Corruption and overvalued M&A’s
University of Groningen Faculty of Economics and Business MSc International Financial Management
Master’s Thesis IFM By
THIJS MOOIJ*
June 2018
ABSTRACT
This paper explores the relationship of corruption on the overvaluation of acquiring firms during a merger and acquisition. Furthermore, the effect of corporate governance on this relationship is examined. The final sample consists of 5,892 deals and 1,156 firms in a total of 19 countries during the period 2002-2017. Evidence is found that corruption has a positive effect on the overvaluation of acquiring firms during an M&A, no significant effects are measured that corporate governance weakens this effect. This research is relevant for international studies and shareholders, while overvaluation can harm their investments.
Field keywords: Overvaluation, M&A’s, Corruption, Corporate governance JEL classification: D73, G32, G34, G15
2
1. Introduction
Finding synergies with other companies to grow, increase market power, boost profitability and improve shareholder’s wealth are the primary objectives in merger and acquisitions (M&A) according to Alexandridis, Petmezas, and Travlos (2010). Moreover M&A’s should establish positive net present value projects. However, while one of the main objectives of firm management is to improve shareholders wealth, this will not always be the case.
This study attempts to find out how it occurs that firm management not always behave for the benefit of the shareholder. This paper examines how corrupted environment effects this behavior and whether acquirers are more overvalued during an M&A in this environment. The latter is relevant because during an M&A there seem to be different interests for both governments as managers and shareholders. Finally, I find out whether certain corporate governance practices can weaken this likely positive effect of corruption on the overvaluation of the acquiring firm after an M&A.
It is assumed that managers do not necessarily act for the benefit of shareholders. Jensen (1976) state that the relationship between stockholders and managers of a firm fit the definition of an agency relationship. Within this relationship, the stockholder (principal) engage the manager (agent) to perform a specific task or service on their behalf. However, it is impossible to ensure that the agent will make the best decision in the point of view of the principal. Also, it is not a surprise that the separation of ownership and control in corporations is associated with the agency problem. The cost that is made by the principal to monitoring expenditures, the bonding costs of the agent and the residual loss are together referred as agency cost.
Jensen (2005) stated that equity overvaluation thrives agency costs and that less convincing acquisitions, driven by stock overvaluation, is reflected by these agency costs. Overvalued stock prices imply that a firm is not able to deliver the expected performance. Managers are not able to meet unachievable expectations which makes them take riskier actions, and accounting manipulation might increase. If managers are not able to obtain the required earnings for the firm, expenses are pushed forward, and revenues are taken into the current period. This can eventually result in fraudulent practices and mislead the firm and the stakeholders. All this can cause value destruction which could harm the shareholders’ value. Therefore, overvaluation of M&A’s is relevant information for shareholders
3
report that managers are driven to maximize their profits, maintain their position, and gain personal benefits. According to the literature, the motives of undertaking M&A’s consist of synergy, agency, and hubris. As mentioned, Alexandridis, Petmezas, and Travlos (2010) stated that finding synergies with other companies to grow, increase market power, boost profitability and improve shareholder’s wealth is the main object in mergers and acquisition. Jensen (2005) suggest that agency cost of overvalued equity can cause value-destroying M&A’s while managers try to meet the growth expectations by engaging in overvalued M&A’s instead of eliminating overvaluation. Regarding the hubris hypothesis, Roll (1986) suggests that managers engage in acquisitions even when there are no synergies to gain. They make irrational decisions and end up paying too much for their targets while they are overconfident about their ability to gain from a takeover.
Because M&A is not only a major business transaction for management, it is also a motive and opportunity for governments to apply a certain influence on specific M&A’s (Brockman, Rui, and Zou, 2013). Some governments encourage certain domestic mergers and acquisition while discouraging other, mostly foreign M&A’s (Serdar Dinc and Erel, 2013). Also, Claessens, Feijen, and Laeven (2008) argue that politically connected firms have easier access to bank financing. This easy access to financing could be unfavorable for firms when Jensen (1986) state that a firm is more likely to undertake low-benefit or even value-destroying mergers and thrives agency costs.
Corruption, defined by Cuervo-Cazurra (2006), is the abuse of power for private gain. And with both personal benefits as government motives that are present in specific M&A’s, it is obvious to consider corruption in this description. It is convincing to say that corruption influences the M&A process and therefore the valuation of firms. Whereas these actions benefit others than the firm itself, and this might not always be noticeable for the shareholders. Firms might be valued for a price that does not equal their performance, while on the background managers have motives that are not for the benefit of the firm and/or shareholders. This is eventually bad for the shareholder's investments.
4
In the first part of the methodology, the univariate analysis is examined. The mean and median of the firms’ abnormal returns are compared between high – and low corrupted countries. Moreover, there is a distinction between short- and long-term performance after an M&A. In the second part, the multivariate analysis is examined. This paper allows to show the results of the regression with corruption as the main dependent variable and performance as the independent variable. Later, corporate governance variables and interaction variables are included. While most papers focus on one specific corporate governance measurement (La Porta, Lopez-de-Silanes, Shleifer, and Vishny, 2000; Fu and Officer, 2013; Donadelli, Fasan, Magnanelli, 2014), this paper takes all mechanisms into account.
The findings of this paper are that firms in high corrupted countries showed high prior returns and low post-merger returns. Moreover, post-merger returns are negatively related to corruptions. This indicates acquiring firms are indeed more overvalued in high corrupted countries. Besides, in line with Dinc and Erel (2013), I found that M&As are engaged more in domestic countries, which could indicate of governments interference. Lastly, moderating effects are measured for corporate governance variables on corruption. However, for the latter, no evidence is found.
2. Literature review and hypothesis development
Corruption and M&A’sCorruption, defined by Cuervo-Cazurra (2006), is the abuse of power for private gain. While corruption receives an increasing amount of attention, studies find, in general, a negative impact. Mauro (1995) find that corruption lowers economic growth due to lower investment. Cuervo-Cazurra (2006) supports this by finding evidence that corruption results in lower foreign direct investments. Companies see corruption as a cost and therefore a reason not to invest (Wei, 2000). Still, some investors are willing to pay a certain amount because they value this asset more than the cost it will bring with (Lui, 1985).
