The effect of corruption distance on acquirer’s abnormal
returns from cross-border M&A announcements
MSc International Financial Management Faculty of Economics and Business
University of Groningen
Abstract
This study examines the influence of corruption distance between countries on acquirer’s abnormal returns around the announcement of a cross-border merger or acquisition. Country pairs of acquirer’s home country and target’s host country are divided in different subsamples in order to analyze the effect of corruption distance, a concept which builds upon internationalization theory. The sample consists of 2,045 cross-border M&A deals and the results provide no significant evidence. Firms acquiring a closer target in terms of corruption distance do not significantly realize higher abnormal returns.
JEL classifications: G14, G34
Keywords: announcement returns, mergers and acquisitions, corruption
Master’s Thesis IFM Author: Marlon Berkhoff Student number: S2350777
E-mail: m.a.r.berkhoff@student.rug.nl Date: 09/06/2017
2
Content
1. Introduction ...3
2. Literature review ...6
2.1 Corruption ...6
2.2 Corruption and FDI ...7
2.3 Corruption, cross-border M&As, and announcement returns ...9
2.3.1 Industrialized acquirers ... 11
2.3.2 Developing acquirers ... 12
3. Research method and data selection ... 14
3.1 Data... 14 3.1.1 Sample ... 14 3.2 Methodology ... 16 3.2.1 Regression model ... 16 3.2.2 Dependent variable ... 16 3.2.3 Independent variable ... 17 3.2.4 Control variables ... 18 4. Results ... 21 4.1 Descriptive statistics ... 21 4.2 Multicollinearity ... 23 4.3 Heteroscedasticity ... 23 4.4 Regression results ... 25 4.5 Robustness check ... 30
5. Discussion and conclusions ... 31
6. References ... 34
Appendix A ... 38
3
1.
Introduction
In the year 1986, a remarkable trend initiated as significant growth in foreign direct investment (FDI) flows can be seen. Many FDI recipient countries liberalized their regimes and this period marked the start of an era where firms sought for cross-border investment opportunities. Firms started to substantially globalize their operations. In the period 1986-2001, the volume of FDI has been growing over 20% annually (UNCTAD, 2001). When firms want to become multinational companies (MNEs), they constantly have to evaluate both whether and how to enter foreign markets. According to Pan and Tse (2000), different entry modes are associated with varying degrees of resource commitment, risk exposure, control, and profit returns. Mergers and acquisitions (M&As) are important strategic corporate initiatives to internationalize (Di Guardo, Marrocu, and Paci, 2016). Firms have moved away from more traditional greenfield investments and M&As have become the primary mode of internationalization (UNCTAD, 2000). According to UNCTAD (2007), for the year 2006 global FDI outflows were valued at $1,215 billion and cross-border M&As accounted for $880 billion (72%). This amount was a 131% increase compared to the 2004 figure.
4 Literature on corruption and FDI mainly analyzes the consequences of corruption at the country level and there is consensus corruption negatively affects a country. Mauro (1995) finds lower economic growth for higher corruption levels, Ades and Di Tella (1997) relate higher corruption levels to lower public policy effectiveness and Wei (2000a) reports lower FDI inflows for corruptive countries. However, at the company level, theorists are less in consensus regarding the effects of corruption (Cuervo-Cazurra, 2016). Similarly, according to Qian and Sandoval-Hernandez (2016) empirical studies are inconclusive in their findings, with studies finding both positive and negative effects of corruption. Empirical research which supports the negative effects of corruption on FDI finds that corruption deters FDI by ‘‘making a host country unattractive to foreign companies via the high costs of entry and uncertainty, and distorting incentives to invest’’ (Qian and Sandoval-Hernandez, 2016, p.419). Contrary, others find that corruption may act as a ‘helping hand’ increasing companies’ revenues (e.g. Egger and Winner, 2005) and may stimulate FDI inflows from MNEs because of increased efficiency by speeding up bureaucratic processes (Lui, 1985) or gaining access to publicly funded projects (Tanzi and Davoodi, 2000).
5 This paper will study the influence of corruption distance on the firm value of an acquirer. More specifically, the influence of corruption distance on the wealth effects of an acquirer’s cumulative abnormal returns around the announcement of a cross-border merger or acquisition will be analyzed, which leads to the following research question:
What is the effect of corruption distance on an acquirer’s abnormal returns around the announcement of a cross-border merger or acquisition?
The empirical results show that there is no significant relationship between corruption distance and acquirer’s cumulative abnormal returns. The full sample of 2,045 cross-border mergers and acquisitions provide, as expected, consistently negative coefficients, indicating that a closer target country in terms of corruption results in higher abnormal returns for the acquiring firm. The insignificance of the results indicates that other factors, such as host country’s economic conditions, need to be considered in addition to a host country’s level of corruption.
6
2.
Literature review
Firms can create value through internationalizing as a result of imperfections in product and factor markets, different taxation regimes, and imperfections in international financial markets (Doukas and Travlos, 1988). While internationalization through mergers and acquisitions has emerged over recent decades, researchers and business practitioners have explored the effect of multinationalism on firm value. Short-term wealth effects for target firms have been well-established in previous literature, which acknowledge that target firms reap positive market reactions and thus are the main beneficiaries of an M&A announcement (e.g. Lang, Stulz, and Walkling, 1989). Researchers find both positive and negative market reactions for acquirers’ M&A announcements. As firms and markets become more internationally integrated, one factor that has drawn attention and of which firms increasingly have become aware of is corruption in host countries.
2.1 Corruption
7 advantage of firms in exchange for bribes) it would, like an auction, get awarded a contract based on the bribe’s size. In a similar vein, Egger and Winner (2005) provide empirical evidence that although corruption raises costs in the short run, in the long run there is a helping hand outweighing the initial short-term costs. The negative view towards corruption is more prominently defended by scholars. An important implication in this view is the difficulty of limiting corruption in areas where it actually might be economically desirable (Rose-Ackerman, 1978). Corruption is therefore rarely restricted to areas where it may increase welfare. The negative effects of corruption are referred to as sand in the wheels of commerce and corruptive practices would result in the wasteful use of resources devoted to both corruption practices itself as well as to the fighting of it (Cuervo-Cazurra, 2006). Firms which participate in bribery practices are not sure whether promised contracts or goods are delivered, and these firms are not able to go to court to demand fulfillment of a contract. Mauro (1995) and Tanzi (1998) find, on empirical grounds, negative consequences of corruption on economic growth, business development, and domestic and foreign investment.
