationship between completion and risk-taking. After running the subgroup analysis, in high protection countries, the completion of cross-border M&As has significantly negative effects on risk-taking; however, in low protection countries, the completion has significantly positive effects on risk-taking. Lastly, the cultural dimension of individualism suggests a positive and significant impact on risk-taking, which supports expectation. However, the cultural dimension of uncertainty avoidance and the acquirer’s home country shareholder protection do not have significant effects on corporate risk-taking.

This thesis makes several contributions. Firstly, it examines the important period before

(5)

the completion of cross-border M&As. Many announced cross-border M&As collapse before completion, but many scholars ignore this important stage. This thesis confirms two country- level determinants for completion of cross-border M&As: the economic freedom and cultural distance. Secondly, this thesis uses a broader sample by incorporating observations form developed and emerging markets, which provides better generality for this kind of research.

Thirdly, this thesis goes further than extant literature by examining the effect of completion of cross-border M&As on an acquirer’s subsequent risk-taking. Hence, this thesis starts a discussion of the potential effects of completion/withdrawal of cross-border M&As on firm activities and performance.

The rest of this thesis is organized into following sections: Section 2 will present the literature related to this thesis and formulation of corresponding hypotheses. Section 3 will present the data collection and methodology applied while Section 4 will discuss the results and data analysis. Section 5 will present the conclusion and limitations of this thesis.

2. Literature review and hypotheses 2.1. Procedure for cross-border M&As

In order to investigate the impact of country-level factors on completion of cross-border

M&As, the detailed procedure for general M&As is discussed in this section. Due to the fact

that an M&A is not a one-step activity, a completion of M&As comprises several phases, but

in general, there are two main phases: the pre-completion stage and the post-completion stage

(Boone and Mulherin, 2007; Dikova et al., 2010; Muehlfeld et al., 2012; Zhou et al., 2016). As

shown in Figure 1, Zhou et al. (2016) summarize the process for completion a M&A. The first

step in the pre-completion stage involves a complex private takeover process which takes place

before the public announcement (Boone and Mulherin, 2007). During this process, the target

firms will hire a dealer (mostly from investment banks) to help companies to search for potential

buyers. Next, the selling firm will organize an auction with those intended buyers, and the final

winner of bidding firm will be determined during this process. After the public announcement,

the M&As will enter the public takeover period. During this second section of the pre-

(6)

completion phase, the acquirer will conduct deeper investigations regarding the due diligence of the target firm (Zhou et al., 2016). Due to the complexity and uncertainty of M&A related issues, this process can take several months or more time to finalize (Dikova et al., 2010). For cross-border deals in particular, more country-level factors should be taken into consideration, and the process may take more time than a domestic M&As. When such investigations are completed, both parties will determine whether the M&A will be completed or cancelled. Once the M&A is announced to be completed, the date of completion is called the “date effective”, and the post-completion phase of M&A will begin, including the detailed integration plan for both firms.

Figure 1. The procedure for M&As (source from Zhou et al., 2016)

The focus of this study is the period between the public announcement date and “data effective”. During this time, the acquirer will conduct in-depth activities regarding the further issues of the M&A. Muehlfeld et al. (2012) state that the activities in the public takeover period include two major types: strategic and administrative activities. According to Muehlfeld et al.

(2012), the strategic activities refer to compliance with regulations and future strategies, while administrative activities include issues relating to financial assessment, final negotiations, and communications. In cross-border acquisitions, those activities can significantly affect the process of M&As, because the potential risks may arise after the evaluation and the acquirer may terminate the deal to avoid such risks. When additional information is added for the bidder’s consideration, both parties will renegotiate the current contracts to avoid future problems. As a consequence, bidders may withdraw from the deals if the existing problems cannot be resolved (Zhou et al., 2016). Specifically, bidders in cross-border M&As deals will bear a lot of costs for firm-specific, deal specific or country-specific factors (Aybar and Ficici, 2009).

Empirical evidence suggests that the distance between two countries’ formal institution

(7)

(i.e. regulation, legal systems) and informal institutions (i.e. language, culture) will negatively affect the completion of cross-border M&As (Dikova et al., 2010; Zhou et al., 2016). Moreover, the industry relatedness between the acquiring firm and the target firm will increase the likelihood of completion of cross-border M&As (Lim and Lee, 2016). However, the extant literature ignores the important role of the acquirer’s home country and the study on this theme is still in its infancy. This thesis will fill some gaps by investigating the impact of the home country’s economic freedom on the completion of cross-border M&As.

2.2. Corporate risk-taking

In the corporate decision-making, risk is commonly perceived as the potential variation in the allocation of possible outcomes and their likelihoods (March and Shapira, 1987). For example, a cross-border M&A is a risky decision making for the firm because it involves a lot of uncertainties. At the same time, the success (completion) of a cross-border M&A will bring advantages to the firm as well, including entering the new markets and acquiring new knowledge. Brockhaus (1980) defines the propensity for risk-taking as “the perceived probability of receiving the rewards associated with the success of a proposed condition, which is required by people before they are willing to subject themselves to failures, the alternative condition providing fewer rewards and less severe outcomes than the proposed one”. According to this definition, corporate risk-taking can reflect the managerial trade-off of between risks and the benefits of their investments. Managerial risk-taking is, to some extent, fundamental to decision-making, and is critical to firm survival and development (Shapira, 1995; Li and Tang, 2010; Boubakri et al., 2013).

Moreover, corporate risk-taking can be affected by several factors. Current studies on the determinants of the corporate risk-taking mainly focus on the corporate governance (John et al., 2008; Phan and Hegde, 2013), political institutions (Dey, 2010; Boubakri et al., 2013), national cultures (Li et al., 2013) and ownership structure (Acharya et al., 2011; Chittoor et al., 2015).