5
While M&A is not only a major business transaction for management, it is also a motive and opportunity for governments to apply a certain influence on specific M&A’s. (Brockman, Rui, and Zou, 2013) Some governments encourage certain domestic mergers and acquisition while discouraging other, mostly foreign M&A’s. (Serdar Dinc and Erel, 2013). Moreover, Brockman, Rui, and Zou (2013) stated that governments often interfere in M&A’s because they are afraid of unemployment during and after the restructuring process and they suggest the government has the power as well to control the process of an M&A transaction.
While the main objectives of an M&A is to find synergy gains, increase of market power, profitability boosts and improve shareholders ‘wealth (Alexandridis, Petmezas, and Traylos, 2010), these objectives will be threatened if the government influences the process of M&A’s, because these objectives are seen from a firm’s perspective. Also, manager of firm’s might have different interests as Johnson (2009) stated (maximizing profits, maintaining their position and gain personal benefits). However, these personal gains do not have to be beneficial for the firm per see. Thus, it would be valid to say corrupt actions occur during M&A’s in the benefit of the government and not the shareholders of the firm itself.
Overvaluation
Equity is overvalued when a firm’s stock price is higher than its intrinsic value (Jensen, 2005) Chi and Gupta (2009) stated that an overvalued firm cannot justify the market value due to the insufficient positive net present value investment opportunities they have. The subsequent decrease in acquirer’s stock might be triggered by overpayment or the missing synergies or even both (Fu, Lin, and Officer, 2013).
6
Rhodes-Kropf and Vishwanathan (2005) support the misvaluation theories based on behavioral explanations or asymmetric information between otherwise rational manager and markets. They explain that misvaluation affects the buyers and the way the transaction is paid. Mainly they confirm that acquirers that use stock are more overvalued than their targets before they announce the M&A. Furthermore, the overvaluation of equity increases the probability that a firm is the bidder using stock as payments.
Weitzel and Berns (2006) found that high level of corruption in host countries is associated with low target premiums. While this implies that the valuation discount is based on the higher uncertainty in corrupt countries, it might as well show signs of an overvalued target. Also, Fu, Lin, and Officer (2013) report that overvalued acquirers are related to weak corporate governance, and an increasing option-based compensation for CEO’s appears to be a motive for these acquisitions. The increase for the CEO’s compensation is often higher than the decrease in the CEO’s equity holding. Chi and Gupta (2009) state that overvaluation is related to income-increasing earnings management, while overvaluation intensifies these earning management.
Considering that the governments have motives for interfering with M&A’s it is sensible to say that the government could contribute to specific income-increasing activities. Moreover, while managers motives can be irrational (i.e. hubris hypothesis (Roll, 1986)) and not always in favor of the firm, the following hypothesis is formulated:
H1: Corruption is positively related to the overvaluation of acquiring firms during a merger or
acquisition
Abnormal returns
Following Barber and Lyon (1997) I calculate the abnormal returns to measure the return of firms that are different from the expected rate of returns. For short-term performance, I look at the cumulative abnormal returns with a 3-month window (CAR).
7
little importance in a short time span such as a few days. However, it can be of great importance if long-term returns are used.
So, it is understandable to ask why it is useful to use the BHAR approach to measure long-term returns, while there are alternatives available. For instance, the calendar time portfolio approach. However, Lyon, Barber, and Tsai (1999) found that the calendar time approach has misspecified test statistics in non-random samples, meaning it could be biased. The BHAR approach turned out to be relatively robust. Moreover, Barber and Lyon (1997) favor the use of BHAR in predicting long-term returns, while the BHAR measures the underlying parameter of interest. This is the long-run performance of the common stock of the sample firms compared to a decent reference group.
Shleifer and Vishny (2003) state that the market misvaluation theory predicts that firms that are overvalued will have a high pre-merger return and abnormal negative returns after the merger. This is because the market corrects overvaluation. In the same way, undervaluation occurs when firms are being valued less than they should, resulting in higher post-merger return. This happens while firms perform better than the shareholders expect them to do, and the market errs in its initial response (Ikenberry, Lakonishok, and Vermaelen, 1995)
Corporate governance
Literature finds that the quality of corporate governance appears to be positively related to the value creation from M&A’s (Alexandridis, Antypas, and Traylos 2017; Dahya, Goluboy, Petmezas, and Traylos, 2016). Salama and Putnam (2013) found that high-quality corporate governance is associated with positive financial outcomes, while the agency theory says that effective corporate governance can improve the quality of managerial decision making. This benefits the outcomes of corporate investments. Also, Jensen (2005) concluded that monitoring by intelligent people with integrity in a sound governance system is necessary for effective control of corporate agency problems. He agrees that this might be the only solution for the agency problems caused by overvalued equity.
8
state that weaker shareholder rights increase agency costs. Basically, the weaker the shareholder rights, the more power managers have. Gompers, Ishii, and Metrick (2003) say that when the market underestimates this extra cost, the operating performance and the stock returns would have been worse than expected. What would be observed, is that the firms are overvalued at the beginning of the period. Fu, Lin, and Officer (2013) even found that overvalued acquirers are related to weak corporate governance, and an increasing option-based compensation for CEO’s appears to be a motive for these acquisitions. Finally, Donadelli, Fasan, and Magnanelli (2014) stated that the corporate governance mechanism as the size of the board and independent directors positively affect the performance of a firm in corrupt-sensitive industries.
However, Core, Guay, and Rusticus (2006) argue this, using shareholder rights as their main focus of corporate governance. They find that, despite having lower operating performance, weak governance firms also have a bit higher abnormal stock returns than companies with strong corporate governance. They explain this based on the assumption that shareholders forecast the lower operating firms and therefore not affect the abnormal returns. This is because investors expect the difference in operating performance.
Still, corporate governance has a positive effect on financial outcomes and a positive effect in corrupt sensitive environments. It is reasonable to argue that corporate governance moderates the relationship between corruption and over valuated M&A’s. Thus, the following is hypothesized:
H3: When corporate governance becomes stronger, the relationship between corruption and
overvaluation of acquiring firms during a merger or acquisition weakens.
3. Methodology
Measuring overvaluation and undervaluation
To measure over- and undervalued firms the alpha (intercept) are estimated using the market model:
Rit = αi + βi * Rmt + εit
(1)In this model, Rit is the return of the stock of observation i on day t, which in this case is the
9
it tells you how much better firm i performed than the model predicted. A negative alpha tells you how much worse it did. βi is the slope and tells you the sensitivity of Rit on the reference
market. Both the alpha as the beta are measured over a period of the last four years.