2.2 Corruption and FDI
8 flows from 12 source countries to 45 host countries to study the effects of corruption. Wei (2000a) finds statistically significant relationships supporting the negative notion that corruption hurts FDI flows and states that ‘‘an increase in the corruption level from that of Singapore to that of Mexico would have the same negative effect on inward FDI as raising the tax rate by eighteen to fifty percentage points’’ (Wei, 2000a, p.8). Taking properly into account government policies towards FDI, including host governments’ restrictions on FDI and governments’ incentives to attract FDI, Wei (2000b) finds again statistically significant negative relations between corruption and FDI. Habib and Zurawicki (2001) add to the generalizability of the negative effects of corruption on FDI flows by studying inward FDI flows of the entire world, whereas previous studies mainly focused on flows from developed countries to developing countries.
9 2.3 Corruption, cross-border M&As, and announcement returns
As mentioned earlier, cross-border M&As have become the primary mode of internationalization. The most important motive for cross-border M&A deals is the same as for domestic M&A deals: the creation of value, because the acquiring firm’s managers want to merge or acquire when the combined value of two firms is greater than the combined value of the firms individually (Erel, Liao, and Weisbach, 2012; Xie, Reddy, and Liang, 2017). However, contrary to domestic deals, international deals are associated with unique challenges and frictions that can impede M&As post-deal performance and value. Erel et al. (2012) list several factors that are related to national boundaries affecting the costs and benefits of a cross-border merger or acquisition. First, nations have their own specific cultural identities, with different languages, different religions, and non-similar ways of conducting business (Ahern, Daminelli, and Fracassi, 2010). Second, physical distance, measured as geographic distance, decreases the likelihood of two firms from different countries merging. Furthermore, different corporate governance considerations can affect international deals. Countries that have better governance policies in place through higher quality accounting or legal standards are more likely to acquire firms in target countries where legal and accounting standards are of lower quality. Lastly, the market’s level of development is a factor that influences cross-border acquisitions. Erel et al. (2012) do not explicitly mention corruption as an important host-country factor influencing a cross-border deal, although a country’s corporate governance will include, among other variables, a country’s level of corruption. Malhotra et al. (2010) acknowledge the importance of corruption as a specific country variable and, in line with Habib and Zurawicki (2001) and Cuervo-Cazurra (2006), consider the importance of similarity or dissimilarity between home-country corruption and host-country corruption.
10 commitments and decisions regarding to which extent equity in a host country is utilized can be influenced by country-specific factors such as corruption. Acquisitions, after greenfield investments, require the highest equity commitment by firms when internationalizing and firms have to assess investment risk, return, location choice, and adaptation to the local environment (Pan and Tse, 2000). By acquiring a target in the local market, knowledge about the local environment regarding corruption in the host country can be obtained. As mentioned earlier, Habib and Zurawicki (2001) find empirical evidence that local business is used to deal with the corruption on their own turf. A managerial implication provided by Habib and Zurawicki (2001) based on this finding is that foreign investors should reach for some local ‘connectivity’ through, for example, the acquisition of a local company in order to adapt to or resist local corruption. However, Weitzel and Berns (2006) stress that a target which acts as an insider may have a stronger bargaining position when corruption levels are higher, which can be disadvantageous for the acquirer.
Qian and Sandoval-Hernandez (2016) introduce the concept of corruption distance, derived from the idea of psychic distance. In internationalization theory, psychic distance is based on the notion that when psychic distance is low, a foreign market tends to be more similar to one’s home country which in turn reduces uncertainty (Johanson and Vahlne, 1977). Therefore, when internationalizing, firms tend to move first to ‘closer’ countries. This psychic closeness would reduce learning costs when entering a foreign market and would promote investment activities. As stressed above, uncertainty resulting from corruption has to be dealt with by acquirers. Following the concept of psychic distance, lower corruption distance between an acquirer’s home country and the target’s host country would mean less uncertainty for the acquirer. Therefore, it is likely that acquirers would prefer to move to countries with a similar level of corruption when internationalizing in order to cope with uncertainty and learning costs.
11 internationalizing can be seen as a firm’s inability to extract additional profits and benefits from its existing domestic operations (Doukas and Travlos, 1988). Internationalizing may be perceived as a matter of survival. This will be negatively perceived by investors. Furthermore, other explanations for negative market reactions resulting from internationalizing include higher agency costs (Bodnar, Tang, and Weintrop, 1997) and costs related to information asymmetries (Myerson, 1982). Contrary, researchers also find positive market reactions. Fuller, Netter, and Stegemoller (2002) find positive abnormal returns for firms which acquire a private firm and returns are higher when the target is larger. According to Kogut (1983), multinationalism creates value when an acquirer benefits from distortions in international markets. These distortions relate to the acquirer’s ability to arbitrage institutional restrictions, to capture informational externalities, and costs savings resulting from joint production and marketing (Kogut, 1983). Positive market reactions would acknowledge the value creating potential when firms seek to internationalize through cross-border M&As.
Uncertainty, informational asymmetries, and higher agency costs related to internationalizing and the amplifying effect of corruption may harm acquirer’s abnormal returns when moving to corruptive host countries. As companies engage in M&As to create and increase firm value and in turn increase shareholder value, valuation is often derived from potential synergies which can be reached. Targets countries of low corruption distance may be expected to have a more similar institutional environment because corruption levels will be of comparable height. Based on the theoretical implications and findings from previous literature, the first hypothesis for this study is:
H1: There is a negative relationship between corruption distance and an acquirer’s abnormal returns.
2.3.1 Industrialized acquirers
12 in institutional circumstances will be great and, in terms of corruption, there is low uncertainty related to entering the foreign market.
H2: The negative effect of corruption distance will not have a large effect on the announcement returns for an acquirer from an industrialized country announcing a cross-border M&A of a target in another industrialized country.