However, they neglect the potential impacts of the success or failure of previous investment

(e.g. cross-border M&As). Based on this situation, this thesis argues that the completion of a

(8)

cross-border M&A deal can affect the subsequent risk-taking behaviors of the acquirer. At the same time, several country-level factors, such as shareholder protection and national culture, also worn on corporate risk-taking. Further discussion and detailed theory are presented in Section 2.4.

2.3. Country-level factors and the completion of cross-border M&As 2.3.1 Formal institutions in the acquirer’s home country

According to the institutional theory, the institutional differences between countries refer to “the different rules of game” in specific countries (North, 1990). In cross-border M&As, those “rules of game” can be regulatory (laws and regulations), cognitive (widely accepted and shared social cognition), and normative (national cultures) (Scott, 1995; Du and Boateng, 2015).

Formal institutional factors are, otherwise, existing laws and rules that vary across jurisdictions.

A recent example shows that cross-border M&As made by Chinese acquirers were cancelled because of regulatory restrictions by the Chinese Government (Financial Times, 2017). Among those withdrawals, the biggest one was the attempted $14 billion takeover of Starwood Hotels

& Resorts in the U.S by Chinese Anbang Insurance. The reason for cancelling this deal was that the Chinese government judged the attempt to be an “irrational” acquisition” (Financial Times, 2017). Given the above example, it can be seen that the acquirers are profoundly affected by their formal institutional environment (Buckley et al., 2007).

Considering this recent news, the completion of cross-border M&As is affected by the

acquirer’s home country institutions. This study is related to the economic freedom across

countries. Economic freedom is used to measure a country’s long-term economic development

based on a macroeconomic perspective (Wang and Wang, 2012). Heckelman (2000) states that

the economic freedom of a nation can boost domestic economic growth. Domestic economic

growth can further promote successful cross-border investment opportunities (e.g. cross-border

M&As, cross-border venture capital). After studying 10205 cross-border venture capital

investments, Wang and Wang (2012) find that firms located in countries that have greater

economic freedom are more likely to successfully exit through IPO (initial public offering) or

(9)

M&As. However, in countries with less economic freedom, business opportunities are diminished by poor institutional infrastructures (La Porta et al., 1998; Aybar and Ficici, 2009).

Dikova et al. (2010) state that the environmental complexity will harm the completion of cross- border acquisitions. Cui and Jiang (2012) state that a home country’s regulatory restriction constrains the outward FDI of firms. For example, Chinese firms operate in an economy that is not as free as developed countries because they are profoundly subject to the government regulations. In this home country institutions, cross-border M&As are very likely to be delayed or cancelled if the M&A is not in line with national interests (Cui and Jiang, 2012).

Current literature focuses on the formal institutional distances between the home and host country (Dikova et al., 2010; Zhou et al., 2016), and they generally find that formal institutional distance will decrease the likelihood of completion. Other studies investigate the impacts of formal institutions in target countries (Meyer et al., 2009; Zhang et al., 2011). Using the economic freedom index as the proxy of formal institutions, Meyer et al. (2009) find that the better host country institutional quality leads to more international acquisitions. Similarly, Zhang et al. (2011) find that the quality of formal institutions in the target country has a positive influence on the likelihood of completion of cross-border deals. The interpretation of these findings is that the high quality of institutions can facilitate the process of M&As. Referring to the case of “Chinese cross-border deals cancelled” (Financial Times, 2017), it is plausible to expect that a home country with more economic freedom can boost domestic firms’

internationalization because there are fewer barriers and restrictions from the home country’s government. Therefore, this study hypothesizes:

Hypothesis 1: the economic freedom of home country has positive effects on the likelihood of completion of cross-border M&As.

2.3.2. National cultural distance

Based on institutional theory, national culture belongs to the informal institutions of a

country. The most commonly used definition for national cultural distance is the collective

programming of the mind that differentiates one group of members from another (Hofstede,

(10)

2001). In cross-border M&As, such “collective programming of the mind” refers to the shared language, norms, educations, and religion of specific countries (Reus and Lamont, 2009).

Specifically, due diligence before deals’ completion will focus on issues such as future strategy, adjustment of organizational structure in the target firm, and regulatory compliance.

However, these important issues are all always rooted in their own national cultures which will significantly affect the M&A process. Negotiations between the acquirer and the target managers always cover controversial issues, such as valuation and pricing, as well as its process (Popli et al., 2016). When there is a very high cultural distance between acquirer and target, managers of acquirers will predict greater conflicts and higher costs of integration in the future (Brock, 2005). Lack of cross-cultural knowledge can easily result in misunderstandings between two parties and make the negotiations more complex (Popli et al., 2016). Under such tense pre-completion circumstances, the trust between the two parties will be depressed. Dikova et al. (2010) point out that the trust of target managers can act as an important deal-breaker in the cross-border M&As process. Malhortra and Gaur (2014) suggest that one party’s inability to adoption or understanding to the particular cultures of the other party is the major determinant of many cross-border M&As failures.

Moreover, in terms of other dimensions of informal institutions, Riad and Vaara (2011)

state that the combination of national metonymy also contributes to differences in national

cultures. At the same time, since M&As require acquirers to complete the multiple pre-

completion activities in a limited time with limited information (Very and Schweiger, 2001),

different languages will lower the bidder’s information collection and communication between

the two parties, which will hinder the cross-border M&As process. Empirical evidence provided

by current literature also proves those arguments. Dikova et al. (2010) find the cultural

differences between home and host countries (at Power Distance and Uncertainty Avoidance

dimensions) have significantly negative effects the completion of deals by using 2389

announced cross-border acquisition. Popli et al. (2016) investigate the effects of cultural

distance between India and target countries on the completion of cross-border acquisitions, and

they find a cross-border deal is less likely to be completed when a great cultural distance exists.

