Target firm’s earnings quality and acquisition
performance: an international perspective
Bram Fiselier
S2226669
Supervisor: Dr. S. Wang
Second assessor: Dr. T. A. Marra
Word count: 13.158
November 2017
Master’s thesis
University of Groningen
Faculty of Economics and Business
Abstract
An extensive body of research investigated the performance implications of international acquisitions and what triggers this performance. This study attempts to contribute to this field of research by focusing on the role of earnings quality in international acquisition performance. This performance is examined using the event-study methodology. Multivariate clustered robust regressions are conducted to investigate the role of several variables on the acquisition performance of 735 firms that engaged in a cross-border acquisition between 2000 and 2016. The results show no initial effect of differences in accounting systems, target country’s enforcement system strength, and target firm’s earnings quality on international acquisition performance. However, evidence is found for an interaction effect of earnings quality and accounting distance. That is, when differences in accounting standards increase, acquirers fail to recognize low earnings quality and therefore, fail to discount their bids. This negatively affects international acquisition performance. Furthermore, the results provide evidence that accounting system differences do not negatively affect international acquisition performance if the earnings quality of the target is high.
JEL-Classifications: G14, G34, M41
Keywords: international acquisitions, accounting distance, enforcement systems, earnings quality
1. Introduction ...1
2. Literature review and hypothesis development ...4
2.1 Mergers and acquisitions ...4
2.2 International acquisitions ...5
2.3 Acquisition performance ...7
2.4 Accounting standards ...7
2.5 Earnings quality ... 10
3. Data and Methodology ... 13
3.1 Data Collection ... 13 3.2 Variables ... 14 3.2.1 Dependent variable. ... 14 3.2.2 Independent variables ... 18 3.2.3 Control Variables ... 20 3.3 Statistical methodology ... 25 4. Results ... 27
4.1 Multivariate clustered regression ... 28
4.2 Additional analysis ... 29
4.2.1 Subsample analysis ... 29
4.2.2 Different earnings quality measure ... 31
4.3 Robustness tests ... 33
4.3.1 Different event window ... 33
4.3.2 Constant mean return model. ... 35
5. Discussion ... 35
5.1 Accounting distance and enforcement strength ... 35
5.2 Earnings quality ... 36
6. Conclusion ... 39
6.1 Summary ... 39
6.2 Implications for practice ... 39
6.3 Limitations ... 40
Table of contents
1
1. Introduction
The popularity of acquisitions increased heavily in recent decades. Although this is mostly attributable to domestic acquisitions, the importance of cross-border acquisitions is rising. The share of international acquisitions doubled to almost fifty percent in the first decade of this century. In 2014, the total value of all acquisitions crossing a nation’s border was $3.4 trillion (Raice, 2015). For comparison, only the United States, Germany, China and Japan had a GDP higher than the global value of cross-border acquisitions in that year.
The increased popularity of international acquisitions resulted in an extensive body of research focusing on the performance implications of this type of acquisition and what triggers this performance. According to Anand et al. (2005), cross-border acquisitions are more likely to create value than domestic acquisitions due to greater access to geographically distributed resources embedded in target firms. Some studies empirically support this notion by providing evidence of increased performance after international acquisitions (Chakrabarti et al., 2009; Goergen and Renneboog, 2004; Gubbi et al., 2010). However, Datta and Puia (1995) present negative performance effects for these acquisitions. So, although international acquisitions seem to provide more opportunities than domestic ones, these acquisitions are not always performance enhancing. This study attempts to contribute to this extensive body of research by addressing the following research question: does target firm’s earnings quality affect international acquisition performance?
International acquisitions are different than their domestic counterparts. For example, firms need to tackle the issues related to differences in accounting standards and practices when valuing foreign targets (Angwin, 2001). These differences could, for instance, complicate the already complex process of assigning a value to intangible assets. Furthermore, critical environmental conditions, like governmental regulations applicable to the target, should be evaluated to accurately value a foreign target (Hitt and Pisano, 2003).
According to Straub (2007), the value of an acquisition is equal to the combined value of the firms after the acquisition minus the combined value of the firms before the acquisition, the acquisition premium, and other process related costs. This equation suggests that the higher the acquisition premium, the lower the acquisition value. Literature empirically supports this notion (Hayward and Hambrick, 1997; Malmendier and Tate, 2008). Thus, to create value in an acquisition, an acquirer should accurately value and not overpay for the target.
2 In a regular acquisition, the target sets a reservation price and accepts any bid higher than this price (Hansen, 1987). The more accurately the acquirer can value the target, the more able it is to bid to the reservation price closely (McNichols and Stubben, 2015). One of the factors that could affect the estimation accuracy in this process is the accounting quality of the target firm. The higher the accounting quality of a company, the more decision-useful this information is for outsiders, and thus, the better outsiders can estimate the fundamental value of a firm (Dechow et al., 2010). Evidence for this positive relationship between the target’s accounting quality and valuation accuracy and therefore, acquisition performance, is already provided by McNichols and Stubben (2015). However, their sample consists of domestic acquisitions only.
Besides the potential positive effect, an adverse effect of earnings quality is also possible. The reasoning behind is that firms with low earnings quality have a higher cost of capital. This higher cost of capital is reflected in a company’s share price, resulting in a built-in discount. This built-in discount lowers the likelihood of an acquirer overpaying for the target (Black et al., 2007). Thus, in this case, lower earnings quality leads to higher acquisition performance and vice-versa.
As stated above, international acquisitions are more complex than domestic ones. This complexity could make it more difficult for acquiring firms to value a foreign target appropriately. For example, acquiring companies tend to view earnings and other financial data from a home country perspective and do not take the effects of accounting differences into consideration (Radebaugh and Gray, 1997). If the acquiring firms do not have sufficient knowledge on the accounting standards and usual accounting practices of the country the target firms are headquartered in, they might be unable to judge the quality of the targets' earnings accurately. This could lead to an inaccurate valuation of the target. The less the management of an acquiring firm is able to value the target appropriately, the more likely it is that they overpay for the target and the more likely it is that the acquisition will be value destroying. Therefore, high target firm’s earnings quality could be even more important in international acquisitions characterized by a high level of accounting system differences.
The way this study distinguishes itself from the previous research on the effect of earnings quality on acquisition performance by McNichols and Stubben (2015) is by considering a cross-border effect. Black et al. (2007) already investigated the role of earnings quality in international acquisitions. However, they determined earnings quality on a country-level. Gaio (2010) showed that country-level data only explains a small portion of level earnings quality and that
firm-3 level data has significant incremental explanatory power over country-level variables in determining earnings quality. Therefore, this study attempts to contribute to the academic literature by investigating the role of earnings quality in international acquisitions using firm-level data in determining a target firm’s earnings quality.
This study hypothesizes both positive and negative effects of target firm’s earnings quality on international acquisition performance. Furthermore, value-destroying effects of accounting system differences and value-enhancing effects of target countries’ enforcement system strength are hypothesized. To test the performance implications of these variables in international acquisitions, an event-study is conducted, as is common in studies focusing on the performance implications of acquisitions (Datta and Puia, 1995; Li et al., 2016). The three-day market-adjusted stock return of the acquirer centered on the date of the acquisition is used to determine the economic benefits of the acquisition. A multivariate clustered robust regression is conducted to investigate the relationship between the several variables and international acquisition performance of the 783 acquirers1 in the sample. Initially, the results show no effect of differences in accounting systems, target country’s enforcement system strength, and target firm’s earnings quality on international acquisition performance. However, evidence is found for an interaction effect of earnings quality and accounting distance. That is, when differences in accounting standards increase, acquirers seem unable to recognize low earnings quality and therefore, fail to discount their bids, resulting in lower international acquisition performance. Furthermore, the results provide evidence that accounting system differences do not negatively affect international acquisition performance if the earnings quality of the target is high.