Corruption is seen as a fundamental problem for developing economies (Bardhan, 1997). Therefore, an industrialized acquirer will experience, compared to its home country, a higher level of corruption when acquiring a target in a developing country. This situation is referred to as negative corruption distance by Qian and Sandoval-Hernandez (2016). High corruption has mainly been linked to developing economies where the administrative situation fosters excessive and discretionary power, and where laws are barely transparent (Tanzi, 1998; LaPolambara, 1994). Furthermore, an acquirer from a high corruption environment gained expertise and experience to operate in corruptive foreign environments (Habib and Zurawicki, 2002). Acquirers from industrialized countries will generally lack this experience when entering a country with a higher level of corruption. Research provided evidence that high corruption in the host country results in lower stock returns to shareholders of the acquiring firm following a M&A announcement (Francis, Hasan, Sun, and Waisman, 2014). As corruption distance increases, industrial acquirers will lack the ability to handle higher levels of corruption, which negatively influences a firm’s abnormal returns.
H3: Negative corruption distance is negatively related with abnormal returns for acquirers from industrialized countries when announcing a cross-border M&A of a target in a developing country.
2.3.2 Developing acquirers
13 Cuervo-Cazurra and Genc (2008) state that multinationals from corrupt countries often lack firm-specific competitive advantages which are related to firms from less corrupt countries, such as technology or branding. Firms from corrupt environments benefit from skills specifically for dealing with corruption and will view firm-specific investments in branding or technology of less competitive value. In line with the psychic distance theory, developing acquirers’ home countries and developing targets’ host countries will both have high corruption, there is low corruption distance and norms and skills for doing business will be more similar.
H4: The negative effect of corruption distance will not have a large effect on the announcement returns for an acquirer from a developing country announcing a cross-border M&A of a target in another developing country.
Acquirers from developing countries, where corruption levels are generally high, have to deal with low host country corruption when acquiring a target in an industrialized country. Qian and Sandoval-Hernandez (2016) refer to this situation as positive corruption distance. Developing acquirers seeking to acquire targets in industrialized countries may be searching for efficiency (Qian and Sandoval-Hernandez, 2016). These firms may want access to managerial know-how and cutting-edge technologies which is available in industrialized environments. Such efficient seeking behavior may offset the effects of corruption in the home environment and a more stable host environment could be perceived positively by a firm’s investor leading to positive market reactions. Therefore, the larger the corruption distance, the less corrupt the host country environment will be for the developing acquirer.
14
3. Research method and data selection
The effect of corruption on the wealth effects of announcement returns is undiscovered. In order to extend the research on abnormal returns and to contribute to the scarce literature on the effects of corruption on firms announcing a cross-border merger or acquisition, an event study will be conducted. Event studies are used to measure whether and how the market incorporates new information regarding an important event such as the announcement of a cross-border merger or acquisition. This section will elaborate on the data selection procedure, the regression model, and the variables included in this study.
3.1 Data
Data on mergers and acquisitions is collected from the electronic database Zephyr. The Zephyr database is one of the most comprehensive databases regarding M&A deals. A total of 442,428 mergers and acquisitions were completed for the time period from 1997, the year the OECD convention combating corruptive practices, was signed, until 2016. Country-specific corruption data is obtained from the Corruption Perception Index (CPI) which is annually reported by Transparency International. CPI data is openly accessible through Transparency International. Stock return data of the firm, and the related stock market, and data on financial variables of the firms is obtained from the Thomson DataStream database. The capital control data is obtained from the Economic Freedom Ranking by the Fraser Institute.
3.1.1 Sample
15 CPI by Transparency International. (8) An acquirer’s stock data and the related market data need to be available in Thomson DataStream.
The criteria above may lead to biased results in several ways. Malhotra et al. (2010) find that firms from countries with low corruption levels tend to avoid highly corruptive countries which can be seen both in the transaction values and the number of acquisitions towards these environments. A minimum deal value of one million dollar may not grasp a number of deals involving acquirers from low corruption countries with targets in high corruption countries which are of smaller value, while there may be positive or negative wealth effects related to these deals. Furthermore, in general, listed acquirers are scarce compared to private acquirers and this is especially the case when the acquirer is from a developing country. Due the less developed financial markets in developing countries, relatively few listed acquirers from developing countries can be found in the given time period and the sample therefore consists mainly of acquirers from industrialized countries.
In line with Qian and Sandoval-Hernandez (2016), the sample is arranged in the following groups in order to test hypotheses 2, 3, 4, and 5:
- Industrialized acquirer – Industrialized target - Industrialized acquirer – Developing target - Developing acquirer – Developing target - Developing acquirer – Industrialized target
16 3.2 Methodology
3.2.1 Regression model
In order to test the hypotheses, an ordinary least squares (OLS) regression will be performed. The dependent variable (i.e. cumulative abnormal returns) will be estimated as a function of the independent variable (i.e. corruption distance) and control variables. The regression model follows in Equation (1):
(1)
CARjt is the cumulative abnormal return for acquirer j for event period t. The constant is represented by . CDhkt represents the corruption distance between the acquirer’s home country corruption level h and the target’s host country corruption k for the year of event period t. APjt is the accounting performance of acquirer j prior to event period t. FSjt is acquirer’s j firm size for the year of the event period t. ECt-1 is target country’s economic conditions measured as GDP growth one year prior to event period t. CCt is the capital controls in the target’s host country for the year of event period t. IDi represents whether
target i is in the same industry as the acquiring firm. The random error is represented by et.
3.2.2 Dependent variable
The dependent variable is the cumulative abnormal return (CAR) of the acquiring firm. Using the daily stock returns of a listed firm, the market’s returns, and the announcement date of a merger or acquisition, a firm’s abnormal announcement returns can be measured. The market-adjusted return model for event studies by Brown and Warner (1985) is used. Appendix A provides a comparison with another event study, namely the market model. An event study measures the effects of an event on the value of the firm based on the notion that, when markets are rational, specific events will immediately affect the prices of a security (MacKinlay, 1997). The method by Brown and Warner (1985) uses the mean adjusted return as the expected return. The related equation is shown in Equation (2):
(2)
cross-17 border M&A, a firm’s returns can be subtracted from the market’s return as shown below in Equation (3):
(3)
ARjt is the abnormal return for firm j on day t, Rjt is the rate of return on security j for event day t, and Rmt is the market return on event day t.