To allocate each firm in the category overvalued, fair valued, or undervalued, a distribution based on the firm’s alpha is made. Within the high corrupted sample and low corrupted sample, all firms with a negative alpha are allocated to the overvalued category, while they perform less than expected. This is considered the bottom tail which contains 493 firms in high corrupted countries and 679 firms in low corrupted countries (see table 2). For the upper tail, the same number of firms are selected, however this time with the highest alphas. These are considered undervalued and also contain 493 firms in high corrupted countries and 679 firms in low corrupted countries. The rest of the sample is considered fair valued and count respectively 1207 firms in high corrupted countries and 2340 in low corrupted countries (see table 2).
Measuring short- and long-term performance
Short-term
To measure the short-term performance, I use the abnormal return (AR) during the event and calculate the cumulative abnormal returns (CAR) over a 3-month window. Following Barber and Lyon (1997) the AR and CAR are calculated as follows:
ARit = Rit – Rmt
(2)CARit = ∑
𝑡𝑡=1AR
it (3)Where Rit is defined as the monthly stock return of firm i in month t and Rmt is the benchmark
return in month t. Firm’s return index and market index of the firm’s country are obtained from Datastream..
Long-term
10
Following Barber and Lyon (1997) pp, 369: long-run abnormal returns should be calculated ‘‘as the long-run buy-and-hold return of a sample firm less the long-run return of an appropriate benchmark, to which I refer as a buy-and-hold abnormal return.’’
The BHAR of firm i in month t is calculated using the following formula:
BHARit = ∏
𝑡𝑡=1[
1+ Rit] - ∏
𝑡𝑡=1[
1+ Rmt]
(4)
Measuring corruption
Despite that corruption is illegal and difficult to measure, there is a way to measure corruption using the Transparency International’s Corruption Perception Index. This index ranks countries based on their degree of corruption and is based on how public officials and politicians perceive these countries; this is in line with the study of Brockman, Rui, and Zou (2013). Because comparison between high and low corruption is made, a dummy variable is created. A selection of countries within the CPI scale of 80-100 are considered low corruption countries, whereas countries within the CPI scale of 0-40 are considered high corruption countries. As a dummy variable, low corrupted countries equal 0 and high corrupted countries equal 1.
Measuring corporate governance
To measure corporate governance, a composite measure (CGVS TOTAL) will be used. While DeFond, Hann, and Hu (2005) believe that a composite measurement better captures the overall environment of the firm’s governance than individual measures. However, Donadelli, Fasan and Magnanelli (2014) only found that the corporate governance mechanism as the size of the board and independent directors positively affect the performance of a firm in corrupt-sensitive industries. Moreover, La Porta, Lopez-de-Silanes, Shleifer, and Vishny (2000) state that shareholder protection is the key protection, therefore category measurements will be used as well. Measuring on category level enables to see which specific mechanisms effects our main relationship. The corporate governance quality measurement is based on board functions (CGBF), board structure (CGBS), compensation policy (CGCP), vision and strategy (CGVS) and the right of shareholders (CGSR) which are obtained from the asset 4 part of DataStream database.
Sample construction
11
should be obtained. Furthermore, only merger and acquisitions are selected while this is our focus, all other deals are excluded.
The sample got reduced to 25,197 deals after the sample got filtered for high- and low corrupted countries. In this study, I concentrate on the tails and allocate all firms with a CPI of 0-40 categorized as high corrupted and firms with a CPI of 80-100 as low corrupted countries. Moreover, every deal before 2002 got excluded, because corporate governance data is only available after 2002.
The final sample that is created to compare the high corrupted and low corrupted countries, consist of a total of 5,892 deals (see table 2). The final sample includes 1,156 firms in a total of 19 countries. Table 1 illustrates the distribution of the sample of M&A deals made in either high or low corrupted countries.
Table 2 shows the univariate analysis of the sample of high and low corrupted countries with the corresponding alphas and betas. The alpha act as a performance measure, because it tells you how much a firm performed better than the model predicted. As allocated, overvalued firms have negative alphas. Fair valued firms have a median of 0.0109 for high-corrupted countries and 0.0097 for low-corrupted countries, and the undervalued firms have a median alpha of 0.0363 for high corrupted and 0.0298 for low corrupted countries. The latter differ the most, however not exceedingly. While both the difference in means and medians are significant, it is an appropriate sample to compare.
Looking at the beta, we can see that all betas for low- and high corruption countries are below 1. Indicating that this sample consist of rather low volatile investments compared to the market, or their price movements do not correlate well with the market. While some outliers were present for high corrupted countries, the betas are winsorized at the 5% level. Interesting is that, for undervalued firms were the mean and meduan significantly differs, systematic risk is higher for corrupted countries. Suggesting that corrupt environments are related to higher systematic risk. Systematic risk is related to the sensitivity to market movements or more generally to the common risk of the market. Brockman, Rui, and Zou (2013) state that weak institutions give rise to high levels of corruption and low-quality legal systems. So, it is plausible that these environments come with a certain risk.
12
corrupted countries and 59.8% for high corrupted countries, the total amount of deals is considerably higher in the low corruption countries.
Table 1 Description of sample
Country Number of firms Percentage of all firms Number of deals Percentage of all deals
Low corrupted countries
New Zealand 166 13.0 1375 23.3 Denmark 95 7.4 1137 19.3 Finland 90 7.0 463 7.9 Norway 88 6.9 491 8.3 Switzerland 74 5.8 232 3.9 513 40.2 3698 62.8
High corrupted countries
Thailand 227 17.8 571 9.7 Russia 168 13.2 841 14.3 Philippines 112 8.8 261 4.4 Indonesia 89 7.0 234 4.0 Mexico 58 4.5 136 2.3 Egypt 41 3.2 59 1.0 Pakistan 27 2.1 37 0.6 India 15 1.2 19 0.3 Nigeria 10 0.8 14 0.2 Turkey 8 0.6 8 0.1 Brazil 5 0.4 5 0.1 Kazakhstan 2 0.2 5 0.1 Bahrain 1 0.1 2 0.0 Colombia 1 0.1 1 0.0 764 59.8 2193 37.2 Total 1277 100 5891 100
This table describes the sample of all M&A deals made in high – and low corrupted countries. The sample consists of 5,892 deals and 1,156 firms in a total of 19 countries during the period 2002-2017
Control variables
Table 3 shows the descriptive statistics of the control variables that are used. These variables follow Brockman, Rui, and Zou (2013) and are based on previous studies (see e.g. Chen, Ding, and Kim, 2010; Masulis, Wang, and Xie, 2007; Hope, Thomas, and Vyas, 2011). All variables are either obtained from Datastream or Orbis database. For all the firm characteristic variables are winsorized at the 5% level to reduce the effect of outliers. Except for Market CAP, where the log is used.