(11)

In summary, cultural distance is more likely to make firms avoiding risks. Therefore, this study hypothesizes:

Hypothesis 2: cultural distance between the acquirer’s home country and the host country has negative impacts on the likelihood of completion of cross-border M&As.

2.3.3. Moderating effects of prior experience

As discussed above, cross-border acquirers are subject to specific formal and informal institutional settings. These institutional settings not only affect the activities before the completion of M&As, but also shape the acquiring firm’s adaptive efficiency (Dikova et al., 2010). Therefore, different institutional settings provide firms with different motivations to learn tacit knowledge and to develop specific skills (North, 1990; Dikova et al., 2010). M&A experience enables firms to build their specific routines on how to screen, select, and acquire a company (Vermeulen and Barkema, 2001; Zhang et al., 2011). As successful experiences accumulate, firms are able to integrate other companies at an appropriate level, and deal with troublesome administrative problems (Vermeulen and Barkema, 2001; Zhang et al., 2011).

Furthermore, the prior M&As experience can be split into domestic or international

experience and successful or failed experience. The focus of this study is the completed cross-

border deals, and what is more likely to increase the likelihood of completion. In the context of

cross-border M&As, acquiring firms have to learn how to survive and operate in diverse

institutions and cultural settings (Vermeulen and Barkema, 2002; Collins et al., 2009). In

contrast, firms with domestic M&A experience only obtain the general knowledge about

acquiring a company, but such experience does not enable them adapt to complex international

activities. Moreover, according to the organizational learning perspective (Very and Schweiger,

2001; Rabbiosi et al., 2012), a failed international experience can also force organizations to

evolve. However, learning from failure is a complex process. Muehlfeld et al. (2012) state that

there are three questions when firms want to learn from failure: which of routines need to be

modified, how can the original routines be improved, and how can the changes be implemented

smoothly? Learning from failure can be a more complex process than learning from success.

(12)

Therefore, this thesis focuses on the completed cross-border M&As as the previous experience measurement in this study.

In the “Chinese cross-border deals cancelled” (Financial Times, 2007) example, Chinese acquirers are severely subject to the government’s administrative system (Cui and Jiang, 2010;

Cui and Jiang, 2012). The main purposes of this administrative system are to guard state assets, to prevent capital flight, and to direct the international activities of firms in ways that are consistent with national interests (Cui and Jiang, 2010; Cui and Jiang, 2012). In this institutional environment, the firms will follow the past practices approved by the government in order to obtain the legitimacy (Cui and Jiang, 2012). Therefore, firms that have a rich experience of completed cross-border M&As are more likely to survive in a market that has less freedom.

However, in home countries with highly free markets, there is less intervention and regulatory restrictions from the government.

Zhang et al. (2011) state that the transaction costs and regulatory uncertainty remain stable and transparent in an advanced institution. In this case, the market performs better and can accurately offer the information that is necessary for firms to conduct a cross-border M&A.

Hence, a higher level of market freedom in the home country will facilitate an acquirers’ ability to purchase foreign firms. Therefore, the prior cross-border M&A experience will be less effective in home countries with higher economic freedom relative to countries with lower freedom. Also, since the acquirers must depend on past practices in a poor institutional environment to make decisions, the experience plays a more predominant role in home countries with low economic freedom than in countries with high economic freedom.

In addition, the previous experience can be helpful for bidders to mitigate the negative

effects of cultural distance because acquirers can more efficiently evaluate target firms (due

diligence) and become familiar with the international environment and foreign markets (Dikova

et al., 2010). Moreover, Collins et al. (2009) explain that such positive effects from previous

M&As result from enhanced knowledge and broadened networks and richer communications

in the global venues. Also, Zhou et al. (2016) suggest that bidders can better negotiate and

generate strategies in host countries with the help of previous completion experience. Prior

(13)

experience can enable acquirers to develop a correct model to deal with high cultural distance circumstances. Also, experienced acquirers can more efficiently hire the local experts and professionals or build a multi-cultural team to help them to overcome the cultural barriers.

Empirical evidence provides support for the moderating effects of prior successful experience. Using a sample of 2389 cross-border acquisitions from 1981 to 2001, Dikova et al.

(2010) find that past completed experience can significantly increase the likelihood of the completion of cross-border M&As as well as shorten the duration of the completion process.

They also find that the moderating effects of experience on the relationship between cultural distance and completion are significant. Similarly, Zhou et al.’s (2016) study of 15 years’ of cross-border M&As in BRIC countries finds that the acquirer’s successful M&A experience can significantly increase the likelihood of completion of both inbound and outbound deals.

Moreover, after studying 1324 Chinese cross-border M&A announcement, Zhang et al. (2011) find that the experience has a predominant positive impact on the completion of cross-border M&As in countries with relatively poor institutions. Furthermore, none of the above studies examine the impacts of experience on the relationship between home country economic freedom and cross-border M&A completion.

In summary, based on the above discussion, this thesis hypothesizes:

Hypothesis 3a: the prior completion experience of bidders moderates the effect of home

country economic freedom on the cross-border M&A completion. The positive relationship between home country economic freedom and the likelihood of deal completion becomes weaker as past experience increases.

Hypothesis 3b: the prior completion experience of bidders moderates the relationship

between cultural distance and completion of cross-border M&As. The negative relationship between cultural distance and the likelihood of deal completion becomes weaker as past experience increases.

Figure 1 (below) presents the first conceptual model of this study, including two country-

level variables, one moderator, and one dependent variable.

(14)

Figure 2. Conceptual model (1): country-level factors and completion of cross-border M&As (source: the author creates this model)

2.4. Completion of cross-border M&As, country-level factors, and corporate risk-taking 2.4.1. Completion and corporate risk-taking

Literature suggests that the past M&A experience is positively related to a firm’s subsequent M&A activities (Haleblian et al., 2006; Collins et al., 2009; Rabbiosi et al., 2012).