The structure of this study is as follows. Section 2 presents an analysis of the relevant literature and provides the hypotheses of this study. Section 3 discusses the data collection, variables, and the methodology. Section 4 provides an overview of the regression outcomes, additional analyses, and robustness tests. Section 5 provides a discussion of the results. Finally, the conclusion is provided in section 6.
4
2. Literature review and hypothesis development
This section elaborates on the relevant literature related to cross-border acquisitions and earnings quality. The section begins with a discussion on acquisitions and the way they can create value, followed by an overview of the challenges and opportunities related to international acquisitions. Lastly, the role of earnings quality in this type of acquisition will be discussed.
2.1 Mergers and acquisitions
Mergers and acquisitions (M&As) are both ways for a company to expand. However, some major differences exist. After two companies merge, one company ceases to exist as the two companies are combined into one entity. This does not hold for acquisitions. In an acquisition, one firm obtains a controlling interest in another firm, meaning that over fifty percent of the voting rights are acquired. This could be achieved through the purchase of assets or shares by the acquirer. In this case, the target company remains to exist as a subsidiary of the acquiring firm (DePamphilis, 2009).
Firms engage in acquisitions because they believe synergies can be captured and value can be created. Synergies are present when the combined value of the two entities is higher than the sum of values of the individual companies prior to the acquisition. In this case, the value added through an acquisition is determined by subtracting the value of both firms before the acquisition, the acquisition premium, and the expenses made in the process from the combined post-acquisition value of both firms (Straub, 2007). Thus, if the value gained through synergies is higher than the premium paid and other costs made in the acquisition process, an acquisition is value enhancing. In their review of the M&A literature, Haleblian et al. (2009) identified several ways through which acquisitions could provide synergies. Firstly, synergies occur through higher market power for the newly combined firm. By combining two firms, the number of firms in the industry will decline, resulting in increased pricing power for the remaining firms. Since the newly combined firm is bigger than the individual firms before the acquisition, and the number of firms in the industry is lower after the acquisition, it is easier for the newly combined firm to set its prices at a higher level. Furthermore, the increased size enables the new firm to demand lower prices from its suppliers due to an enhanced bargaining position. Previous literature empirically supports the increased market power hypothesis (Prager, 1992; Kim and Singal, 1993). Secondly, the newly combined firms could benefit from economies of scale. As production volume of the
5 newly combined firm increases, marginal costs decrease. In this case, companies could adopt different and more cost-efficient production methods. Previous literature empirically supports this theoretical notion (McGucking and Nguyen, 1995; Banker et al., 2003). Besides, overlapping departments can be merged after two firms are combined, resulting in savings on salaries (Straub, 2007). Thirdly, economies of scope could arise through competency transfers and redeployment of assets. A company’s product line might be bigger after an acquisition. If this is the case, intangible assets like reputation, product names, customer bases, and distribution channels can be used for multiple products. This enables the firm to enter new markets more easily and combine strategies, resulting in cost savings (Capron et al., 1998; King et al., 2008). Lastly, incompetent managers being disciplined by the market could create acquisition value (Jensen, 1986; Jensen and Ruback, 1983). This theoretical notion is supported by a study of Agrawal and Walkling (1994). They show that CEOs of acquired companies often get dismissed after the acquisition.
2.2 International acquisitions
The previously explained ways through which companies could achieve synergies apply to acquisitions in general. International acquisitions, however, provide additional opportunities in comparison to domestic ones, but firms also have to cope with certain challenges.
The first opportunities provided by international acquisitions are the extension of a company’s current product market to other countries and the access to lucrative foreign markets. Because of the possible entry barriers, entering markets abroad is often difficult for firms. Companies lack relationships with distributors and suppliers in foreign countries. Furthermore, regulations and legislation discouraging foreign firms to enter a certain nation could hinder international expansion (Hitt and Pisano, 2003). By acquiring a foreign company, a firm could overcome these barriers and use international acquisitions as a way to enter an otherwise difficult-to-enter market. By entering these overseas markets, a company could benefit from economies of scale and expand their current product market. Thus, international acquisitions could enable growth and enhance profitability for firms.
Secondly, firms could obtain new knowledge and capabilities by acquiring foreign companies (Barkema and Vermeulen, 1998; Very and Schweiger, 2001). These opportunities arise because of differences in corporate and societal cultures across countries. Acquiring firms could exchange knowledge of operational methods with the target and learn from their managerial practices and capabilities and vice-versa (Javidan et al., 2005). Furthermore, firms could use
6 international acquisitions to get access to specific strategic resources that are not available domestically. Markets are becoming increasingly global and competitive. Therefore, to successfully compete and obtain a sustainable competitive advantage, companies need valuable and unique resources (Barney, 1991). By expanding their focus to foreign markets, the availability and variety of new and valuable resources will be higher. Thus, by combining a firm’s resources with resources acquired abroad, a firm could create a unique and valuable resource base, needed for a competitive advantage (Makino et al., 2002).
Finally, foreign acquisitions could create value through increased possibilities of risk diversification. Operating internationally allows for risk spreading over multiple markets and countries. In this case, companies are less reliant on the market of one single nation and fluctuating income streams caused by different economic conditions worldwide will offset each other.
Besides opportunities, international acquisitions also come with specific challenges. A significant challenge firms should overcome is the differences in culture between the acquirer’s country and the target’s country. According to Hofstede et al., (2010, p.6), “Culture consists of the unwritten rules of the social game. It is the collective programming of the mind that distinguishes the members of one group or category of people from others”. Hofstede et al. (2010) defined culture using six dimensions: individualism, power distance, masculinity, uncertainty avoidance, long-term orientation, and indulgence. The more countries differ on these dimensions, the bigger the differences in culture are. The degree to which countries are culturally different is defined as the cultural distance between two countries. Previous research highlighted the important role of cultural distance in international business, and thus, international acquisitions (Hofstede et al., 2010; Trompenaars and Hampden-Turner, 2011; House et al., 2004). Firstly, cultural distance could hinder the integration process, which is crucial for acquisition success (Gates and Very, 2003; Lajoux, 2006; Lemieux and Banks, 2007). Secondly, different leadership styles between countries and cultures could hinder successful integration and thus negatively affect acquisition performance (Van de Vliert, 2006; Datta, 1991). Thirdly, information sharing is hindered because culturally different executives communicate inefficiently (Sales and Mirvis, 1984; De Long and Fahey, 2000; Lin and Germain, 1998). Other challenges arising in international acquisitions are poor communication and lower commitment of the employees of the acquired firm (Very et al., 1996), misunderstanding about assignments because of limited understanding between groups that differ culturally (Heiman et al., 2008), and the lack of relationships and knowledge about local
7 markets, needed to succeed abroad (Zaheer and Mosakowski, 1997; Hitt and Pisano, 2003). Lastly, the differences in accounting standards between countries could hinder the valuation process in acquisitions. The issues resulting from these differences will be explained in section 2.4.