13
Table 2 Matched sample: Univariate analysis
Alpha
Mean Median
High Low Difference High Low Difference
corruption corruption (p-value) corruption corruption (p-value) Overvalued -0.0103 -0.0091 0.0011** -0.0080 -0.0058 0.0022*** Fair valued 0.0133 0.0102 0.0031*** 0.0109 0.0097 0.0011*** Undervalued 0.2965 0.0364 0.2601*** 0.0363 0.0298 0.0064***
Beta
Mean Median
High Low Difference High Low Difference
corruption corruption (p-value) corruption corruption (p-value) Overvalued 0.8963 0.9283 0.0320 0.8659 0.9291 0.0632 Fair valued 0.8659 0.8555 0.0104 0.8983 0.8462 0.0521 Undervalued 0.8391 0.7016 0.1375*** 0.8514 0.6529 0.1985***
This table shows the mean and median values of the alphas and betas from high- and low corrupted countries. The sample M&A deals are grouped based on the alpha whether they are overvalued, fair valued or undervalued. Respectively 493, 1207 and 493 deals are included for high corrupted countries and 679, 2340 and 679 deals for low corrupted countries. The betas are winsorized at 5% level. The two sample t-test is used to test the significance of the differences in mean against the hypothesis of 0. The Wilcoxon signed-rank sum test is to test the significance of the differences in the median against the hypothesis of 0. *, **, and *** denote significance at 10%, 5%, and 1% level respectively.
median. For leverage the mean (median) is 25,91% (25,79%) for low corrupted countries and 23,36% (22,68%) for high corrupted countries. Both the t-test as the Wilcoxon signed-rank test find significant differences in the mean and median. This significance is found as well for the rest of the firm characteristics, where the Market to Book, TobinsQ and the buy-and-hold returns for the previous 12 months are somewhat higher in low corrupted countries than high corrupted countries. That both low as high corrupted countries experience positive returns 1-year before the announced date of the M&A is in line with the managerial hubris hypothesis (Roll, 1986). Which state that acquirers tend to display superior performance before an announcement.
14
This table reports descriptive statistics on firm and deal characteristics made within low corrupted countries and high corrupted countries. The number of deals included for high corrupted countries is 2193, for low corrupted countries 3698 deals. Market cap is the market capitalization at the most recent fiscal year-end preceding the merger. Leverage is the ratio of total debt to total assets. Market to book contains the ratio of market value of equity to book value of equity. Tobin’s Q is the ratio of market value of equity plus book value of debt, divided by total assets. Prior return is the buy-and-hold return over 12 months before the mergers. Cross-border is whether a deal is closed between an acquirer and target firm from different countries. The two-sample t-test is used to test the significance of the differences in mean against the hypothesis of 0. The Wilcoxon signed-rank sum test is to test the significance of the differences in the median against the hypothesis of 0. *, **, and *** denote significance at 10%, 5%, and 1% level respectively.
4. Results
Univariate analysisTable 5 shows the post-merger performance of companies in low corrupted countries and high corrupted countries. Moreover, they are classified as overvalued, fair valued and undervalued. The post-merger performance is expressed in short-term performance (AR, CAR) and long-term performance (BHAR 1,2,3-year). The differences in mean and median are tested for significance with the two-sample t-test for the mean and the Wilcoxon signed-rank sum for the median.
For classification undervalued and fair valued, both the mean as the median have a significant difference in short –and long-term performance compared with different corruption levels. However, the difference in overvalued is only significant for the Abnormal Returns and is not significant for the CAR and all the BHAR’s. Looking at the returns for undervalued and fair valued firms, it is evident to say that shareholders obtain better returns at firms in low corrupted countries while in each classification the low corrupted countries get better returns than high corrupted countries. For fair valued firms, the differences in returns increases while the time after the announced date increase.
It seems that corruption has a negative effect on the short- and long term returns of a firm. The table shows worse returns in high corrupted countries, what could be in line with the market misvaluation theory (Shleifer and Vishny, 2003) that predicts that firms which are overvalued
Table 3 Descriptive statistics
Mean Median
High Low Difference High Low Difference
corruption corruption (p-value) corruption corruption (p-value)
15
will have a high pre-merger return and abnormal negative returns after the merger. This is because the market corrects overvaluation. Differences of prior return are 0.0334 in favor of low corrupted countries with a significance level at the 1 % level. Table 5 shows that the significant differences for long-term performance are even stronger in favor of low corrupted countries. Based on these results I could say this is in line with the misvaluation theory, indicating that corruption has a positive effect on the overvaluation of acquiring firms during an M&A. However, at this stage, it is premature to make conclusions.
One concern is, however, the skewness in the returns as they are not normally distributed. The descriptive statistics from the distribution of the returns are shown in table 4. Although the raw return data is already winsorized at a 5% level, still some skewness seems to be present. While the mean is somewhat higher than the median, the data seems positively skewed. This is the case for both returns in low corrupted as high corrupted countries. It is in line with Lyon, Barber and Tsai (1999) who explained that skewness bias arises while long-term stock return distribution is typical positively skewed. They state that, to control the bias, a skewness adjusted t-test is the best solution. In table 6 results are shown regarding the skewness adjusted t-test, where t-test1 is the outcome of Johnson’s corrected t-test for skewed data (Johnson, 1978) and t-test 2 is a modified Johnson test by Chen (Chen, 1995).
Table 6 shows slightly reducing t-values, indicating that variance decreases and somewhat control the skewness bias. And although the results of the skewness adjusted t-test lose some significance on the differences of the means, the general outcome is the same as shown in table 5. The significant different means show that low corrupted countries get a greater return than high corrupted countries. Moreover, it stays evidently higher than the differences of prior return of 0.0334 in favor of low corrupted countries.