Cross-border M&A is more likely a process of exploration, which includes activities such as search, risk-taking, discovery, and innovation (March, 1991; Rabbiosi et al., 2012). Since a cross-border M&A itself is a risky investment, it is plausible to expect the success of a firm’s attempt to invest abroad would affect their subsequent risk-taking behavior. Based on the behavioral perspective (Li and Tang, 2010; Chittoor et al., 2015), studies link the corporate risk- taking with a CEO’s personal risk propensities. This study extends this perspective from CEO to the entire management team. Haleblian et al. (2009) suggest that the past international experience of management may shape the strategic and acquisitive behavior of firms. On the one hand, some scholars state that the successful cross-border M&As can promote the

Country-level

Independent variable (1) cultural distances

Independent variable (2) acquirer nation's economic

freedom

Moderator

prior experience of cross-

border M&As completion

(15)

organizational learning and improve the international knowledge when dealing with risky investments (e.g. Vermeulen and Barkema, 2001; Collins et al., 2009). From this point of view, the positive effect of learning will help the rational investment of a company’s management.

On the other hand, based on Li and Tang’s (2010) summary of about CEO hubris, the international experience may lead to managerial overconfidence. Under this negative impacts, a firm’s might overestimate their problem-solving abilities and underestimate the uncertainties and risks the firm will face in the future. Li and Tang (2010) find that the CEO hubris (overconfidence) has a positive impact on corporate risk-taking. Similarly, Chittor et al. (2015) find that the international experience of the firm is positively correlated with the firm’s risk propensities.

Although no study directly examines the effects of completion/withdrawal (as an independent event) of cross-border M&As on an acquirer’s risk-taking, the literature about the relationship between past international experience and subsequent international activities sheds some light on this question. When a firm completes a cross-border M&A, the successful experience enables the firm’s management to participate in risky investments in the future (e.g.

conducting more takeovers). At the same time, the past success can also foster the managerial overconfidence, which leads to more unreasonable risk-taking behaviors. Therefore, the author proposes the following hypothesis:

Hypothesis 4: the completion of cross-border M&As has a positive impact on an acquirer’s risk-taking.

2.4.2. Home country shareholder protection and corporate risk-taking

Based on institutional theory (North, 1990; Scott, 1995; Buckley et al., 2007; Du and

Boateng, 2015), firms are subject to specific institutions (“rules of game”). A country’s

institutional settings also have an impact on firms’ risk-taking behavior. Boubakri et al. (2013)

introduce a double-mechanism approach to study how the political institutions affect corporate

risk-taking. Specifically, the first-stage impact of a political institution may affect corporate

risk-taking by influencing a firm’s operations. The second-stage effects of political institutions

(16)

result from their influences on the legal systems and corporate governance. In this study, the focus of formal institution is the shareholder protection by the acquirer’s nation. According to Bris and Cabolis (2008), shareholder protection is the protection stipulated by the corresponding law or code of a firm’s home country. In terms of shareholder protection, John et al. (2008) state that shareholders suffer from the expropriation of corporate insiders who only pursue self-interests at the expense of minority shareholders. On the one hand, in a country with better shareholder protection, conservative behaviors of management are mitigated by better protection and monitoring (John et al., 2008). The relationship between shareholder protection and corporate risk-taking can be related to the agency theory (Jensen and Meckling, 1976). In a country with better shareholder protection, the corporate insiders will make managerial decisions in the shareholders’ interest, which enhances the corporate risk-taking in order to maximize the shareholder values. On the other hand, stronger corporate governance may decrease the corporate risk-taking because managers will have greater discretion to choose a conservative investment in an environment with better investor protection (John et al., 2008).

With regard to the empirical evidence, using ten year’s data from 39 countries, John et al.

(2008) find that the stronger shareholder protection can significantly improve a firm’s riskiness.

Moreover, Phan and Hegde’s (2013) study on the defined benefit asset allocation of 329 firms finds that the firms with better external governance take more risks than firms with poor external governance, where the external governance is measured by the anti-takeover provisions (the more provisions the worse external governance).

In summary this thesis proposes a hypothesis as follows:

Hypothesis 5a: the shareholder protection of acquirer’s home country is positively associated with acquirers’ risk-taking.

Apart from the direct effects on risk-taking, this thesis argues that the home country

shareholder protection moderates the relationship between completion and subsequent risk-

taking. Based on the argument in Section 2.4.1, the completion is viewed as a successful

experience that may increase the subsequent risk-taking behaviors of acquirers. Agency theory

(17)

(Jensen and Meckling, 1976) suggests that shareholders’ interests should be protected from being expropriated by managers or block holders. In strong protection countries, there are fewer opportunities for managers or block holders to exercise control for their own interest. After the completion of cross-border M&As, strong shareholder protection endows shareholders’ rights that urge the managers to duplicate past success in cross-border M&As.

In a weak shareholder protection environment, managerial discretion allows the managers to seek their self-interests and perform risk-aversely. Without proper monitoring, it is hard to recall the management when managers engage in value decreasing activities (e.g. empire building, shirking) (Core et al., 2006). Moreover, with regard to the principal-principal problems, in weak protection country, the block holders (e.g. large family owners) exercise their control by controlling the board (Bammens et al., 2011). Large shareholders can hamper the value-added risk-taking behaviors through imposing their relatives, who are no longer qualified to manage the firms, as managers (Anderson and Reeb, 2003). Therefore, managers would still behave conservatively to protect their self-interests after the completion of a focal cross-border M&A.

On the other hand, shareholder protection in home country may negatively moderate the relationship between completion and risk-taking. According to Shleifer and Vishny (1986), the presence of dominant ownership can be helpful to reduce the managerial discretion. However, the improved shareholder protection may lead to reduced dominant ownership of a firm, then result in greater managerial discretion (John et al., 2008). Hence, the completion of a cross- border M&A can be just an accidental event for acquirers, and the risk-taking may still decrease.