2.3 Acquisition performance
Although firms engage in acquisitions to capture value through potential synergies, research has shown that in most cases domestic acquisitions are value-diminishing. Some studies find value enhancing effects of acquisitions (Wright et al., 2002; Healy et al., 1992). However, the majority of studies show value-destroying effects (King et al., 2004; Agrawal et al., 1992). Cross-border acquisitions seem to differ from domestic ones in this aspect. Although an old study of Datta and Puia (1995) reports value-diminishing effects, more recent studies find value enhancing effects of international acquisitions (Goergen and Renneboog, 2004; Chakrabarti et al., 2009; Gubbi et al., 2010). A logical explanation for this might be the increased level of globalization and technical developments that take away some of the barriers of operating abroad. So it seems that in most cases, the opportunities offered by international acquisitions outweigh the challenges. Another explanation for scholars finding more value enhancing effects for international acquisitions compared to domestic ones is better due diligence in international acquisitions. Companies are aware of the potential difficulties coming with these acquisitions and deals will only materialize when they have substantial economic potential. Aguilera et al. (2004) provide empirical support for this claim by showing that acquisition announcements are more often withdrawn as the level of cultural differences between acquirer and target increases.
So, the different conclusions of the studies on acquisition performance show that an acquisition is a complex event and the studies do not provide prerequisites for estimating acquisition performance. However, research has shown that international acquisitions tend to perform better than domestic ones. This study attempts to contribute to this research by considering the role of firm-level earnings quality and accounting system differences in international acquisition performance.
2.4 Accounting standards
The most interesting challenge raised by international acquisitions regarding this study is the challenge emerging from the different accounting standards and practices between cultures and countries. A study of Gray (1988), was one of the first studies linking differences in national culture to variation in accounting systems. In his paper, he identifies several accounting values
8 useful in evaluating a country’s accounting system (professionalism, uniformity, conservatism, and secrecy) and links these values to the dimensions of national culture as identified by Hofstede (1983). Gray suggests a close link between the cultural dimensions and the patterns of accounting systems. This implies that every culture should develop its own accounting systems based on its own unique requirements. Chanchani and Willet (2004) operationalize the constructs developed by Gray (1988) and try to determine the empirical validity of them. Their results provide some support for the usefulness of the accounting values as empirical constructs, although they may require some reinterpretation and adaptation. Nevertheless, there seems to be a connection between the variation in national cultures and the variation in accounting systems.
Although Gray (1988) suggests that every country should develop their own accounting systems tailored to their own requirements, accounting systems worldwide are becoming increasingly convergent (Land and Lang, 2002). The best example of this is the widespread adoption of the International Financial Reporting Standards (IFRS). Since 2005, the European Parliament requires companies listed on stock exchanges based in the European Union to prepare their consolidated financial statements in accordance with the IFRS (ICAEW, 2016). Besides the EU countries, other countries started adopting the IFRS as well. In 2016, 89 percent of the nations with a stock exchange required or permitted the adoption of IFRS for their listed companies (PwC, 2017). While the IFRS is becoming increasingly popular across the globe, firms publicly listed in the United States are still required to prepare their financial statements in accordance with the US GAAP. However, the American FASB is working with the regulatory body responsible for the IFRS, the IASB, to reduce or eliminate the differences between the IFRS and US GAAP (FASB, 2017). The reasoning behind this convergence is that by providing participants in the worldwide capital market with a similar set of high-quality accounting standards, comparability of financial statements worldwide will be enhanced, which helps these participants in making economic decisions (IASC Foundation, 2005).
However, convergence does not necessarily lead to a higher comparability. According to Sunder (2011), the argument that the IFRS results in global comparability of financial statements are pretentious. He states that the adoption of a worldwide standard disallows the tailoring of financial reporting to country-specific variations in economic, business, legal, auditing, commercial, regulatory, and governance conditions, and therefore, fails to capture the underlying reality. Some empirical support for this statement is provided by Cascino and Gassen (2015). They
9 state that while IFRS adoption increases the comparability of balance sheet items across countries, it does not clearly affect the cross-country comparability of earnings attributes. The outcome of the reporting process is still shaped by country- and firm-level incentives, which limits the cross-sectional comparability of financial accounting information. Especially the quality of enforcement on country-level and the incentives to comply with accounting standards on firm-level affect the degree to which the adoption of the IFRS results in higher cross-country comparability (Cascino and Gassen, 2015).
The variation in accounting systems and standards is one of the difficulties firms should overcome in cross-border acquisitions. Acquirers need to interpret the GAAP of the target firm’s country. And even if both countries use the same set of international standards, differences in the institutional environments of both countries could hinder comparability. This often leads to problems. For example, acquiring firms tend to view earnings and other financial data from a home country perspective and do not take the effects of accounting differences into consideration (Radebaugh and Gray, 1997). This could negatively affect the accurate valuation of a foreign target. In a regular acquisition setting, a target sets a reservation price and accepts every bid by a bidding firm, equal to or higher than that price. Since the bidder has imperfect information about the target, it is hard for the bidder to value the target accurately (Hansen, 1987). As mentioned before, an accurate valuation of the target is essential. If an acquirer does not estimate the value of a potential target accurately, it might pay a premium that is too high, which results in lower acquisition performance. Scholars provide empirical evidence for adverse effects of acquisition premiums on acquisition performance (Hayward and Hambrick, 1997; Malmendier and Tate, 2008). The valuation process is complex in domestic acquisitions (Hitt et al., 2001), but even more complicated in international acquisitions (Angwin, 2001). So, if an acquirer is not fully aware of the accounting systems of the target firm’s country, or the institutional environment affecting a country’s accounting systems, it is likely that the acquirer’s estimate of a target’s reservation price will not be as accurate as the estimate in a domestic acquisition process would have been. Thus, the more significant the differences in accounting practices between the acquirer and the target country, the harder it is for the acquirer to value the foreign target accurately, and the lower acquisition performance will be. This leads to the first hypothesis:
H1a: Differences in accounting systems between the acquiring firm’s country and the target firm’s country have a negative effect on international acquisition performance
10 As mentioned before, an increasing number of countries are adopting the IFRS. Having similar standards in both the acquiring firm’s country and the target firm’s country could ease the valuation process for the acquirer. However, having the same set of accounting standards in both countries does not necessarily mean that the comparability of the financial statements of both companies is high. A factor influencing the comparability is the degree to which companies comply with the IFRS. The degree to which companies comply is dependent on the strength of the enforcement systems in a country (Cascino and Gassen, 2015). Even if both the acquirer’s nation and the target’s nation do not share the same set of accounting standards, strong enforcement might still enhance international acquisition performance. The stronger the investor protection, and thus enforcement in a country, the more decision-useful the accounting information of firms in that country (Siekkinen, 2016; Ali and Hwang, 2000). This enables the acquirer to estimate the reservation price of the target more accurately. Therefore, the strength of the enforcement system in the target country is likely to have a positive effect on acquisition performance, which results in the following hypothesis:
H1b: There is a positive relationship between the strength of the enforcement systems in the target country and international acquisition performance
2.5 Earnings quality
Another factor that could affect the acquisition-performance relationship is the earnings quality of the target firm. Dechow et al. (2010) provide a definition of earnings quality: ‘Higher quality earnings provide more information about the features of a firm’s financial performance that are relevant to a specific decision made by a specific decision-maker’ (p.1). High-quality earnings are said to be more relevant to the users of this information (Scott, 2015). Previous research shows that accounting information helps in explaining equity prices (Ball and Brown, 1968; Francis et al., 2005), and that several measures of earnings quality have a positive effect on this explanatory power of the accounting information (Kormendi and Lipe, 1987; Li, 2011; Francis et al., 2005; Ecker et al., 2006). Thus, the higher the earnings quality, the higher the relevance of the accounting information.