Previous literature used a variety of event windows to calculate CARs. A three-day, five-day or seven-day window can be used to calculate the CAR to reflect the value creation or value reduction caused by the announcement of a cross-border M&A. As mentioned before, lags in responses of prices to new events are short-lived. However, no clear explanation is provided what event period is most closely related to ‘short-lived’. Therefore, this paper studies the most common used event windows, which are mentioned above. A firm’s CAR is the sum of the abnormal returns over the event period, specified by:
3.2.3 Independent variable
The corruption distance is constructed using CPI data by Transparency International. In line with Qian and Sandoval-Hernandez (2016), corruption distance (CD) is measured by the logarithm of the absolute value difference between the corruption indices CPI of the acquirer’s home country h in year t and the target’s host country k in year t, plus 2. This can be written as in Equation (5):
CDhk = CPIh,t – CPIk,t ] + 2) (5)
18 International ratings and the International Country Risk Guide index. All of the mentioned indices are by definition subjective. However, according to Lee and Ng (2009), the patterns these ratings reveal need to be taken seriously. The different ratings are highly correlated, despite the fact that organizations use different techniques to derive ratings. This reduces the chance that results of a particular rating reflect biases of a different measuring system of a particular institution. Moreover, empirical research confirms that ‘‘these subjective ratings are correlated with a wide variety of economic and social phenomena’’ (Lee and Ng, 2009, p.25). The CPI reflects aggregated information from up to 12 individual ratings and surveys, with respondents including business people, risk analysts, and members of the general public.
3.2.4 Control variables
As the importance of firm- and transaction-level besides macroeconomic data in the field of corruption was stressed by Habib and Zurawicki (2002), Weitzel and Berns (2006) find only few empirical studies which control for country-, firm-, and transaction-level characteristics. Omitted variables would lead to inconsistent and biased coefficients for the other variables. Standard errors would be biased, which could lead to inappropriate inferences for the hypothesis tests (Brooks, 2014).
First, in line with Uysal (2011), this paper controls for the acquirer’s accounting performance. Accounting performance is computed as a company’s EBITDA over total assets. Harford (1999) finds that better performing companies experience significantly negative announcement returns because these companies have higher cash reserves and are more likely to make inferior acquisition decisions. Furthermore, informational asymmetries concerning a foreign asset’s payoff make it unable for firms to finance such assets fully with externally obtained finance (Froot and Stein, 1991). Firms acquiring targets in more corrupt environments are therefore likely to finance acquisitions with excess cash reserves as managers will avoid costly external finance due to market imperfections and informational asymmetries (Demirgüc-Kunt and Maksimovic, 1998; Harford, 1999).
19 (2005) find that small and medium-sized firms are particularly constrained by obstacles related to corruption and small firms benefit the most from a reduction in the level of corruption. Large firms’ announcement returns may outperform those of small firms due the level of corruption in the target’s host country. Firm size is measured as the natural log of a firm’s total sales of the year of the announcement of a cross-border merger or acquisition.
Third, a target country’s economic conditions, measured as gross domestic product (GDP) growth, is controlled for. Kiymaz (2004) found statistically significant evidence that improving economic conditions negatively affect the wealth effects of a cross-border M&A announcement. High GDP growth in a foreign country would signal positive growth of the economy. This is perceived as an opportunity for a new market by a foreign acquirer. However, Kiymaz (2004) argues that this market potential would increase negotiating power of target firms leading to higher premiums to be paid by the acquirers, which in turn negatively affects the acquirer’s announcement returns. Another explanation for overpayment of a target firm is that acquirers are more optimistic about the future potential of the acquisition and, in turn, are willing to pay a higher premium. These situations are likely to be the case in developing countries where economic conditions are improving but where corruption still remains an important issue. As described in the literature review, Lui (1985) and Egger and Winner (2005) positively relate corruption to economic growth.
Fourth, a target country variable, which is rather new and not included as a control variable in announcement return literature yet, is the level of capital controls. Capital controls relate to the extent to which there are constraints on foreign ownership and investment restrictions. Wei (2000b) positively correlates corruption and a country’s capital restrictions. This is based on the finding by Kaufmann and Wei (2000), who found that firms which face more bureaucratic intrusion pay more bribes. Barbopoulos, Paudyal, and Pescetto (2012) find that high capital controls result in higher abnormal returns. Acquiring firms which face high capital controls benefit from less competition in the market because market entry is likely to be restricted and costly. The potential value of the acquisition is therefore higher in countries with stringent capital controls (Barbopoulos et al., 2012).
Lastly, in line with Weitzel and Berns (2006), this paper controls for industrial
diversification. When the acquiring firm is in a different industry than the target, the acquiring
20 unfamiliarity will be larger when corruption distance is higher, industry relatedness may overcome some of the cultural or organizational barriers raised by higher corruption (Di Guardo et al., 2016). Moeller and Schlingemann (2005) find significant lower announcement returns for companies acquiring a target in a different industry.
Table 1 provides an overview of the variables included in this paper.
Variable Measurement Description
CAR Cumulative abnormal return Abnormal returns on acquirer's stock
Corruption distance Natural logarithm
Absolute difference in corruption index between home and target country
Accounting performance Ratio Measured as acquirer's EBITDA over total assets
Firm size Natural logarithm Acquirer's total sales
Economic conditions Percentage
Target country GDP growth rate one year prior to the announcement date
Capital controls Rating
Rating ranging from 1 (high capital controls) to 10 (low capital controls)
Diversifying Dummy variable 0 for same industry, 1 for different industry
21
4.
Results
4.1 Descriptive statistics
Table 2 provides the descriptive statistics of the full sample and the different subsamples. The table includes statistics which relate to the number of observations, the mean, the median, and the standard deviation. In order to limit the influence of potential outliers, the data is winsorized at 98% level. This means that the upper and bottom 1% values, the tails of the data, are replaced by less extreme values derived from percentile values from each end.
Focusing on the different event windows, on average there are positive market reactions resulting from the announcement of a cross-border M&A. This is in line with the positive stance of literature which state that firms can receive abnormal returns by announcing an international M&A. Average abnormal returns are for the three-day event windows the lowest compared to the other event windows. Three-day abnormal returns are insignificant for developing acquirers with developing targets and are the largest (0.6%) for industrialized acquirers with industrialized targets. The five-day and seven-day event periods show similar patterns, with industrialized acquirers which acquire industrial target firms on average realizing the largest returns (0.7% and 0.9% respectively) and developing acquirers which acquire developing target firms experiencing the lowest abnormal returns (0.3% and 0.2% respectively).