Moreover, since a cross-border M&A does not necessarily improve a firm’s performance, the completion can lead to a decrease in shareholders’ value. Core et al. (2006) state that the weak shareholder protection can protect managers from the results of the poor performance of good projects, hence encouraging managers to perform more risk-taking. Therefore, it also plausible to believe that managers can behave in a risk-taking manner in weak shareholder protection country after the completion of cross-border M&As.

In summary, this thesis proposes the following hypothesis:

(18)

Hypothesis 5b: the home country shareholder protection moderates the relationship between cross-border M&A completion and acquirers’ risk-taking.

2.4.3. National culture and corporate risk-taking

Apart from shareholder protection, national culture may also affect corporate risk-taking as well. Hofstede (2001) defines the national culture as the collective programming of the mind that differentiates the groups of members from each other. Based on the institutional theory, cultures can not only influence individual behaviors, but also corporate activities, such as management styles and the decision-making process (Kirkman et al., 2006; Popli et al., 2016).

This study focuses on two of Hofstede’s cultural dimensions that may affect corporate risk- taking: uncertainty avoidance and individualism. In line with Kwok and Tadesse (2006) and Li et al. (2013), culture difference across countries can shape the financial activities of firms. For example, Kwok and Tadesse (2006) indicate that countries like Germany and Japan have higher values in uncertainty avoidance, so their financial systems are dominated by bank-based institutions. However, Anglo-Saxon countries such as the U.S. and the U.K. have equity-based markets. Li et al. (2013) explain that their differences result from the acceptance of ambiguity.

Hofstede (1991) defines the uncertainty avoidance as “the extent to which the members of a culture feel threatened by uncertain conditions.” In a high uncertainty avoidance environment, individuals avoid the risky investments because they feel anxious and uncomfortable (Li and Zahra, 2012). In contrast, in low uncertainty avoidance, individuals can adopt the ambiguity of their investments and take more risks. Furthermore, individualism means a loose tie between the society and individuals (Hofstede, 1991). In an individualistic society, members depend more on contractual rules and arm’s length transactions (Steensma et al., 2000; Li and Zahra, 2012), instead of collective decision making. Therefore, the individualistic managers may take more risks than collectivistic managers because they want to achieve personal accomplishment by using their own judgment (Hofstede, 1980; Kreiser et al., 2010).

Based on Hofstede’s (1980, 2001) cultural dimensions and Schwartz’s (1994, 2004)

cultural value orientation, Li et al. (2013) find individualism is positively related corporate risk-

(19)

taking, whereas uncertainty avoidance and cultural harmony are negatively related to risk- taking.

In summary, this study proposes the following hypotheses:

Hypothesis 6a: the uncertainty avoidance in an acquirer’s country has a negative impact

on an acquirer’s subsequent risk-taking.

Hypothesis 6b: the individualism in an acquirer’s country has a negative impact on an

acquirer’s subsequent risk-taking.

Figure 3 (below) presents the conceptual model of acquirer’s risk-taking after the completion of focal cross-border M&As, including completion and country-level factors.

Figure 3. Conceptual model (2): cross-border M&As completion, country-level factors, and acquirer’s risk- taking (source: the author creates this model)

3. Sample and methodology 3.1. Sample

For this thesis, the cross-border M&As deals are collected from the Thomson Reuters mergers and acquisitions database, including the completed and uncompleted (without a completed date provided) deals. Moreover, firm-level financial data are collected from DataStream by Thomson Reuters. This thesis aims to catch the recent cross-border M&As characteristics and also ensures the sufficient data (i.e. the research requires at least five years’

information about acquirers’ return on assets). Hence, this research applies the timeframe from

Dependent variable Acquirer's risk-taking after the completion/withdraw of cross-border

M&As

Independent variables (1) Completion of cross-border

M&As

Independent variables (3) uncertainty avoidance and

individualism Independent variable (2) and

moderator Shareholder protection

(20)

01/01/2006 to 31/12/2010. The sample will be selected based on the following criteria:

(1) Financial institutions are excluded

(2) The acquiring firms must be publicly traded firms while target firms can be listed or unlisted (3) According to Muehlfeld et al. (2012) and Bollaert and Delanghe (2015), Thomson Reuters database requires at least a 3% stake with $1 million deal value for a M&A deal to be included.

This thesis also applies these two criteria.

Table 1. The sample selection process

Process Criteria Result

Step 1 Cross-border Thomson Reuters database provides 15,364 cross-border mergers and acquisitions during the period 1990-2013. In this step, the author has excluded the financial-related firms.

Step 2 Time frame 6,923 cross-border M&As remain for the period 2006-2010.

Step 3 Listing status Following previous studies (e.g. Zhou et al., 2016), in order to obtain sufficient firm-level data, the author only includes the publicly traded acquiring firms. Moreover, the deal is excluded if the target is a subsidiary. This is because a subsidiary can be either a private or public firm, which is difficult to define. After this step, there are 3,947 deals left in the sample.

Step 4 Deal value and stake Deal with a low transaction value and sought stakes may not create a significant effect on the acquiring firms. In line with previous studies, applying the $1 million deal value and 3% sought stake criteria, the sample has 3,932 observations left.

Step 5 Matching with database In this step, the author matches those 3,932 observations with various databases, such as DataStream, World Bank, Hofstede’s national cultural scores, Heritage Foundation. Finally, 1,903 observations are selected, including 1,519 completed deals and 3,84 uncompleted deals.

Following the criteria for sample selection, Table 1 (above) provides detailed steps for screening the final sample of this study.

3.2. Sample characteristics

Table 2 provides an overview of the distribution characteristics of the sample in this study.