In the case of an acquisition, the value relevance of accounting information depends on the usefulness of this information in estimation the value of the target company. In line with previous literature, the higher the earnings quality of a target company, the more useful it will be for the acquirer. When the management of a firm realize that they are a potential acquisition target, they
11 might have incentives to manipulate accounting information. However, a study of Erickson and Wang (1999) does not find evidence of discretionary reporting behavior by target firms before acquisitions. They state that targets do not have enough time to manipulate earnings prior to an acquisition. Another important assumption in this study is that acquiring firms select acquisition targets for business reasons and not for the quality of a target’s accounting information. By including target’s accounting quality in the acquisition prediction model of Palepu (1986), McNichols and Stubben (2015) indeed find that the accounting quality of target firms does not significantly increase the likelihood of a firm being acquired.
In an acquisition process, an acquirer should identify the intrinsic value of a target. The value an acquisition will provide to the acquirer is equal to the intrinsic value of the target firm, plus the synergies produced in the acquisition, minus the price paid for the target firm. So, the acquiring company is willing to pay as much as the intrinsic value of the target, plus the potential synergies. Assuming that the target firm knows its intrinsic value with certainty, the target will face a certain trade-off in setting the reservation price. By setting a high reservation price, more acquisition rents could be extracted, but it increases the risk of not selling if the potential synergies are too low. The value created for the acquirer depends on how accurately the acquirer can estimate the target’s reservation price. If the acquirer knows this price with certainty, it can bid very closely to this price. If the acquirer cannot accurately estimate this price, the uncertainty about the target’s value and reservation price will be higher. This leads to more varying bids and the bids accepted by the target will be higher. Therefore, acquirers pay more for a target, and create less value, if uncertainty is high (McNichols and Stubben, 2015). Even though acquirers should discount a bid in the case of high uncertainty, research has shown that this does not always happen in practice (Andrade et al., 2001; Moeller et al., 2005).
If the acquirers use target firms’ accounting information in estimating the value of the target, high target’s earnings quality will benefit the acquirer. Higher quality accounting information is more relevant to the acquirer since it enables the acquirer to estimate the value of the target accurately. This enables the acquirer to place a bid close to the target’s reservation price, which lowers the chances of the acquirer overpaying for the target and increases the likelihood of a successful acquisition, leading to the following hypothesis:
H2a: There is a positive relationship between the quality of a target’s earnings and international acquisition performance
12 However, it is also possible that high earnings quality is related to low international acquisition performance. Even though high earnings quality might help the acquirer in accurately estimating the value of the target, it also drives up the value of the target company. As said before, an acquirer overpaying for a target is one of the main reasons for acquisitions being value destroying. When earnings quality and value relevance of accounting information of the target company are low, the cost of capital of the target is high because of the higher information asymmetry. This leads to a lower firm value (Gaio and Raposo, 2011). The high cost of capital is reflected in the target company’s share price. Therefore, firms with lower earnings quality have a built-in discount, resulting in a lower likeliness of overpaying by the acquirer. This has positive implications for acquisition performance. Furthermore, after the acquisition, the target’s assets will be under the supervision of the acquirer. This could enable the acquirer to value the target’s assets more accurately. This decrease in information asymmetry will lead to a lower cost of capital for the target firm, and ultimately, remove the discount associated with the lower earnings quality, and increase the value of the target firm, and thus, the value of the acquisition (Black et al., 2007). Thus, target’s earnings quality could be negatively related to international accounting performance as well. This leads to the following hypothesis:
H2b: There is a negative relationship between the quality of a target’s earnings and international acquisition performance
As said before, the possible differences in accounting standards and practices are one of the challenges to overcome in international acquisitions. Even though differences in accounting systems are linked to the cultural differences between countries (Gray, 1988; Chanchani and Willet, 2004), accounting standards worldwide are becoming increasingly convergent (Land and Lang, 2002), apparent by the worldwide adoption of the IFRS. IFRS adoption, however, might not necessarily lead to a higher cross-country comparability of accounting information (Cascino and Gassen, 2015). Furthermore, acquiring firms tend to view earnings and other financial data from a home country perspective and do not take the effects of accounting differences into considerat ion (Radebaugh and Gray, 1997). Therefore, it is even more important that the quality, and value relevance, of accounting information is high in cross-border acquisitions since it is likely that acquiring firms will be less able to judge the quality of the earnings of foreign targets, especially when the target firm prepare their financial statements in accordance with a different accounting standard. This will complicate the process of estimating the value of a foreign target and could
13 lead to inaccurate valuations and therefore, low acquisition performance.
An important assumption when testing this relationship is that the comparison of earnings quality of firms from different countries with different accounting standards is meaningful. One could argue that earnings quality is dependent on the accounting standards of the various countries. However, as said before, accounting standards are becoming increasingly convergent (Land and Lang, 2002). This could make the comparison more meaningful. Additionally, a study by Gaio (2010) on the cross-country comparability of earnings quality measures shows that firm-level variables explain the majority of the variation in earnings quality worldwide. These firm-level variables explain around 32.7 percent of this variation while country-level variables only explain 8.5 percent. Furthermore, the variables related to differences in accounting systems are found to be insignificant in explaining this variation. Thus, it seems that cross-country comparison on firm-level earnings quality measures is meaningful. Therefore, the final hypothesis of this study is:
H3: As differences between acquirer’s country GAAP and target’s country GAAP increases, the role of earnings quality in international acquisition performance increases
3. Data and Methodology
This section presents an overview of the data and methodology employed in this thesis. Firstly, an overview of the data collection and data sources will be provided. Secondly, the independent, control and dependent variables will be explained. Lastly, a description of the methodology will be provided.
3.1 Data Collection
Data on acquisitions is gathered from the Thomson ONE database, developed by Thomson Reuters. The sample criteria are based on Moeller et al. (2004), with the exception that this study focuses on international acquisitions only. Only acquisitions in which the acquirer obtains a controlling interest (that is, more than 50 percent of the target’s shares) in the target company are considered. Furthermore, acquisitions that are not completed within 1000 days after the announcement, have a deal value lower than 1 million USD, and observations with missing data or overlapping event periods are excluded. Target firms headquartered in so-called tax havens (e.g., Bermuda) are also excluded from the sample since, in general, companies expand to these countries for tax purposes and not for operational purposes. These requirements yield a sample of
14 783 usable acquisitions2, including acquirers and targets worldwide, conducted between 2000 and 2016. Table 1 includes a matrix presenting all the acquiring countries and target countries included in the sample.
Acquirer returns and market returns are obtained from Thomson’s Datastream, firm-level data is extracted from Thomson’s Worldscope and the Compustat database, corporate governance data is gathered from BoardEx, country-level data, like GDP per capita, import- and export numbers, are obtained from the Worldbank database, data on bilateral trade flows is gathered from the United Nations’ Comtrade database, the distances between the capital cities of the acquiring firm’s and the target firm’s country is obtained from Kristian Gleditsch’s ‘Distance between Capital Cities Data’ dataset, and data necessary to calculate the cultural distance between acquiring firm’s and target firm’s country is gathered from the Hofstede et al. (2010) dataset.