This table represents the characteristics of the distribution of the abnormal returns (AR), cumulative abnormal returns (CAR), buy-and-hold returns after 1-year (BHAR1), 2-year (BHAR2) and 3-year (BHAR3). All returns have been winsorized at the 5% level. The p represents the percentiles
Table 4 Distribution: descriptive statistics
16
This table shows the mean and median values of the post-merger performance from high- and low corrupted countries. The sample mergers and acquisitions are grouped based on the alpha whether they are overvalued, fair valued or undervalued. Respectively 493, 1207 and 493 deals are included for high corrupted countries and 679, 2340 and 679 deals for low corrupted countries. The two-sample t-test is used to test the significance of the differences in mean against the hypothesis of 0. The Wilcoxon signed-rank sum test is to test the significance of the differences in the median against the hypothesis of 0. *, **, and *** denote significance at 10%, 5%, and 1% level respectively.
Table 5 Post-merger performance: Univariate analysis
Panel A: AR
Short-term performance Mean Median
High Low Difference High Low Difference
corruption corruption (p-value) corruption corruption (p-value) Overvalued -0.0129 0.0044 0.0173*** -0.0234 -0.0010 0.0224*** Fair valued 0.0069 0.0153 0.0084*** -0.0018 0.0121 0.0140*** Undervalued 0.0205 0.0369 0.0164*** 0.0105 0.0335 0.0230*** CAR Mean Median
High Low Difference High Low Difference
corruption corruption (p-value) corruption corruption (p-value) Overvalued 0.0011 0.0096 0.0085 -0.0222 -0.0074 0.0148 Fair valued 0.0183 0.0323 0.014*** 0.0006 0.0250 0.0244*** Undervalued 0.0459 0.0737 0.0278*** 0.0246 0.0675 0.0429***
Panel B: BHAR
Long-term performance 1-year
Mean Median
High Low Difference High Low Difference
corruption corruption (p-value) corruption corruption (p-value) Overvalued 0.0578 0.0677 0.0099 0.0172 0.0284 0.0112 Fair valued 0.0404 0.0642 0.0238** -0.0072 0.0423 0.0495*** Undervalued 0.0291 0.0863 0.0572*** -0.0129 0.0744 0.0873*** 2-year Mean Median
High Low Difference High Low Difference
corruption corruption (p-value) corruption corruption (p-value) Overvalued 0.1884 0.1688 0.0196 -0.0100 0.0868 0.0968 Fair valued 0.1222 0.1647 0.0425** 0.0322 0.0929 0.0607*** Undervalued 0.0744 0.1754 0.101*** -0.0832 0.0167 0.1900*** 3-year Mean Median
High Low Difference High Low Difference
17
Table 6 Post-merger performance: Univariate analysis skewness adjusted
Panel A: AR
Short-term performance Mean T-value
High Low Difference
corruption corruption (p-value) t-test t-test1 t-test2 Overvalued -0.0129 0.0070 0.0199 -4.13*** -1.67* 49.83*** Fair valued 0.0069 0.0212 0.0143 -4.78*** -2.63*** 29.77** Undervalued 0.0205 0.0371 0.0166 -2.94*** -1.85* 11.20*** CAR (+1. 0. -1) Mean T-value
High Low Difference
corruption corruption t-test t-test1 t-test2
Overvalued 0.0007 0.0117 0.0110 -1.15 -1.02 -0.85
Fair valued 0.0184 0.0446 0.0262 -4.70*** -3.57*** -0.92 Undervalued 0.0462 0.0758 0.0296 -2.76*** -2.24** -1.33
Panel B: BHAR
Long-term performance 1-year
Mean T-value
High Low Difference
corruption corruption t-test t-test1 t-test2
Overvalued 0.0566 0.0318 0.0248 1.08 1.13 1.15 Fair valued 0.0389 0.0893 0.0426 -3.68*** -3.36*** -3.29*** Undervalued 0.0290 0.0691 0.0506 -1.62 -1.52 -1.51 2-year Mean T-value
High Low Difference
corruption corruption t-test t-test1 t-test2
Overvalued 0.1585 0.1439 0.0146 0.33 0.34 0.34 Fair valued 0.1178 0.1866 0.1272 -2.74*** -2.62*** -2.62*** Undervalued 0.0686 0.1583 0.0961 -2.03** -1.94* -1.94* 3-year Mean T-value
High Low Difference
corruption corruption t-test t-test1 t-test2
Overvalued 0.2168 0.2032 0.0136 1.47 1.52 1.52
Fair valued 0.1692 0.2908 0.0174 -0.73 -0.72 -0.72 Undervalued 0.1181 0.1766 0.0585 -0.65 -0.64 -0.64
18 Multivariate analysis
A regression analysis is performed to investigate the relationship between corruption and the performance of the merger in short – and long-term. Moreover, different corporate governance variables are included as well as an interaction variable which measures the interaction effect of corporate governance and corruption on the after-merger performance.
As Brockman, Rui, and Zou (2013) I will use the following multivariate regression model:
CARit = βo + β1 Host country corruption + β2 Corporate governance + β3 Host country
corruption * Corporate governance + Controls + year fixed effects + industry fixed
effects + ε (5)
BHAR 1-year = βo + β1 Host country corruption + β2 Corporate governance + β3 Host
country corruption * Corporate governance + Controls + year fixed effects + industry
fixed effects + ε (6)
BHAR 2-year = βo + β1 Host country corruption + β2 Corporate governance + β3 Host
country corruption * Corporate governance + Controls + year fixed effects + industry
fixed effects + ε (7)
BHAR 3-year = βo + β1 Host country corruption + β2 Corporate governance + β3 Host
country corruption * Corporate governance + Controls + year fixed effects + industry
fixed effects + ε (8)
As described above the dependent variable is the performance measure. The most important explanatory variable is corruption, which takes the value of 1 when it is a high corrupted country and a value of 0 when it is a low corrupted country. The firm and deal control variables are included as well, which were described in table 3 and explained in the Appendix. The test statistics are based on standard errors adjusting for clustering at the firm level. Moreover, year fixed effects and industry fixed effects (based on Fama-French’s 12-industry classification) are included in every regression.