Specifically, in terms of region distribution, cross-border M&As take place most frequently in

Western Europe, followed by Asia and North America. With regard to the uncompleted rate,

(21)

Asia ranks first, followed by Oceania. In terms of markets specification, developed market dominates the cross-border M&As in this sample (accounting for 82.92%). With regard to the uncompleted rate, cross-border M&As taking place in emerging markets seem more likely to be cancelled. In terms of time span, the uncompleted rate of cross-border M&As is around 20%

on average, and remained stable through the sample period.

Table 2. Distribution of cross-border M&As

Cross-border M&As Deal completion status

Number Percentage (%) Completed Uncompleted

Regions

Asia 458 24.1 305 153 (33.41%)

North America 384 20.2 336 48 (14.29%)

Western Europe 830 43.6 698 132 (15.90%)

Oceania 144 7.6 108 36 (25%)

Others 87 5.5 72 15 (17.24%)

Markets

Developed market 1578 82.92 1279 299 (18.95%)

Emerging market 325 17.08 240 85 (26.15%)

Time Span

2006 439 23.1 350 89 (20.27%)

2007 504 26.5 400 104 (20.63%)

2008 405 21.3 321 84 (20.74%)

2009 236 12.4 198 38 (16.10%)

2010 319 16.8 250 69 (21.63%)

Total 1903 100 1519 384 (20.18%)

Note: This table presents an overview for distribution of cross-border M&As by region, markets, time frame.

3.3. Research Method

3.3.1. Model for testing the impacts on completion of cross-border M&As T

This model is formulated to answer the research question about the impacts of country-

level factors on completion of cross-border M&As, and the moderating effects of prior

experience. Because the dependent variable here is the dummy variable M&A completion,

which takes 1 if the announced deal is completed and 0 if it is abandoned. Following Dikova et

(22)

_-

represents the explanatory variables and 3 represents the corresponding coefficients. i stands for M&A deals, j stands for country and t stands for the year of announcement of the cross-border M&As. Prob denotes the probability of completion and 1 represents the cumulative distribution function of standard normal distribution (Zhou et al., 2016).

Multicollinearity can be a problem in this mode. A variable will be excluded if it has a high correlation (β>0.5) with another variable.

3.4.1. The model for the completion of cross-border M&As

The dependent variable is the completion or withdrawal of cross-border M&As, which is a dummy variable where value equals one if deals are completed and zero otherwise.

The independent variables include cultural distance (CD) and economic freedom index (EFI). In line with Malhotra et al. (2011), the cultural distance between home and host country is applied using the method developed by Kogut and Singh (1988) with Hofstede’s cultural scores:

%`(`"L( 7+aL,G)

_8,b

=

^(c^d,e_> × h^;c^d,f)^g

d

>8i6

(3)

In this equation, (V

_-8

− V

_-b

) refers to the difference in the score of ith cultural dimensions between country j and k. W

_-

indicates the variance of the difference of ith cultural dimension. %7

_8,b

means the cultural difference between country j (home country) and k (host country). Moreover, in line with Gubbi et al. (2010) and Zhou et al. (2016), this thesis applies economic freedom index developed by Heritage Foundation to measure the formal institutional quality in home country. The economic freedom index measures the ease with which firms conduct their business activities in specific countries.

The moderator is the past completion experience (EXP) of acquirers, which is measured by counting the numbers of previously completed cross-border deals (tracking back to 1990 - the earliest data the author could access).

Control variables include several country-, deal-, and firm-level variables. GDP per capita is used because prior literature suggests that different economic growth may affect the process of cross-border M&As (Malhotra and Gaur, 2014; Kedia and Bilgili, 2015; Liou et al., 2016).

Growth of GDP will drive more cross-border M&As. GDP is measured by the current USD

amount of GDP per capita over one year prior to the deal announcement. Following the

literature (Zhou et al., 2016), geographic distance between two countries in a cross-border

M&A may also affect the process as well. This variable is measured by calculating the natural

logarithm of geographic distance between two capital cities involved in the deal. Moreover, the

stock market capitalization has a positive effect on a firm’s propensity for cross-border M&As

(di Giovanni, 2005). This thesis computes the ratio of stock market capitalization as a

(24)

percentage of a country’s GDP over over year prior to the deal. In terms of firm-level characteristics, following Lim and Lee (2016), this thesis also controls for firm size, leverage ratio, and earning ratio (ROA), because these factors may affect an acquirers’ desire and ability to complete a M&A.

With regard to the deal-specific characteristics, this thesis controls for sought stake, which is the percentage of stake sought by an acquirer in a deal (Zhang et al., 2011). Stake is associated with the interests of shareholders, which may affect the approval of the takeover. Moreover, industry relatedness is also controlled because shareholders’ interests could be damaged if a M&A takes place in an unrelated industry. In addition, cash payment is used because cash payment can facilitate the valuation and increase the likelihood of completion of the M&A (Muehlfeld et al., 2012). Also, the attitude of managers is controlled, because it is difficult to complete a M&A if the managers think it is a hostile deal (Lim and Lee, 2016). Following Dikova et al. (2010), the public status of target may delay the process of a M&A because public firms must comply with stricter regulations. Relative size is controlled because the larger the deal size, the greater the negotiating power of the target firm (Starks and Wei, 2013), which may affect the process of cross-border M&As as well. Lastly, this thesis uses a year dummy, excluding year 2006 to control for fixed year effect. All information regarding the measurement and sources of control variables can be found in Table 3.

3.4.2. The model for an acquirer’s subsequent risk-taking

The Dependent variable is an acquirer’s risk-taking that measures the earnings volatility of a company. In line with John et al. (2008), Boubakri et al. (2013), and Li et al. (2013), corporate risk-taking is measured by calculating the standard deviation of ROA (or R&D if necessary) for each firm with available data for five years.