3.2 Variables
3.2.1 Dependent variable. The dependent variable of this study, acquisition performance,
is calculated using the short-term event-study methodology. The abnormal share returns of a company surrounding the announcement of a specific event are measured to determine the performance implications of this event. The event-study is a common way to measure international acquisition performance (Datta and Puia, 1995; Li et al., 2016). Within the event-study methodology, several ways exist to calculate the abnormal returns. Two widely used methods are the market model and the market-adjusted abnormal returns model (Brown and Warner, 1985; MacKinlay, 1997). The market-adjusted returns model is a restricted market model, often used in studies with low data-availability, for example, in the case of initial public offerings. Since stock data is widely available for the sample
firms, this restriction does not apply to this study. Therefore, this study adopts the market model as developed MacKinlay (1997). This methodology calculates the abnormal returns as follows:
𝐴𝑅𝑖,𝑡 = 𝑅𝑖,𝑡 − 𝐸(𝑅𝑖,𝑡) (1)
Where ARi,t is the abnormal stock return of acquiring firm i at time t, Ri,t denotes the actual
stock return of acquiring firm i at time t, and E(Ri,t) is the expected normal stock return of acquiring
firm i at time t.
2 Because of the extensive data requirements for calculating earnings quality at firm-level, the sample size is reduced
15 Ta b le 1 S a m p le b re a k d o w n b y a c q u ir er - a n d t a rg e t c o u n tr y T a rg et c o u n tr y A c q u ire r co u n tr y AUS A U T BE L BR A CA N CH L CZ E DNK E G Y F IN F R A D E U G RC H K G H U N IN D ID N IRL IS R IT A JP N M Y S A rg en ti n a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 A u st ra li a 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 A u st ri a 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 Be lg iu m 0 0 0 0 2 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 Bra zi l 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ca n a d a 6 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 Ch in a 1 0 0 0 1 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 D e n m a rk 1 0 0 1 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 F in la n d 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 F ra n c e 0 1 0 1 5 0 0 0 0 0 0 3 2 0 0 0 0 0 1 2 1 0 G er m a n y 2 1 0 0 1 0 0 0 0 1 3 0 0 0 1 0 0 0 0 2 0 1 H o n g K o n g 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 In d ia 0 0 1 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 In d o n e si a 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ire la n d 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 Is ra el 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 It a ly 0 0 1 0 1 0 0 0 0 0 1 2 0 0 0 0 0 0 0 0 0 0 Ja p a n 5 0 0 0 0 0 0 0 0 0 2 0 0 1 0 0 0 0 0 0 0 1 L u x e m b o u rg 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M a la y si a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 M e x ic o 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N et h erl a n d s 0 0 2 0 2 0 0 0 0 0 1 1 0 0 0 0 0 0 0 1 0 0 N e w Z ea la n d 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N o rw a y 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 P eru 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 P h il ip p in e s 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 P o la n d 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 P o rt u g a l 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ru ss ia 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S in g a p o re 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 2 0 0 0 0 2 S o u th A fr ic a 3 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 S p a in 0 0 0 1 2 1 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 S w e d e n 1 0 0 0 1 1 0 3 1 1 1 0 0 0 0 0 0 0 0 0 0 0 S w it z erl a n d 3 0 0 0 3 0 0 0 0 0 3 5 0 0 0 0 1 0 1 1 0 0 T a iw a n 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 T h a il a n d 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 U n it e d K in g d o m 9 0 0 1 7 0 0 3 0 0 4 1 1 0 0 0 1 2 0 0 0 0 U n it e d S ta te s 9 0 0 1 67 3 1 2 1 1 8 12 0 1 0 1 0 1 12 1 2 0 T o ta l 44 2 4 7 105 5 3 9 2 4 29 29 3 6 1 2 6 3 15 7 3 5
16 Ta b le 1 ( Co n ti n u e d ) T a rg et c o u n tr y A c q u ire r co u n tr y M E X N L D N Z L NOR P A K P E R P H L P O L P RT S G P Z A F K O R E S P S W E CH E T W N T H A T U R G BR U S A T o ta l A rg en ti n a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 A u st ra li a 0 0 3 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 3 2 14 A u st ri a 0 2 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 5 Be lg iu m 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 4 4 15 Bra zi l 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 Ca n a d a 0 0 0 0 0 3 0 0 0 1 0 0 0 1 0 0 0 0 8 52 73 Ch in a 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 D e n m a rk 0 2 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 2 10 F in la n d 0 0 0 2 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 5 11 F ra n c e 0 4 0 0 0 0 0 1 0 0 0 0 0 1 1 0 0 0 8 29 60 G er m a n y 0 0 0 2 0 0 0 0 0 0 0 0 1 0 4 1 0 1 7 21 49 H o n g K o n g 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 3 3 8 In d ia 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 3 7 In d o n e si a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 Ire la n d 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 5 10 Is ra el 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 8 It a ly 0 2 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 1 2 7 19 Ja p a n 0 2 0 0 0 0 0 0 0 1 1 1 0 2 0 0 2 0 4 16 38 L u x e m b o u rg 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 M a la y si a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 M e x ic o 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 N et h erl a n d s 0 0 0 1 1 0 0 0 0 0 0 0 0 2 1 0 0 0 4 16 32 N e w Z ea la n d 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 N o rw a y 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 3 8 P eru 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 P h il ip p in e s 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 P o la n d 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 P o rt u g a l 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 2 Ru ss ia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 5 S in g a p o re 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 8 S o u th A fr ic a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 2 9 S p a in 0 1 0 0 0 0 0 1 2 0 0 0 0 0 0 0 0 0 4 5 19 S w e d e n 0 4 0 3 0 0 0 1 0 0 0 0 0 0 2 0 0 0 2 15 36 S w it z erl a n d 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 4 15 38 T a iw a n 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 T h a il a n d 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 U n it e d K in g d o m 0 1 0 0 0 1 0 0 0 1 7 0 2 3 0 0 0 2 0 41 87 U n it e d S ta te s 1 7 1 3 0 0 0 2 0 0 2 2 0 8 3 2 0 0 34 0 188 T o ta l 1 27 5 11 1 5 1 6 2 5 13 4 8 20 11 3 3 7 93 263 783
17 To determine the expected stock returns of the acquiring companies, it must be assumed that a stable, linear relationship exists between an acquiring company’s stock returns and the return on the market (MacKinlay, 1997). Since the sample consists of acquirers from different countries, a suitable benchmark should be adopted. Stock markets are different around the world. For example, share markets could differ in size, accounting standards, securities regulation, corporate governance, and in many more ways. These different characteristics could affect the statistical properties of a company’s share returns (Campbell et al., 2010). Therefore, applying one single benchmark to every firm in the sample would not suffice. In their study on multi-country event studies, Campbell et al. (2010) conclude that using national market indices instead of using American or international indices, produces powerful and well-specified tests of share return effects. Therefore, the expected stock returns of the acquiring companies in this sample are calculated by determining the market returns of the indices of the acquiring firms’ countries individually, instead of taking one single index. This study uses an estimation window of 120 trading days (-140, -20), which should be sufficient (MacKinlay, 1997). For each acquiring firm, the market model is defined as:
𝑅𝑖,𝑡 = 𝛼𝑖 + 𝛽𝑖𝑅𝑚,𝑡 + 𝜀𝑖,𝑡 (2)
Where Ri,t denotes the actual stock return of acquiring firm i at time t, Rm,t denotes the return
of the acquiring firm’s country-related market portfolio m at time t, and εi,t is the disturbance term for acquiring firm i’s stock at time t, which value is expected to be zero, with a variance of 𝜎𝜖2𝑖.