19
and 3-year BHAR, however, there is no significant effect of corruption on the post-merger performance. These findings on the CAR and 1-year BHAR confirm what is shown in the univariate analysis. Based on these findings and taking the market misvaluation theory (Shleifer and Vishny, 2003) into account the first hypothesis can be accepted. While corruption has a significant negative effect on the post-merger performance, it valid to say that during the event the acquiring firm was overvalued while they had a high prior return (see table 3) and abnormal negative returns after the merger. This is because the market corrects overvaluation.
Tables 8, 9, 10 and 11 show the results of the multivariate analysis including corporate governance variables. As stated earlier, CGVS TOTAL is included while DeFond, Hann, and Hu (2005) believe that a composite measurement better captures the overall environment of the firm’s governance than individual measures. Moreover, board functions (CGBF), the right of shareholders (CGSR) and compensation policy (CGCP) are included while these variables have a significant effect as corporate govenance variable.
The first thing that stands out is that some of these corporate governance variables have a significant negative effect on the post-merger performance. CGBF has a significant negative effect at the 10% level for 1- and 3-year BHAR and 5% for 2-year BHAR (table 9). Moreover, CGSR has a significant negative effect at the 10% level for 2-year BHAR (table 10). You could interpret this as unfavorable for shareholders to invest in firms that have a decent corporate governance policy. However corporate governance can still have a positive effect on performance, while abnormal returns are somewhat different. In line with Core, Guay, and Rusticus (2006), who say that shareholders forecast the lower operating performance and therefore have no impact on the abnormal returns, higher abnormal stock returns were found for weak governance (significant negative effect) firms compared with strong corporate governance firms. This can happen because investors expect the difference in operating performance. However, these effects are small and therefore of low impact.
20
Table 9 shows the insertion of variable CGBF. The interaction variable corruption * CGBF shows no significance. Meaning that following the best practice corporate governance principles related to board activities and functions does not moderate the effect of corruption on the post-merger performance. Table 10 shows the insertion of CGSR. The interaction variable corruption * CGSR shows no significance either. This shows that following the best practice corporate governance principles related to a shareholder policy and equal treatment of shareholders does not moderate the effect of corruption on the post-merger performance either. The second hypothesis cannot be accepted, considering the lack of evidence for this hypothesis. In addition to CGVS TOTAL, CGBF, and CGSR variables, other corporate governance variables are used to examine the effect of corporate governance and corruption on the short- and long-term performance of a firm. The variable CGCP, which measures the compensation policy, got some surprising, significant results which are shown in table 11. The interaction variable corruption * CGCP has no significant effect on the CAR. However, it does on the 1- and 2-year BHAR at the 5 % level. The coefficient of corruption in the 1- and 2-year BHAR is significant at the 5 % level for 1-year BHAR and 10% level at the 2-year BHAR and has a positive effect on the long-term performance of the firm. Suggesting that CGCP and the interaction variable has an opposite effect on corruption compared to the effect of corruption only in table 7.
21
This table reports the regression results of post-merger stock returns on corruption. This regression includes year fixed effects and industry fixed effects.The dependent variable is CAR, 1, 2 and 3-year BHAR. The CAR is the cumulative abnormal returns over a 3-month window. The BHARs are calculated as the long-run buy-and-hold return of a sample firm less the long-run return of the country index, which is referred to as a buy-and-hold abnormal return. See Appendix for definitions of independent variables. *, **, and *** denote significance at 10%, 5% and 1% level respectively
Table 7 Regression results of post-merger stock returns on corruption
Panel A: CAR Short-term performance Corruption -0.021*** (0.007) Market CAP -0.017*** (0.004) Leverage 0.005 (0.017) Market to Book 0.001 (0.002) TobinsQ. 0 (0.003) Cross-border 0.006 (0.007) Prior return 0.144*** (0.009) Constant 0.079*** (0.029) R-squared 0.173 Observations 3095 Panel B: BHAR
Long-term performance 1-year 2-year 3Year
This table reports the regression results of post-merger stock returns on corruption including corporate governance. This regression includes year fixed effects and industry fixed effects.The CGVS TOTAL is the corporate governance total score. See table 7 for dependent variables. See Appendix for definitions of independent variables. *, **, and *** denote significance at 10%, 5% and 1% level respectively.
Table 8 Regression results of post-merger stock returns on corruption including CGVS TOTAL
Panel A: CAR Short-term performance Corruption 0.006 (0.017) CGVS TOTAL 0 (0.000) Corruption * CGVS TOTAL 0 (0.000) Market CAP -0.009 (0.006) Leverage 0 (0.019) Market to Book 0 (0.003) TobinsQ. -0.002 (0.003) Cross-border 0.011 (0.010) Prior return 0.135*** (0.014) Constant 0.058 (0.051) R-squared 0.152 Observations 1389 Panel B: BHAR
Long-term performance 1-year 2-year 3Year
This table reports the regression results of post-merger stock returns on corruption including corporate governance. This regression includes year fixed effects and industry fixed effects.The CGBF represents corporate governance board functions. See table 7 for dependent variables. See Appendix for definitions of independent variables. *, **, and *** denote significance at 10%, 5% and 1% level respectively.
Table 9 Regression results of post-merger stock returns on corruption including CGBF
Panel A: CAR Short-term performance Corruption 0.015 (0.016) CGBF 0 (0.000) Corruption * CGBF 0 (0.000) Market CAP -0.01 (0.006) Leverage 0 (0.019) Market to Book 0 (0.003) TobinsQ. -0.002 (0.003) Cross-border 0.010 (0.010) Prior return 0.134*** (0.014) Constant 0.065 (0.050) R-squared 0.155 Observations 1394 Panel B: BHAR
Long-term performance 1-year 2-year 3Year
24
This table reports the regression results of post-merger stock returns on corruption including corporate governance. This regression includes year fixed effects and industry fixed effects. The CGSR stands for the corporate governance shareholder rights. See table 7 for dependent variables. See Appendix for definitions of independent variables. *, **, and *** denote significance at 10%, 5% and 1% level respectively.