The independent variables in this model include the completion status (Completion), index

of shareholder protection of acquirer’s home country and scores of cultural dimensions in

acquirers’ home country. The status of completion is a dummy variable where value equals one

(25)

Table 3. Definition of variables

Variables Symbol Measurement Source

Risk-taking Std. ROA

Ratio. The standard deviation of earnings (EBITDA/TOTAL ASSETS) over five years after the completion/withdrawal of cross-border M&As. The higher value of this ratio

the higher earnings volatility.

DataStream

Completion Completion Dummy. One stands for a completed deal and zero otherwise. Thomson Reuters

Cultural distance CD Applying Kogut and Singh’s (1988) method of calculating the cultural distance between acquirer nation and target nation with Hofstede’s four cultural dimensions.

Hofstede (1980, 2001); Geert Hofstede.com

Investor protection ADR Investor protection index ranges from 0-5. Djankov et al. (2008)

Economic freedom

index EFI The economic freedom index that measures the quality of formal institutions. The higher the

index, the better the quality of formal institutions. Heritage Foundation Cultural dimensions UA, IND Two national cultural dimensions from Hofstede. The value ranges from 0-100. Hofstede (1980, 2001); Geert

Hofstede.com Geographic distance GEO Geographic distance between the two capital cities of acquirer and target nations. Geobytes Database

GDP per capita Acq GDP, Tar GDP

Natural logarithm of GDP per capita in current USD of home and target country one year

prior to the deal (Taiwan’s information comes from the Knoema database) World Bank; Knoema database Stock Market

Capitalization SMC The ratio of stock market capitalization of GDP One year prior to the deal. Data from Taiwan

is collected from Taiwan’s Financial Supervisory Commission R.O.C World Bank; Knoema database

Acquirer size SIZE Natural logarithm of acquirer’s total assets prior to deal DataStream

Relative size RLS Ratio of the dollar value of the transaction to the bidder’s market value DataStream Public status of target PUB Dummy variable. One for a public target and zero otherwise Thomson Reuters

Method of payment CASH Dummy variable. One for all-cash payment and zero otherwise Thomson Reuters

Stake sought SOU Ratio. Percentage of stake sought by acquirers Thomson Reuters

Industry relatedness IR Dummy variable. One for the M&A in the same industry and zero otherwise Thomson Reuters

Earnings ROA Ratio. EBITDA/TOTAL assets of acquirer one year prior to the deal DataStream

Attitude FRI Dummy variable. One for friendly M&A, and zero otherwise Thomson Reuters

Leverage LEV Ratio. Total debts to total assets of acquirer One year prior to the deal. DataStream Year and industry Fixed-year,

industry Year and industry fixed effects are applied but not reported in the table. Thomson Reuters

(26)

if a deal is completed, otherwise zero. Investor protection is measured by the anti-director rights (ADR) index established by Djankov et al. (2008). Scores of cultural dimensions of uncertainty avoidance (UA) and individualism (IND) are collected from Hofstede (1980;2001) and Geert Hofstede.com.

Control variables in this model include several country-, and firm-level characteristics.

Following the prior literature (John et al., 2008; Li et al., 2013), this thesis incorporates GDP per capita, stock market capitalization, earnings, leverage, and firm size as control variables.

GDP per capita reflects the economic growth of a country and the national income level. Higher economic growth and income level may boost corporate risk-taking (Li et al., 2013). In order to minimize the influences of country-level financial structure, this thesis also controls for stock market capitalization as percentage of country’s GDP, because different risk-sharing mechanisms may affect firm-level risk-taking as well (John et al., 2008; Li et al., 2013). With regard to firm-level factors, this thesis also controls for earnings (ROA) and firm size.

Comparing to a small company, a large company (a company with high earnings) that own stable operations and resources will have less volatile earnings (John et al., 2008; Li et al., 2013). Leverage ratio also influences the risk-taking because different types of investors may lead to different propensity for risk-taking behaviors, such as cross-border M&As (Chittoor et al., 2015). Control variables used in this model are collected over one year prior to the cross- border M&As. Moreover, the fixed effects of year and industry are included in this thesis.

Detailed information about the measurement and sources of variables can be found in Table 3 (above).

4. Empirical results and analysis 4.1. Descriptive statistics

Table 4 below provides summary statistics. Panel A presents the descriptive statistics of variables used in this thesis. Regarding the main variables in this thesis, firstly, the mean and median of Std. ROA are 5.961% and 3.220%, respectively. Secondly, 79.8% of cross-border M&As are completed. Thirdly, the mean and median of cultural distance are 1.590 and 1.146, respectively. Fourthly, the mean of cultural dimensions of uncertainty avoidance and individualism are 53.887 and 67.717, respectively. Fifthly, the mean 4.651 of experience suggests that the acquirers have completed more than four cross-border M&As on average.

Sixthly, the mean and median of anti-director rights index are 3.904 and 4, respectively. Lastly,

the mean and median of economic freedom index are 73.859 and 78.000, respectively.

(27)

In order to avoid the potential multicollinearity problems, the author run two separate correlation analyses for all variables. Panel B presents the correlation matrix of variables used for cross-border M&A completion model. Specifically, there is high correlation between Acq GDP and EFI (0.692), and between SOU and IR (-0.643). Therefore, Acq GDP and SOU are excluded in this model. Panel C presents the correlation matrix of variables used for the completion of cross-border M&As model. Therefore, this thesis excludes Acq GDP and SOU in the following analysis. Specifically, the correlation between Acq GDP and IND is high (0.537). Hence, this study excludes Acq GDP in the analysis.