Furthermore, αi and βi are the parameters of the market models. The market model is able to isolate the effects of a certain event by correcting the return for the return on the markets and therefore, it is preferred over the constant mean return model (MacKinlay, 1997).
In order to draw generic inferences, the next step is the aggregation of the abnormal returns across the different firms in the sample and through time. Firstly, the abnormal stock returns for each separate acquirer will be aggregated for the specific event windows of interest using the following formula:
𝐶𝐴𝑅𝑖(𝑡1, 𝑡2) = ∑ 𝐴𝑅𝑖,𝑡
𝑡2=1
𝑡1=−1
(3)
18 (-1, 1) and ARi,t is the abnormal stock return of acquiring firm i at time t, as calculated in Eq. (1).
The event-windows utilized in this study are (-1, 1), in line with McNichols and Stubben (2015). After determining the CARs of every acquirer separately, all the CARs of the sample firms are aggregated to calculate the cumulative average abnormal return using the following formula:
𝐶𝐴𝐴𝑅(−1, 1) = 1
𝑁∑ 𝐶𝐴𝑅𝑖(−1, 1)
𝑛
𝑖=1
(4)
Where CAAR(-1, 1) is the cumulative average abnormal return for all the sample firms in event period (-1, 1), and N is the number of firms included in the sample. In order to draw inferences from the CARs obtained from Eq. (3), the CAAR should be significantly different from zero.
3.2.2 Independent variables. The first independent variable in this study is a measure that
captures the differences in accounting standards between two countries, the so-called accounting distance. This measure, developed by Yu and Wahid (2014), is an improved version of the Bae-score (Bae et al., 2008). Bae et al. (2008) developed a measure that captures the differences between local accounting standards and the IFRS on 21 key accounting items based on a global survey of seven globally operating accounting firms. Yu and Wahid (2014) modified this measure to determine differences in accounting standards between countries. However, an increasing number of countries require companies to prepare their financial statements in accordance with the IFRS (PwC, 2017). If an acquisition takes place after one year following the year a target firm’s country adopts the IFRS, the distance between the acquirer’s country GAAP and the IFRS is taken, not the distance between the two local accounting standards. The same applies when the acquirer’s country is adopting the IFRS, in this case, the distance between the IFRS and the target country is taken. If both the target’s and the acquirer’s country adopted the IFRS at least one year before the acquisition was announced, the accounting distance is equal to zero. Appendix B provides an overview of the accounting distances between several countries and between countries and the IFRS.
The second independent variable in this study captures the strength of enforcement systems in the target country. The measure used in this study is developed by Brown et al. (2014). Most studies use a proxy based on the legal origins of countries, like the ones developed by La Porta et al. (1998) and Kaufmann et al. (2011), because they are available for a wide range of countries.
19 However, these proxies mainly focus on elements of legal systems or securities laws and fail to capture the factors affecting the degree of compliance with accounting standards as a result of the activities of independent enforcement bodies and external audit (Brown et al., 2014). The measure of Brown et al. (2014) addresses this deficiency by calculating measures of the degree of accounting enforcement activity and the quality of company auditors’ working environment, based on several proxies. The enforcement index is calculated by summing the audit and the enforce score, resulting in the following formula:
𝐸𝐼𝑋𝑙 = 𝐴𝑈𝐷𝐼𝑇𝑙+ 𝐸𝑁𝐹𝑂𝑅𝐶𝐸𝑙 (5)
Where EIXl is the total score on the enforcement index for target firm’s country l, AUDITl
is the score of target firm’s country l on the audit dimension, and ENFORCEl is the score of target
firm’s country l on the enforce dimension.
Lastly, a variable measuring the earnings quality of the target company will be included. The measure included in this thesis is developed by Jones (1991) and captures earnings quality by focusing on the relationship between a company’s accruals and its fundamentals. Jones (1991) determines abnormal, or discretionary, accruals by modeling total accruals against PPE and changes in revenues. This study adopts a cross-sectional variant of the Jones (1991) model as proposed by DeFond and Jiambalvo (1994). This model provides an industry-relative measure of abnormal accruals. Cross-sectional methods are preferred over time-series variants because of their better specification and higher statistical power (Jeter and Shivakumar, 1999). Furthermore, the data requirements for the cross-sectional method are less stringent. The model defines earnings quality as the absolute value of a firm’s discretionary accruals. This value is multiplied by -1 so that a higher score implicates higher earnings quality:
𝐴𝑄𝑗,𝑡 = −|𝐷𝐴𝐶𝐶𝑗,𝑡| = −|𝑇𝐴𝐶𝐶𝑗,𝑡− 𝑁𝐷𝐴𝐶𝐶𝑗,𝑡| (6)
Where AQj,t is a target firm j’s earnings quality in year t, -|DACCj,t| is the absolute value
of target firm j’s discretionary accruals in year t multiplied by -1, TACCj,t =(∆CAj,t - ∆CLj,t -
∆CASHj,t + ∆STDEBTj,t - DEPNj,t)is the total accruals of target firm j in year t. ∆CAj,t is the change
in current assets of target firm j between year t and year t-1, - ∆CLj,t is the change in current
liabilities of target firm j between year t and year t-1, ∆CASHj,t is the change in cash of target firm
j between year t and year t-1, ∆STDEBTj,t is the change in short-term debt of target firm j between
20 and NDACCt is a target firm j’s non-discretionary accruals in year t. The non-discretionary accruals
are estimated as follows:
𝑁𝐷𝐴𝐶𝐶𝑗,𝑡 = 𝛽1 1 𝑇𝐴𝑗,𝑡−1+ 𝛽2 ∆𝑅𝐸𝑉𝑗,𝑡 𝑇𝐴𝑗,𝑡−1 + 𝛽3 𝑃𝑃𝐸𝑗,𝑡 𝑇𝐴𝑗,𝑡−1 (7)
where TAj,t-1 are the total assets of target firm j in year t-1, ∆REVj,t is the change in revenues
of target firm j in year t, and PPE is the property plant and equipment of target firm j in year t. This implies that the abnormal, or discretionary, accruals are equal to the absolute values of the residuals of the following model:
𝑇𝐴𝐶𝐶𝑗,𝑡 𝑇𝐴𝑗,𝑡−1 = 𝛽1 1 𝑇𝐴𝑗,𝑡−1+ 𝛽2 ∆𝑅𝐸𝑉𝑗,𝑡 𝑇𝐴𝑗,𝑡−1 + 𝛽3 𝑃𝑃𝐸𝑗,𝑡 𝑇𝐴𝑗,𝑡−1+ 𝜀𝑗,𝑡 (8)
Eq. (8) is estimated for all the industries included in the sample identified by the one-digit SIC code that have data available for at least 20 firms in year t. Annual cross-sectional estimations of Eq. (8) provide firm-specific residuals and year-specific residuals. The absolute value of these residuals, multiplied by -1, is taken to determine accrual quality for every firm in the sample.
3.2.3 Control Variables. The analysis includes several variables controlling for acquiring
firm’s characteristics. Firstly, the natural logarithm of an acquiring firm’s age is added to the regression. Older companies are more likely to have engaged in previous international acquisitions compared to younger companies. These differences in experience might affect the acquisition-performance relationship. For example, Fowler and Schmidt (1989) showed that older companies perform better in acquisitions, compared to younger ones.
Secondly, a control variable for firm size is added to the model. Moeller et al. (2004) showed that, in general, small acquisitions by small acquirers were value enhancing, whereas large acquisitions by large acquirers were value destroying. On the other hand, Healy et al. (1992) provided evidence for positive post-acquisition performance for acquisitions by larger firms. Thus, although scholars might not agree if firm size affects acquisition performance positively or negatively, there seems to be an effect. Therefore, the natural logarithm of the acquiring firm’s total assets in year t-1 is included as a control variable.