Table 10 Regression results of post-merger stock returns on corruption including CGSR
Panel A: CAR Short-term performance Corruption -0.004 (0.016) CGSR 0 (0.000) Corruption * CGSR 0 (0.000) Market CAP -0.008 (0.006) Leverage 0 (0.019) Market to Book 0 (0.003) TobinsQ. -0.001 (0.003) Cross-border 0.011 (0.010) Prior return 0.135*** (0.014) Constant 0.058 (0.05) R-squared 0.153 Observations 1394 Panel B: BHAR
Long-term performance 1-year 2-year 3Year
25
This table reports the regression results of post-merger stock returns on corruption including corporate governance. This regression includes year fixed effects and industry fixed effects. The CGCP is the firm's compensation policy. See table 7 for dependent variables. See Appendix for definitions of independent variables. *, **, and *** denote significance at 10%, 5% and 1% level respectively.
Table 11 Regression results of post-merger stock returns on corruption including CGCP
Panel A: CAR Short-term performance Corruption -0.004 (0.018) CGCP 0 (0.000) Corruption * CGCP 0 (0.000) Market CAP -0.008 (0.006) Leverage 0 (0.019) Market to Book 0 (0.003) TobinsQ. -0.002 (0.003) Cross-border 0.011 (0.010) Prior return 0.136*** (0.014) Constant 0.054 (0.051) R-squared 0.150 Observations 1394 Panel B: BHAR
Long-term performance 1-year 2-year 3Year
26
5. Conclusion
In this study, I investigate the effect of corruption on the overvaluation of acquiring firms during an M&A. By describing the agency relationship between shareholders and managers it is explained why managers do not always act for the benefit of a firm. In addition, underlying motives are described during an M&A consisting of synergy, agency, and hubris. While both managers, shareholders, and governments have interests in M&As, corruption seems to play a role during an M&A. Comparing a sample of M&A deals between high- and low corrupted countries shows that corruption has a significant negative effect on short-and long-term performance using both the univariate as the multivariate analysis. In addition, with positive and higher prior returns it is assumed that corruption has a positive effect on the overvaluation of acquiring firms during an M&A. This is based on the market misvaluation theory (Shleifer and Vishny, 2003). Secondly, I find that considerable more domestic deals are made in high corrupted countries compared to low corrupted countries. These outcomes suggest that domestic mergers and acquisition is encouraged, and foreign M&As are discouraged in corrupted countries. This is in line with what Serdar Dinc and Erel (2013) found. Thirdly, the betas show that in generally systematic risk is higher for corrupted countries, suggesting that corrupt environments are in general more risky investments for shareholders. Finally, I found that the corporate governance mechanism does not show moderating effects on the main relationship. Meaning that implementing these governance mechanisms do not weaken the negative effect of corruption on the long-term performance of the firm. These findings are useful insights for shareholders while investing in overvalued firms can be of negative impact on their investment. Investors should consider that in corrupted countries governments might influence managers in a way that is unfavorable for the firm itself. Moreover, they should notice that corrupt environment brings a certain risk with their investment and not necessarily result in better abnormal returns.
Limitations and further research
27
Appendix
Table A1 Variable definitions
Variable Database Definition
Market CAP Datastream Market capitalization (#WC07120) in US dollars at most recent fiscal year-end preceding the merger
Leverage Datastream Ratio of total debt (#WC03255) to total assets (#WC02999)
Market to Book Datastream Ratio of market value of equity (#WC08001) to the book value of equity(#WC03501) TobinsQ Datastream
Ratio of market value of equity (#WC08001) plus total assets (#WC02999) minus book value of common equity (#WC03501), divided by total assets (#WC02999)
Prior return Datastream Firms' BHAR over 12 months prior to the M&A
Cross-border Orbis database (country code) Indicator equals 1 if the acquirer and target are from different countries
The following tables present the Pearson’s correlation coefficients to show the statistical relationship between two variables. Moreover, the variance inflation factors (VIFs) are included to detect multicollinearity. The VIF is calculated as
𝑉𝐼𝐹 =
11−𝑅𝑓2
(9)
29
Table A2 Correlation table - corruption
1 2 3 4 5 6 7 8 9 10 11 VIF 1/VIF 1. CAR 1.0000 2. BHAR1 0.1968 1.0000 3. BHAR2 0.1458 0.7287 1.0000 4. BHAR3 0.1336 0.5990 0.8199 1.0000 5. Corruption -0.0213 -0.0019 -0.0059 -0.0232 1.0000 1.7900 0.5598 6. MarketCAP -0.1167 -0.0137 -0.0694 -0.0910 0.0558 1.0000 1.5700 0.6354 7. Leverage 0.0197 0.0395 0.0175 -0.0026 -0.0893 0.2008 1.0000 1.0900 0.9147 8. MarkettoBook 0.0382 0.0304 0.0101 0.0354 -0.0619 0.0128 0.0254 1.0000 1.1800 0.8460 9. TobinsQ -0.0083 0.0313 -0.0046 -0.0086 -0.0234 -0.0846 -0.0093 -0.0105 1.0000 1.1200 0.8900 10. Crossborder 0.0298 0.0339 0.0168 0.0102 0.5344 -0.1521 -0.0717 0.0002 0.0191 1.0000 1.7600 0.5687 11. Prior return 0.3828 0.0797 0.0522 0.0702 -0.0194 -0.0734 -0.0063 -0.0065 0.0147 0.0010 1.0000 1.0500 0.9561
30
This table shows the Pearson’s correlation coefficient to show the statistical relationship between two variables. Variance inflation factors (VIF) are included to show multicollinearity. This table includes all dependent, control variables and the independent variables corruption, CGVSCORETOTAL and interaction variable CGVSCORETOTAL*corruption (intcorCGVSCT).
Table A3 Correlation table - corruption including CGVSCORETOTAL
31
This table shows the Pearson’s correlation coefficient to show the statistical relationship between two variables. Variance inflation factors (VIF) are included to show multicollinearity. This table includes all dependent, control variables and the independent variables corruption, CGBF and interaction variable CGBF*corruption (intcorCGBF).