22 Table 2 – Descriptive statistics
Full sample Industrialized - Industrialized Industrialized - Developing
N Mean Median StDev N Mean Median StDev N Mean Median StDev
CAR [-1,+1] 2,045 0.005 0.002 0.046 1,382 0.006 0.002 0.047 310 0.002 0.001 0.030 CAR [-2,+2] 2,045 0.006 0.003 0.052 1,382 0.007 0.004 0.054 310 0.004 0.000 0.044 CAR [-3,+3] 2,045 0.008 0.005 0.057 1,382 0.009 0.006 0.059 310 0.008 0.003 0.047 CorrupDist 2,045 0.552 0.519 0.167 1,382 0.477 0.477 0.107 310 0.749 0.778 0.134 AccPerf 2,045 0.134 0.129 0.074 1,382 0.133 0.130 0.074 310 0.120 0.120 0.065 FirmSize 2,045 6.968 6.999 0.625 1,382 6.993 7.017 0.617 310 7.182 7.204 0.564 EconCondi 2,045 0.028 0.026 0.028 1,382 0.022 0.024 0.020 310 0.050 0.048 0.040 CapControl 2,045 7.332 7.283 1.242 1,382 7.631 7.413 1.117 310 6.422 6.401 1.308 Diversifying 2,045 0.318 0.000 0.466 1,382 0.350 0.000 0.477 310 0.261 0.000 0.440
Developing - Developing Developing - Industrialized
23 4.2 Multicollinearity
Table 3 provides the correlation matrix including all variables used in this study. The correlations between variables are examined to test for multicollinearity. The high correlation coefficients among the three event windows are logical as the seven-day event window consists of data of both the three-day event window and the five-day event window. These high correlations are not a problem, because these event windows are not included in a model simultaneously. The other variables do not show high collinearity coefficients. The highest correlations can be found in corruption distance (CorruptionDist) and capital controls (CapitalControl), -0.296, and corruption distance and target country’s economic conditions (EconCondi), 0.219. These scores are not high enough to detect multicollinearity. Therefore, no variables have to be excluded for further analysis due to problems of multicollinearity.
4.3 Heteroscedasticity
24 Table 3 – Correlation coefficients
25 4.4 Regression results
Table 4 provides the model specification. Table 5 to 7 show the results of the multivariate regressions for the three-day, five-day, and seven-day event windows respectively. Model 1 examines the full sample. The coefficient is negative for all event windows, as expected, but the results are insignificant. Hypothesis 1 is therefore not supported. The control variables show coefficients in line with the literature, with significant coefficients for acquirer’s firm size and target country’s economic conditions. For the seven-day event window, industry diversification is significant and negatively related with acquirer’s abnormal returns.
Model 2 examines the relationship between corruption distance and acquirer’s abnormal returns for the first subsample, which consists of industrialized acquirers with industrialized target firms. Hypothesis 2 did not expect a large impact of the negative effect due the fact that industrialized acquirers which internationalize towards another industrialized country deal with a similar environment. As expected, the coefficients are particularly low and negative for the event windows. The results are, however, insignificant. Hypothesis 2 is therefore not supported. Again, the control variables mainly show expected signs and firm size is significantly related to lowering abnormal returns for larger acquirers.
Hypothesis 3 is tested by model 3, which considers the subsample of industrialized acquirers with developing targets. The results are insignificant and, therefore, hypothesis 3 is not supported. Interestingly, the coefficients for corruption distance in the different event windows do not have the expected negative signs. For all event windows, the coefficients indicate a positive relationship between corruption distance and acquirer’s abnormal returns. Accounting performance has, contrary to the previous models, a positive coefficient which can be expected as acquirers would want to finance deals with excess cash to avoid costly external finance. The coefficients are significant for the five-day and seven-day event periods.
26 CAR [-1,+1] Intercept 0.011*** 0.012*** 0.048*** 0.048*** 0.053*** 0.053*** (2.940) (2.673) (2.856) (2.855) (2.843) (2.829) CorruptionDistance -0.011* -0.011* -0.011* -0.008 -0.010 -0.010 (-1.901) (-1.882) (1.931) (1.440) (-1.591) (-1.597) AccPerformance -0.011 -0.012 -0.011 -0.011 -0.010 (-0.798) (-0.825) (-0.785) (-0.746) (-0.728) FirmSize -0.005** -0.005** -0.005** -0.005** (-2.523) (-2.491) (-2.533) (-2.508) EconCondition -0.067** -0.065** -0.071** (-2.131) (-2.103) (-2.137) CapitalControl -0.001 -0.001 (-0.675) (-0.633) Diversifying -0.001 (-0.476) N 2,045 2,045 2,045 2,045 2,045 2,045
27 CAR [-1,+1]
Model 1 Model 2 Model 3 Model 4 Model 5
Intercept 0.053*** 0.061*** 0.024 -0.007 0.031 (2.829) (2.582) (0.832) (-0.172) (0.631) CorruptionDistance -0.010 -0.003 0.003 -0.020 0.009 (-1.597) (-0.252) (0.189) (-1.216) (0.298) AccPerformance -0.010 -0.015 0.034 0.044 -0.054 (-0.728) (-0.855) (1.368) (0.992) (-1.396) FirmSize -0.005** -0.006** -0.004 0.000 -0.002 (-2.508) (-2.376 (-0.911) (-0.018) (-0.413) EconCondition -0.071** -0.037 -0.058 -0.073 -0.042 (-2.137) (-0.544) (-1.392) (-0.925) (-0.313) CapitalControl -0.001 -0.001 0.000 0.003 -0.002 (-0.633) (-0.924) (0.148) (1.416) (-0.397) Diversifying -0.001 0.001 -0.005 -0.010 -0.004 (-0.476) (0.513) (-1.374) (-1.276) (-0.612) N 2,045 1,382 310 153 200
28 CAR [-2,+2]
Model 1 Model 2 Model 3 Model 4 Model 5
Intercept 0.074*** 0.093*** 0.001 0.026 0.024 (3.673) (3.550) (0.032) (0.582) (0.405) CorruptionDistance -0.009 -0.003 0.017 -0.024 -0.001 (-1.235) (-0.214) (0.739) (-1.341) (-0.039) AccPerformance -0.015 -0.018 0.082** -0.040 -0.060 (-0.949) (-0.871) (1.941) (-0.840) (-1.408) FirmSize -0.008*** -0.010*** -0.004 -0.004 -0.002 (-3.609) (-3.548) (-0.797) (-0.717) (-0.270) EconCondition -0.071* -0.058 -0.041 -0.098 0.025 (-1.919) (-0.782) (-0.656) (-1.141) (0.129) CapitalControl 0.000 -0.001 0.002 0.005** 0.000 (-0.409) (-0.805) (0.646) (2.160) (-0.018) Diversifying -0.004 -0.003 -0.003 -0.011 -0.004 (-1.540) (-0.985) (-0.562) (-1.446) (-0.444) N 2,045 1,382 310 153 200
29 CAR [-3,+3]
Model 1 Model 2 Model 3 Model 4 Model 5