Thirdly, a variable controlling for potential information leakage before the acquisition announcement is added to the regressions, as proposed by Korczak et al. (2010). This variable is the return on the acquirer’s stock in the period from 120 trading days before the announcement
21 until 31 calendar days before the announcement. According to Ahern (2009), a company’s prior return could drive a company’s abnormal returns. He finds that firms with lower prior returns experience higher, but biased, abnormal returns and vice-versa. Therefore, a negative relationship between prior returns and abnormal returns is expected.
Finally, some corporate governance-related variables are included in the regressions. The stronger the corporate governance mechanisms in the acquiring firm, the lower the likelihood of a CEO being able to engage in ‘empire-building’ behavior and make value-destroying acquisitions (Hayward and Hambrick, 1995). The first variable gauging the strength of the acquirer’s corporate governance is board size, which is said to influence the strength of corporate governance (Elsayed, 2011). Therefore, the natural logarithm of the number of directors on the acquirer’s board of directors is added to the model.
Secondly, a variable indicating the degree of independence of the board is added to the model. The more independent the board, the stronger the corporate governance (Hayward and Hambrick, 1995). Board independence is defined by the percentage of outside directors on the board.
Lastly, a dummy variable is included taking value 1 if one of the directors is a financial expert, and value 0 if not. Intuitively, having a financial expert on the board is beneficial in the due-diligence process, especially if earnings quality of the target firm is low. However, Güner et al. (2008) find that companies with financial experts on the board undertake more value-destroying acquisitions because they engage in higher risk-taking behavior.
Several deal characteristics are also considered. Firstly, a variable is included indicating if the acquirer and the target operate in the same industry, as identified by the two-digit SIC code. Even though the influence of relatedness on value creation is unclear, the relatedness of targets to their acquirers is often assumed to affect acquirer’s acquisition performance (Datta and Puia, 1995; King et al., 2004). Therefore, a dummy variable is added to the model. If the acquirer and the target operate in the same industry, this dummy takes value 1, if not, the dummy is equal to 0.
Secondly, a dummy variable indicating if there were any competing bids is added to the model. If any competing bidders are present, the dummy takes value 1, if not, the dummy takes value 0. In the case of competing bids, it is possible that the acquirer fell victim to the ‘winners curse’ (Bazerman and Samuelson, 1983). If the acquirer beat other bidders in the process, it is likely the acquirer overestimated the value of the target company relative to competing bidders
22 (Thaler, 1988). This implies that the acquirer overpaid for the target, which has been shown to negatively affect acquisition performance (Malmendier and Tate, 2008).
Thirdly, a dummy variable indicating if the acquisition was negotiated (i.e. friendly) or hostile is included. This dummy takes value 1 if the deal was hostile, and value 0 if not. Bhagat et al. (1990) show that acquiring firms’ shareholders often lose value in hostile acquisitions.
Fourthly, a dummy variable indicating if the acquisition is a tender offer or not is included in the model. The dummy takes value 1 if the acquisition is a tender offer, and value 0 if not. Jarrell and Poulsen (1989) presented evidence for negative value implications for acquirer shareholders in the case of tender offers.
Finally, a dummy variable indicating if the acquisition is paid with equity is included. This dummy takes value 1 if the deal price is paid with stock, and 0 if not. Heron and Lie (2002) found that announcement and post-acquisition shareholder returns were lower for acquisitions paid in stock.
The last set of control variables consists of country-level variables. By including these control variables the outcomes of this study are not the result of trade flows between countries, but the result of differences in accounting standards and quality of enforcement. Firstly, the cultural distance between the acquirer’s country and the target’s country is included. The cultural distance between two countries is a proxy for the degree two countries culturally differ from each other. Previous literature found opposing effects for cultural distance on international acquisitions (Datta and Puia, 1995; Chakrabarti et al., 2009). The measure used is developed by Kogut and Singh (1988) and calculates cultural distance using the six dimensions of national culture as identified by Hofstede et al. (2010) using the following formula;
𝐶𝐷𝑘,𝑙 = ∑ (𝐼𝑝,𝑘
6
𝑝=1 − 𝐼𝑝,𝑙)/𝑉𝑝
6 (9)
Where CDk,l denotes the cultural distance between acquiring firm’s nation k and target
firm’s nation l, Ip,k is acquiring firm’s nation k’s score on the pth cultural dimension, Ip,l is target
firm’s nation l’s score on the pth cultural dimension, and Vp is the variance of all the scores on
cultural dimension p.
Secondly, a variable controlling for the geographic distance between the acquiring firm’s country and the target firm’s country is included. Previous research highlighted the role of geographic distance in international business (Frankel and Romer, 1999). The natural logarithm of
23 the geographic distance in kilometers between the capital cities of the acquiring and the target firm’s country is added to the model.
Thirdly, a control variable indicating the target firm’s country’s openness to the world economy is included in the regression. The more open a target firm’s country is, the more manageable the newly acquired company is, and the more efficient the new subsidiary’s profits can be employed. This should positively affect acquisition performance (Chakrabarti et al., 2009). The openness of a target’s firm country to the world economy is calculated using the following formula:
𝑂𝑝𝑒𝑛𝑛𝑒𝑠𝑠𝑙 = 𝐼𝑚𝑝𝑜𝑟𝑡𝑙+ 𝐸𝑥𝑝𝑜𝑟𝑡𝑙
𝐺𝐷𝑃𝑙 (10)
Where Opennessl is the openness of target firm’s nation l to international trade, Importl is
the aggregate import of target firm’s country l in year t-1, Exportl is the aggregate export of target
firm’s nation l in year t-1, and GDPl is the GDP of target firm’s nation l in year t-1.
Lastly, a control variable indicating economic synergies between the acquirer firm’s and target firm’s country is added to the regression. Economic synergies between countries could exist when the sum of bilateral trade flows between two countries are high. This could enhance acquisition performance (Chakrabarti et al., 2009). Economic synergies are calculated as follows:
𝐵𝑖𝑙𝑎𝑡𝑒𝑟𝑎𝑙𝑘,𝑙 = 𝑙𝑛(𝐸𝑥𝑝𝑜𝑟𝑡𝑙,𝑘+ 𝐼𝑚𝑝𝑜𝑟𝑡𝑙,𝑘) (11)
Where Bilateralk,l denotes the degree of economic synergies between acquiring firm’s
nation k and target firm’s nation l, Exportl,k is the aggregate export of target firm’s country l to
acquiring firm’s country k in year t-1, and Importl,k is the aggregate import of target firm’s country
l from acquiring firm’s country k in year t-1.