Table 4. Summary statistics

Panel A: descriptive statistics of variables used in the analysis

Variable N Mean Median Std. Dev Minimum Maximum

Std. ROA (%) 1903 5.961 3.220 8.475 0.522 56.985

Completion 1903 0.798 1.000 0.402 0.000 1.000

CD 1903 1.590 1.146 1.548 0.027 11.218

UA 1903 53.887 48.000 21.719 8.000 100.000

IND 1903 67.717 74.000 24.212 13.000 91.000

EXP 1903 4.651 2.000 6.973 0.000 50.000

ADR 1903 3.904 4.000 0.975 1.000 5.000

EFI 1903 73.859 78.000 8.918 49.800 90.000

GEO 1903 8.242 8.683 1.109 4.422 9.829

Acq GDP 1903 10.304 10.622 0.920 6.592 11.560

Tar GDP 1903 9.939 10.546 1.200 5.720 11.560

SMC (%) 1903 128.762 121.111 128.943 13.740 1254.470

SIZE 1903 21.073 21.081 2.324 15.271 25.648

RLS 1903 0.158 0.018 0.958 0.00002 24.599

PUB 1903 0.284 0.000 0.451 0.000 1.000

CASH 1903 0.357 0.000 0.479 0.000 1.000

SOU (%) 1903 64.545 75.000 37.177 3.000 100.000

IR 1903 0.546 1.000 0.498 0.000 1.000

ROA (%) 1903 12.095 12.925 13.864 -68.844 41.498

FRI 1903 0.901 1.000 0.300 0.000 1.000

LEV (%) 1903 22.495 21.525 16.326 0.000 68.924

Note: Panel A presents the descriptive statistics of variables used in this thesis, including the model of completion and the model of risk-taking. Acquirer GDP, target GDP =natural logarithm (GDP per capital in US$); acquirer size=natural logarithm (total assets in US$). Std. ROA is the measurement of corporate risk-taking, which is calculated by the standard deviation of (EBITDA/Total assets) over five years after the completion/withdraw of the focal cross-border M&As. Completion is a dummy variable that equals one when the deal is completed, and zero otherwise. CD refers to the cultural distance between the home and host countries in a deal. UA and IND refer to Hofstede’s cultural dimensions: uncertainty avoidance and individualism. EXP is the measurement of acquirer’s prior completed cross-border M&As in absolute numbers. ADR is the revised anti-director-rights index developed by Djankov et al. (2008). EFI is economic freedom index developed by Heritage foundation. GEO is the geographic distance between two capital cities of acquirers’ home and host country. SMC refers to the stock market capitalization as percentage of country’s GDP. RLS refers to the relative size of transaction value divided by acquirer’s market capitalization. PUB is a dummy variable that equals one if the target is a publicly-traded firm.

CASH is a dummy variable that equals

one

when the deal is paid all by cash. SOU refers to the stake sought by the acquirer. IR is the industry relatedness measured by SIC two-digit code of acquirer and target firms. ROA is the earning ratio that calculated by EBITDA/Total assets one year before the focal deal. FRI is a dummy that equals one when the M&A is friendly. LEV refers to the leverage calculated by total debts/total assets one year prior to the deal. All firm-level variables are winsorized at 1% for both sides.

(28)

Panel B: Correlation matrix (1)

COM CD EFI EXP GEO Acq

GDP

Tar GDP SMC SIZE RLS PUB CASH SOU IR ROA FRI LEV

COM 1.000

CD -0.064 1.000

EFI -0.012 -0.206 1.000

EXP 0.048 0.076 0.046 1.000

GEO -0.029 0.074 0.059 0.018 1.000

AcqGDP 0.056 -0.054 0.694 0.208 -0.086 1.000

TarGDP 0.168 -0.216 0.020 -0.019 -0.067 0.035 1.000

SMC -0.169 -0.130 0.399 -0.083 -0.011 0.065 -0.131 1.000

SIZE 0.083 0.116 -0.134 0.483 0.026 0.127 -0.069 -0.131 1.000

RLS -0.103 -0.032 0.056 -0.086 -0.060 -0.030 -0.041 0.093 -0.220 1.000

PUB -0.014 0.061 -0.127 0.049 -0.024 -0.052 -0.034 0.001 0.329 0.018 1.000

Cash 0.077 0.038 0.036 -0.019 -0.010 0.010 0.049 0.030 0.014 -0.040 0.115 1.000

SOU 0.174 -0.184 0.188 -0.038 0.002 0.122 0.216 -0.033 -0.273 0.005 -0.645 -0.038 1.000

IR 0.010 -0.026 0.027 -0.029 -0.002 0.046 -0.070 -0.072 0.011 -0.051 0.005 0.006 0.047 1.000

ROA 0.095 -0.005 -0.114 0.075 -0.008 -0.067 0.036 -0.102 0.277 -0.262 0.065 0.036 0.005 0.072 1.000

FRI 0.043 -0.029 0.089 -0.021 0.045 0.063 0.088 -0.044 -0.128 0.006 -0.223 -0.057 0.289 0.029 -0.008 1.000

LEV 0.033 0.099 -0.154 0.229 -0.027 -0.113 -0.024 -0.075 0.281 -0.001 0.090 -0.070 -0.121 -0.095 0.050 -0.045 1.000 Panel C: Correlation matrix (2)

Std.(ROA) Completion UA IND ADR Acq GDP SMC SIZE ROA LEV

Std. ROA 1.000

Completion -0.066 1.000

UA -0.096 -0.015 1.000

IND 0.060 0.171 -0.296 1.000

ADR 0.024 -0.026 -0.227 -0.082 1.000

Acq GDP 0.003 0.056 -0.013 0.537 -0.096 1.000

SMC 0.036 -0.169 -0.303 -0.203 0.209 0.038 1.000

SIZE -0.344 0.083 0.345 -0.032 -0.154 0.109 -0.131 1.000

ROA -0.290 0.095 0.050 0.013 -0.037 -0.076 -0.102 0.277 1.000

LEV -0.147 0.033 0.115 -0.125 0.032 -0.117 -0.075 0.280 0.050 1.000

Note: Panel B and Panel C present the correlation matrix of variables used in the model of completion and the model of risk-taking, respectively.