Intercept 0.085*** 0.110*** 0.049 0.015 -0.012 (3.963) (3.897) (1.048) (0.312) (-0.181) CorruptionDistance -0.009 -0.001 0.000 -0.019 -0.027 (-1.204) (-0.031) (-0.006) (-0.931) (-0.678) AccPerformance -0.006 -0.004 0.104*** -0.016 -0.054 (-0.323) (-0.169) (2.738) (-0.313) (-1.142) FirmSize -0.009*** -0.013*** -0.007 -0.002 0.000 (-3.929) (-4.100) (-1.364) (-0.311) (-0.014) EconCondition -0.044 -0.032 -0.018 -0.118 0.020 (-1.085) (-0.380) (-0.274) (-1.251) (0.117) CapitalControl -0.001 -0.001 -0.001 0.003 0.006 (-0.452) (-0.752) (-0.150) (1.305) (0.822) Diversifying -0.005** -0.006* -0.000 -0.010 -0.010 (-2.063) (-1.731) (-0.049) (-1.113) (-0.891) N 2,045 1,382 310 153 200
30 Finally, model 5 considers the subsample of developing acquirers which announced the merger or acquisition of a target in an industrialized country. The different event windows provide mixed results. Only the three-day period shows the expected positive relationship. The five-day and seven-day event windows provide a negative relationship. Hypothesis 5 is not supported by the results. The five-day and seven-day event windows show positive coefficients for target country’s economic conditions and capital controls. Although insignificant, this indicates that developing acquirers tend to benefit from well-developed environments when acquiring a target in an industrialized country.
4.5 Robustness check
In line with the UNCTAD (2016) country classifications, Singapore and Hong Kong are grouped as developing countries. Firms from these countries are thus included in the regression models as either developing acquirer or developing target. However, these countries are characterized by low corruption scores. Singapore has been in the top 10 of the CPI by Transparency International for over a decade, with a number 1 ranking in 2010. Similarly, Hong Kong has perceived corruption scores which are similar to countries such as Norway, Switzerland, and Luxembourg, which are all considered to be low in corruption. Singapore and Hong Kong are two of the four original ‘Asian Tiger’ economies. Between 1965 and 1995, these countries experienced average economic growth rates of 6% annually. Since the 1960s, Singapore’s GDP increased to a level that equals New Zealand and Australia and as off 2005 has higher per capita GDP than these countries (Baten, 2016). Without going into further details, Singapore and Hong Kong are renowned for their strong economic fundamentals and prudent macroeconomic policies which are comparable to those of most industrialized countries. The results of the regression estimations might be influenced by industrialized-like variables of these two countries. Therefore, to check the robustness of the results, the regressions of model 3 to 5 are repeated excluding the deals involving acquirers and targets from Singapore and Hong Kong. The results of this robustness check can be found in model 6 to 8 in table 8 in Appendix B.
31
5.
Discussion and conclusions
This paper has examined the impact of corruption, measured as corruption distance, on acquirer’s abnormal returns when announcing a cross-border M&A. Based on the psychic distance concept from internationalization theory, it was expected that higher corruption distance between an acquirer’s home country and a target’s host country would negatively affect the acquirer’s abnormal announcement returns. More similarity between home and host countries would reduce uncertainty and costs related to asymmetries arising from corruption differences. A sample of 2,045 cross-border M&A deals was used to empirically examine the effect of corruption while controlling for firm-, country-, and transaction-level variables. Three-day, five-day and seven-day event periods were included to not restrict the information flows to a three-day event period solely, as previous literature does not provide an exact definition how long the short-lived lags of information into stock prices are.
32 One unexpected relationship was found in the regression results. Corruption distance is positively related to acquirer’s abnormal returns when industrialized firms acquire a target in a developing host country. Despite the fact that industrialized acquirers are expected not to have the expertise and experience in how to handle corrupt foreign environments (Habib and Zurawicki, 2002), the positive relationships in all three event periods, although insignificant, indicate that these acquirers tend to have higher abnormal returns when acquiring targets in more corrupt countries. This may be explained by the wealth effect of Froot and Stein (1991), who state that firms from wealthier countries with stronger currencies tend to acquire firms from poorer countries. As valuation differences occur due to wealth differences between countries, lower cost of capital can be reached by the acquirer when acquiring a foreign firm in a poorer country compared to acquiring a domestic firm. The positive coefficients might also be explained by the aforementioned findings of Egger and Winner (2006).
The findings in this paper provide implications for managers of firms who are seeking to expand their operations globally through mergers or acquisitions and for policy makers. Closeness in corruption, which means that home country corruption is comparable to host country corruption, not necessarily leads to higher announcement returns. Corruption thus needs to be considered relative to other important dimensions, such as target country’s economic conditions, affecting a firm’s international operations and related wealth effects. This will support managers in refining country selection procedures when seeking for international opportunities. It is likely that, as argued by Wheeler and Mody (1992) and Egger and Winner (2006), classical factors such as infrastructure development and factor endowments are more important to a firm’s international success than a target countries’ level of corruption. For policy makers, despite the OECD Convention of 1997, corruption remains a problem which is difficult to handle. Corruption does not significantly deter a firm’s short-term wealth effects and firms are therefore incentivized to participate in corrupt practices as they may benefit from it when operating in corrupt environments. In certain environments, corruption might be part of the way business is conducted and cannot be controlled by conventions on a political level in countries where FDI outflows mainly originate.