Table 2 presents descriptive statistics of all the variables included in this study’s regressions, including a correlation matrix. Multicollinearity does not seem to be an issue when looking at the correlation matrix, and a non-reported VIF-test confirms this. When necessary, variables are winsorized to limit the influence of extreme outliers. Although multicollinearity is not an issue, the correlation between some variables is high. For example, accounting distance and cultural distance are positively related. This indicates that there might indeed be a linkage between a country’s culture and its accounting systems, as proposed by Gray (1988). Furthermore, the bilateral trade between acquirer- and target countries seems to be positively related to the strength
24 T a b le 2 De sc ri p ti v e st ati sti cs a n d c o rr el ati o n m at ri x o f v ar ia b les i n cl u d e d i n t h e m o d els P a n el a : Des cr ip ti ve s ta ti st ic s In d ep en d en t va ria b les N M ea n S t. De v . M in M ax M ed ia n C A R (-1 ,1 ) 7 8 3 0 .0 0 3 0 .0 6 1 -0 .1 8 4 0 .2 3 2 0 .0 0 0 AC DI F F 7 8 3 4 .8 2 5 4 .1 0 1 0 17 4 AQ 1 2 5 -0 .0 8 8 0 .0 3 4 -0 .1 7 7 -0 .0 0 5 -0 .0 8 6 E IX 7 8 3 4 3 .9 4 6 1 3 .1 7 5 7 56 52 F ir m ag e 7 8 3 3 .7 4 1 1 .2 2 7 0 5 .7 3 7 4 .2 2 0 F ir m size 7 8 3 1 5 .4 1 0 2 .4 0 4 5 .7 8 1 2 1 .5 8 3 1 5 .4 1 3 P as tr et u rn 7 8 3 0 .0 6 3 0 .2 3 0 -1 .0 4 7 1 .6 2 9 0 .0 4 5 B o ar d size 7 8 3 1 1 .4 3 9 4 .8 9 8 3 34 10 B o ar d in d 7 8 3 0 .5 4 5 0 .2 7 6 0 1 0 .5 7 1 F in ex p 7 8 3 0 .4 8 9 0 .5 0 0 0 1 0 R elate d n es s 7 8 3 0 .6 1 4 0 .4 8 7 0 1 1 C o m p eti n g 7 8 3 0 .0 6 0 0 .2 3 8 0 1 0 Ne go ti at e d 7 8 3 0 .0 1 9 0 .1 3 7 0 1 0 T en d e r 7 8 3 0 .3 6 5 0 .4 8 2 0 1 0 E q u it y 7 8 3 0 .0 4 3 0 .2 0 4 0 1 0 CD 7 8 3 1 .1 7 1 1 .0 7 7 0 .0 2 0 6 .9 4 7 0 .9 8 6 GD 7 8 3 8 .0 6 9 1 .2 1 9 5 .3 0 8 1 0 .0 4 6 8 .6 8 8 Op en n es s 7 8 3 0 .5 8 5 0 .4 7 5 0 .1 8 8 4 .3 2 9 0 .5 3 3 P a n el b : C o rr ela ti o n ma tr ix CAR (-1 ,1 ) ACDI FF EIX AQ Fir ma ge Fir msize Pas tre turn Bo ard size Bo ard ind Fin ex p Relate dn ess Co mp eti ng Neg oti ated Te nd er Eq uit y CD GD Ope nn ess Bil ater al C A R (-1 ,1 ) 1 .0 0 0 AC DI F F -0 .0 3 9 1 .0 0 0 E IX 0 .0 3 3 -0 .2 9 4 * 1 .0 0 0 AQ -0 .1 3 1 -0 .0 0 7 0 .0 2 6 1 .0 0 0 F ir m ag e 0 .0 5 3 0 .0 2 4 0 .0 0 1 0 .0 5 6 1.000 F ir m size -0 .0 9 3 * 0 .1 0 4 * -0 .0 5 6 0 .1 2 4 0 .3 7 7 * 1 .0 0 0 P as tr et u rn -0 .0 7 0 * -0 .0 4 1 0 .0 1 9 -0 .0 3 8 -0 .1 1 3 * -0 .0 9 6 * 1 .0 0 0 B o ar d size -0 .0 9 2 * 0 .1 1 1 * -0 .0 7 8 * 0 .0 5 6 0 .3 1 2 * 0 .6 5 8 * -0 .1 0 0 * 1 .0 0 0 B o ar d in d 0 .0 1 1 -0 .2 4 4 * 0 .1 8 9 * 0 .1 2 7 0.002 -0 .0 2 1 0 .0 3 1 -0 .2 5 0 * 1 .0 0 0 F in ex p 0 .0 0 1 -0 .1 7 4 * 0 .2 1 6 * 0 .0 6 7 0.006 0 .0 1 9 -0 .0 0 6 -0 .0 0 8 0 .3 6 7 * 1 .0 0 0 R elate d n es s -0 .0 2 6 -0 .0 0 4 0 .0 2 8 0 .0 2 4 -0 .0 7 4 * 0 .0 0 1 -0 .0 3 9 -0 .0 0 6 -0 .0 2 0 0 .0 2 5 1 .0 0 0 C o m p eti n g -0 .0 2 9 -0 .0 2 1 0 .0 0 0 -0 .1 1 8 -0.030 -0 .0 4 6 0 .0 3 5 -0 .0 4 6 -0 .0 2 9 0 .0 3 2 0 .0 1 3 1 .0 0 0 Ne go ti ate d -0 .0 2 9 -0 .0 1 0 -0 .0 2 9 -0 .0 9 7 0.055 0 .0 6 4 * -0 .0 1 0 0 .0 2 1 0 .0 2 2 0 .0 3 1 -0 .0 0 4 0 .1 6 1 * 1 .0 0 0 T en d e r 0 .0 4 5 0 .0 1 2 0 .0 1 1 -0 .0 4 7 -0.058 -0 .0 3 1 -0 .0 0 9 -0 .0 4 8 -0 .0 0 8 -0 .0 1 5 -0 .0 0 4 0 .1 3 2 * 0 .1 6 5 * 1 .0 0 0 E q u it y -0 .0 1 6 -0 .0 6 6 * 0 .0 5 5 -0 .0 5 2 -0.046 -0 .0 1 0 -0 .0 2 0 0 .0 3 4 -0 .0 1 7 0 .0 3 0 0 .0 4 0 -0 .0 2 8 -0 .0 3 0 -0 .0 0 6 1 .0 0 0 CD 0 .0 0 6 0 .3 9 7 * -0 .2 1 4 * 0 .0 2 4 0.059 0 .1 6 3 * -0 .0 0 9 0 .1 0 8 * -0 .2 7 3 * -0 .0 9 4 * -0 .0 2 8 0 .0 3 1 -0 .0 1 1 0 .0 0 2 -0 .0 3 4 1 .0 0 0 GD 0 .0 5 3 0 .2 3 6 * 0 .0 7 1 * 0 .0 1 4 0.038 0 .0 3 1 -0 .0 0 2 -0 .0 2 1 -0 .0 0 6 -0 .0 5 0 -0 .0 3 7 -0 .0 5 4 0 .0 0 2 0 .0 0 7 0 .0 1 9 0 .3 3 5 * 1 .0 0 0 Op en n es s -0 .0 3 0 -0 .0 4 9 -0 .2 6 6 * -0 .1 2 8 -0.048 0 .0 0 8 0 .0 3 9 -0 .0 1 9 -0 .0 1 5 -0 .0 0 6 0 .0 0 2 -0 .0 2 3 -0 .0 2 4 0 .0 0 5 -0 .0 3 2 0 .0 7 4 * -0 .1 7 1 * 1 .0 0 0 B il ater al -0 .0 1 1 -0 .1 3 1 * 0 .4 0 4 * 0 .0 1 0 -0.035 -0 .0 3 6 0 .0 0 1 -0 .0 1 7 0 .2 1 6 * 0 .1 7 3 * 0 .0 2 0 0 .0 3 5 -0 .0 0 3 -0 .0 3 7 0 .0 2 5 -0 .3 1 3 * -0 .4 2 6 * -0 .1 2 6 * 1 .0 0 0