33 paper did not include acquirer’s previous experience in a corrupt environment. Acquirers might have acquired a target in the host country before or a target in a country with similar corruption levels and may have gained expertise how the handle high corruption to its benefit. Furthermore, the main definition of corruption focuses on illegal corruption, whereas legal corruption may be more prevalent and more frequently ignored (Kaufmann, 2004). Lastly, the sample is unbalanced because it mainly consists of firms from industrialized countries.
34
6.
References
Ades, A. and Di Tella, R. 1999. Rents, competition, and corruption. American Economic Review, 89(4): 982-993.
Agmon, T. and Lessard, D. 1977. Investor recognition of corporate international diversification. The Journal of Finance, 32(04): 1049-1055.
Ahern, K., Daminelli, D. and Fracassi, C. 2015. Lost in translation? The effect of cultural values on mergers around the world. Journal of Financial Economics, 117(1): 165-189.
Barbopoulos, L., Paudyal, K. and Pescetto, G. 2012. Legal systems and gains from cross-border acquisitions. Journal of Business Research, 65: 1301-1312.
Bardhan, P. 1997. Corruption and development: A review of issues. Journal of Economic Literature, 35(3): 1320-1346.
Baten, J. 2016. A history of the global economy. Cambridge: Cambridge University Press.
Beck, T., Demirgüc-Kunt, A. and Maksimovic, V. 2005. Financial and legal constraints to growth: Does firm size matter? Journal of Finance, 60(1): 137-177.
Blonigen, B.A. 2005. A review of the empirical literature on FDI determinants. Atlantic Economic Journal, 33: 383-403.
Bodnar, G., Tang, C. and Weintrop, J. 1997. Both sides of corporate diversification: The value impacts of geographic and industrial diversification. NBER Working paper no. 6224, National Bureau of Economic Research, Cambridge, MA.
Brooks, C. 2014. Introductory econometrics for finance. Cambridge: Cambridge University Press.
Brown, S. and Warner, J. 1985. Using daily stock returns: The case of event studies. Journal of Financial Economics, 14(1): 3-31.
Cuervo-Cazurra, A. 2006. Who cares about corruption? Journal of International Business Studies, 37: 807-822.
Cuervo-Cazurra, A. 2016. Corruption in international business. Journal of World Business, 51: 35-49.
Cuervo-Cazurra, A. and Genc, M. 2008. Transforming disadvantages into advantages: Developing-country MNEs in the least developed countries. Journal of International Business Studies, 29: 957-979.
Demirbag, M., Glaister, K. and Tatoglu, E. 2007. Institutional and transaction cost influences on MNEs’ ownership strategies of their affiliates: Evidence from an emerging market. Journal of World Business, 42: 418-434.
Demirgüc-Kunt, A. and Maksimovic, V. 1998. Law, finance, and firm growth. Journal of Finance, 53: 2107-2137.
35 Doukas, J. and Travlos, N. 1988. The effect of corporate multinationalism on shareholders’ wealth: Evidence from international acquisitions. The Journal of Finance, 43(5): 1161-1175.
Dyckman, T., Philbrick, D. and Stephan, J. 1984. A comparison of event study methodologies using daily stock returns: A simulation approach. Journal of Accounting Research, 22: 2-30.
Erel, I., Liao, R. and Weisbach, M. 2012. Determinants of cross-border mergers and acquisitions. Journal of Finance, 67(3): 1045-1082.
Egger, P. and Winner, H. 2005. Evidence on corruption as an incentive for foreign direct investment. European Journal of Political Economy, 21(4): 932-952.
Egger, P. and Winner, H. 2006. How corruption influences foreign direct investment: A panel data study. Economic Development and Cultural Change, 54(2): 459-486.
Fama, E. 1998. Market efficiency, long-term returns, and behavioral finance. Journal of Financial Economics, 49: 283-306.
Faccio, M., McConnell, J. and Stolin, D. 2006. Returns to acquirers of listed and unlisted targets. The Journal of Financial and Quantitative Analysis, 41(1): 197-220.
Francis, B., Hasan, I., Sun, X. and Waisman, M. 2014. Can firms learn by observing? Evidence from cross-border M&As. Journal of Corporate Finance, 25: 202-215.
Froot, K. and Stein, J. 1991. Exchange rates and foreign direct investment – an imperfect capital-markets approach. Quarterly Journal of Economics, 106: 1191-1217.
Fuller, K., Netter, J. and Stegemoller, M. 2002. What do returns to acquiring firms tells us? Evidence from firms that make many acquisitions. Journal of Finance, 57: 1763-1793.
Habib, M. and Zurawicki, L. 2001. Country-level investments and the effect of corruption – some empirical evidence. International Business Review, 10: 687-700.
Habib, M. and Zurawicki, L. 2002. Corruption and foreign direct investment. Journal of International Business Studies, 33(2): 291-307.
Harford, J. 1999. Corporate cash reserves and acquisitions. The Journal of Finance, 54(6): 1969-1997.
Hair, J., Black, W., Babin, B. and Anderson, R. 2009. Multivariate data analysis. Upper Saddle River, New Jersey: Pearson Education.
Henisz, W. 2000. The institutional environment for multinational investment. Journal of Law, Economics and Organization, 16(2): 334-364.
Hines, J. 1995. Forbidden payment: foreign bribery and American business after 1977. NBER Working Paper No. 5266. National Bureau of Economic Research: Cambridge, MA.
Huntington, S. 1968. Political order in changing societies. London: Yale University Press
36 Kaufmann, D. 2004. Corruption, governance and security: Challenges for the rich countries in the world. In M.E. Porter, K. Schwab, X. Sala-i-Martin and A. Lopez-Carlos (eds.), The World Economic Forum, Global Competitiveness Report 2004-2005: 83-102. New York: Palgrave Macmillan.
Kaufmann, D. 2005. Myths and realities of governance and corruption. In A. Lopez-Carlos, M.E. Porter and K. Schwab (eds.), The World Economic Forum, Global Competitiveness Report 2005-2006: 81-98. New York: Palgrave Macmillan.
Kaufmann, D. and Wei, S. 2000. Does ‘grease money’ speed up the wheels of commerce? IMF Working Paper WP/00/64. International Monetary Fund.
Kiymaz, H. 2004. Cross-border acquisitions of US financial institutions: Impact of macroeconomic factors. Journal of Banking and Finance, 28: 1413-1439.
Klitgaard, R. 1991. Gifts and bribes. In J. Zeckhauser, ed. Strategy and choice. The MIT Press, Cambridge, MA.
Kogut, B. 1983. Foreign direct investment as a sequential process. In C. Kindleberger and D. Andretsch The Multinational Corporation in the 1980s. The MIT Press, Cambridge, MA. 38-56. Lang, L., Stulz, R. and Walkling, R. 1989. Managerial performance, Tobin’s q and the gains from successful tender offers. Journal of Financial Economics, 24(1): 137-154.
LaPalombara, J. 1994. Structural and institutional aspects of corruption. Social Research, 61(2): 325-350.
Lee, C. and Ng, D. 2009. Corruption and international valuation: Does virtue pay? The Journal of Investing, 18(4): 23-41.
Leff, N. 1964. Economic development through bureaucratic corruption. American Behavioral Scientist: 8-14.
Lui, F. 1985. An equilibrium queuing model of bribery. Journal of Political Economy, 93(4): 760-781.
MacKinlay, A. 1997. Event studies in economics and finance. Journal of Economic Literature, 35(1): 13-39.
Mauro, P. 1995. Corruption and growth. Quarterly Journal of Economics, 110(3): 681-712.
Malhotra, S., Zhu, P. and Locander, W. 2010. Impact of host-country corruption on U.S. and Chinese cross-border acquisitions. Thunderbird International Business Review, 52(6): 491-507.
Moeller, S. and Schlingemann, F. 2005. Global diversification and bidder gains: A comparison between cross-border and domestic acquisitions. Journal of Banking and Finance, 29: 533-564.
Moeller, S., Schlingemann, F. and Stulz, R. 2004. Firm size and the gains from acquisitions. Journal of Financial Economics, 73: 201-228.
Myerson, R. 1982. Optimal coordination mechanisms in generalized principal-agent problems. Journal of Mathematical Economics, 10(1): 67-81.
37 Qian, X. and Sandoval-Hernandez, J. 2016. Corruption distance and foreign direct investment. Emerging Markets Finance & Trade, 52: 400-419.
Rose-Ackerman, S. 1978. Corruption. A study in political economy. New York: NY Academic Press.
Smarzynska, B. and Wei, S. 2000. Corruption and the composition of foreign direct investment: Firm-level evidence. NBER Working Paper No. 7969. National Bureau of Economic Research: Cambridge, MA.
Svensson, J. 2003. Who must pay bribes and how much? Evidence from a cross-section of firms. Quartely Journal of Economics, 118: 207-229.
Tanzi, V. 1998. Corruption around the world: Causes, consequences, scope, and cures. Working Paper WP/98/63. International Monetary Fund, Washington, D.C.
Tanzi, V. and Davoodi, H. 2000. Corruption, growth, and public finances. Working Paper No. 2000-2182. International Monetary Fund.
UNCTAD. 2000. World Investment Report. UNCTAD: New York and Geneva. http://unctad.org/en/docs/wir2000_en.pdf; Accessed February 10, 2017
UNCTAD. 2001. World Investment Report. UNCTAD: New York and Geneva. http://unctad.org/en/docs/wir2001overview_en.pdf; Accessed April 17, 2017
UNCTAD. 2007. World Investment Report. UNCTAD: New York and Geneva. http://unctad.org/en/docs/wir2007_en.pdf; Accessed April 18, 2017
UNCTAD. 2016. Economic Groupings.
http://unctadstat.unctad.org/EN/Classifications/DimCountries_DevelopmentStatus_Hierarchy.pdf; Accessed February 14, 2017
Uysal, V. 2011. Deviation from the target capital structure and acquisition choices. Journal of Financial Economics, 3: 602-620.
Wei, S. 2000a. How taxing is corruption on international investors? Review of Economics and Statistics, 82(1): 1-11.
Wei, S. 2000b. Local corruption and global capital flows. Brookings Papers on Economic Activity, 2: 303-354.
Weitzel, U. and Berns, S. 2006. Cross-border takeovers, corruption, and related aspects of governance. Journal of International Business Studies, 36(6): 786-806.
Wheeler, D. and Mody, A. 1992. International investment location decisions: The case of U.S. firms. Journal of International Economics, 33: 57-76.
World Bank. 1997. World Development Report 1997: The State In a Changing World. New York: Oxford University Press.
World Bank. 2017. Enterprise Surveys.
http://www.enterprisesurveys.org/data/exploretopics/corruption#2; Accessed April 30, 2017
38
Appendix A
39
Appendix B
CAR [-1,+1]
Model 6 Model 7 Model 8
Intercept 0.024 -0.016 0.012 (0.796) (-0.308) (0.234) CorruptionDistance 0.004 -0.014 -0.018 (0.274) (-0.563) (-0.548) AccPerformance 0.036 0.005 -0.060 (1.417) (0.096) (-1.612) FirmSize -0.003 0.001 0.000 (-0.792) (0.089) (0.022) EconCondition -0.084* -0.053 0.001 (-1.818) (-0.589) (0.006) CapitalControl 0.000 0.004 0.002 (-0.101) (1.596) (0.421) Diversifying -0.004 -0.010 -0.004 (-1.035) (-1.128) (-0.436) N 295 116 170 CAR [-2,+2]
Model 6 Model 7 Model 8
40 CAR [-3,+3]
Model 6 Model 7 Model 8
Intercept 0.056 0.017 -0.031 (1.158) (0.274) (-0.465) CorruptionDistance -0.004 -0.025 -0.011 (-0.148) (-0.880) (-0.253) AccPerformance 0.102** -0.040 -0.053 (2.305) (-0.619) (-1.069) FirmSize -0.007 -0.002 0.000 (-1.405) (-0.308) (-0.023) EconCondition -0.041 -0.167 0.007 (-0.545) (-1.529) (0.034) CapitalControl -0.001 0.005 0.007 (-0.232) (1.648) (1.251) Diversifying 0.000 -0.013 -0.009 (0.022) (-1.162) (-0.862) N 295 116 170
Table 8 – Robustness check results. The dependent variable is the three-day, five-day or seven-day cumulative
abnormal return (CAR). The independent variable CorruptionDistance is the natural logarithm of the absolute