Table A4 Correlation table - corruption including CGBF
32
Table A5 Correlation table - corruption including CGCP
1 2 3 4 5 6 7 8 9 10 11 12 13 VIF 1/VIF 1. CAR 1.0000 2. BHAR1 0.2159 1.0000 3. BHAR2 0.1757 0.7310 1.0000 4. BHAR3 0.1932 0.5829 0.8125 1.0000 5. Corruption -0.0405 0.0411 -0.0604 -0.0874 1.0000 7.4200 0.1347 6. CGCP 0.0130 -0.0776 0.0001 0.0331 -0.6284 1.0000 2.5000 0.4006 7. intcorCGCP -0.0296 -0.0499 -0.0995 -0.0629 0.7378 -0.2282 1.0000 3.4100 0.2932 8. MarketCAP -0.0880 0.0008 -0.1097 -0.1759 0.3492 -0.3350 0.1039 1.0000 1.9600 0.5098 9. Leverage 0.0561 0.0315 0.0490 0.0479 0.0312 -0.0758 -0.0197 0.2060 1.0000 1.1500 0.8716 10. MarkettoBook 0.0356 0.0188 0.0547 0.1024 -0.0899 0.0620 -0.0456 -0.0627 0.0127 1.0000 1.3500 0.7430 11. TobinsQ -0.0955 -0.0942 -0.0917 -0.0557 0.0446 0.0280 0.1691 -0.1423 -0.0036 -0.0325 1.0000 1.3700 0.7308 12. Crossborder -0.0145 0.0262 -0.0468 -0.0515 0.6827 -0.4423 0.4797 0.1734 0.0045 -0.0132 0.0875 1.0000 2.3200 0.4308 13. Prior return 0.3534 0.0597 0.1007 0.1452 -0.0670 0.0823 -0.0236 -0.1743 0.0353 -0.0126 -0.0682 -0.0605 1.0000 1.1500 0.8721
33
Table A6 Correlation table - corruption including CGSR
1 2 3 4 5 6 7 8 9 10 11 12 13 VIF 1/VIF 1. CAR 1.0000 2. BHAR1 0.2159 1.0000 3. BHAR2 0.1757 0.7310 1.0000 4. BHAR3 0.1932 0.5829 0.8125 1.0000 5. Corruption -0.0405 0.0411 -0.0604 -0.0874 1.0000 6.6100 0.1512 6. CGSR -0.0914 -0.1001 -0.0874 -0.0639 -0.3239 1.0000 2.1300 0.4685 7. intcorCGSR -0.0696 0.0254 -0.0421 -0.0550 0.6673 0.2163 1.0000 3.8500 0.2596 8. MarketCAP -0.0880 0.0008 -0.1097 -0.1759 0.3492 -0.1561 0.0679 1.0000 1.9100 0.5228 9. Leverage 0.0561 0.0315 0.0490 0.0479 0.0312 -0.0697 -0.0374 0.2060 1.0000 1.1300 0.8853 10. MarkettoBook 0.0356 0.0188 0.0547 0.1024 -0.0899 -0.1605 -0.1254 -0.0627 0.0127 1.0000 1.7900 0.5584 11. TobinsQ -0.0955 -0.0942 -0.0917 -0.0557 0.0446 0.2569 0.1811 -0.1423 -0.0036 -0.0325 1.0000 1.1000 0.9112 12. Crossborder -0.0145 0.0262 -0.0468 -0.0515 0.6827 -0.2016 0.4473 0.1734 0.0045 -0.0132 0.0875 1.0000 2.3400 0.4282 13. Prior return 0.3534 0.0597 0.1007 0.1452 -0.0670 -0.0254 -0.0423 -0.1743 0.0353 -0.0126 -0.0682 -0.0605 1.0000 1.1500 0.8692
References
Alexandridis, G., Antypas, N., and Travlos, N. (2017). Value creation from M&A’s: New evidence. Journal of Corporate Finance, 45, 632-650.
Alexandridis, G., Petmezas, D., and Travlos, N. G. (2010). Gains from mergers and acquisitions around the world: New evidence. Financial Management, 39, 1671-1695.
Barber, B. M., and Lyon, J. D. (1997). Detecting long-run abnormal stock returns: The empirical power and specification of test statistics. Journal of Financial Economics, 43, 341-372.
Bessembinder, H., and Zhang, F. (2013). Firm characteristics and long-run stock returns after corporate events. Journal of Financial Economics, 109, 83-102.
Brockman, P., Rui, O. M., and Zou, H. (2013). Institutions and the performance of politically connected M&A’s. Journal of International Business Studies, 44, 833-852.
Chen, C. J., Ding, Y., and Kim, C. F. (2010). High-level politically connected firms, corruption, and analyst forecast accuracy around the world. Journal of International Business Studies, 41, 1505-1524.
Chen, L. (1995). Testing the mean of skewed distributions. Journal of the American Statistical Association, 90, 767-772.
Chi, J. D., and Gupta, M. (2009). Overvaluation and earnings management. Journal of Banking & Finance, 33, 1652-1663.
Claessens, S., Feijen, E., and Laeven, L. (2008). Political connections and preferential access to finance: The role of campaign contributions. Journal of Financial Economics, 88, 554-580.
Core, J. E., Guay, W. R., and Rusticus, T. O. (2006). Does weak governance cause weak stock returns? An examination of firm operating performance and investors' expectations. The Journal of Finance, 61, 655-687.
Cuervo-Cazurra, A. (2006). Who cares about corruption? Journal of International Business Studies, 37, 807-822.
35
DeFond, M. L., Hann, R. N., and Hu, X. (2005). Does the market value financial expertise on audit committees of boards of directors? Journal of Accounting Research, 43, 153-193.
Donadelli, M., Fasan, M., and Magnanelli, B. S. (2014). The agency problem, financial performance and corruption: Country, industry and firm level perspectives. European Management Review, 11, 259-272.
Fama, E. F. (1998). Market efficiency, long-term returns, and behavioral finance1. Journal of Financial Economics, 49, 283-306.
Fu, F., Lin, L., and Officer, M. S. (2013). Acquisitions driven by stock overvaluation: Are they good deals? Journal of Financial Economics, 109, 24-39.
Gompers, P., Ishii, J., and Metrick, A. (2003). Corporate governance and equity prices Quarterly Journal of Economics, 118, 107.
Hope, O. K., Thomas, W., and Vyas, D. (2011). The cost of pride: Why do firms from developing countries bid higher? Journal of International Business Studies, 42, 128-151.
Ikenberry, D., Lakonishok, J., and Vermaelen, T. (1995). Market underreaction to open market share repurchases. Journal of Financial Economics, 39, 181-208.
Jensen, M. C. (2005). Agency costs of overvalued equity. Financial Management, 34, 5-19.
Jensen, M. C. (1986). Agency costs of free cash flow, corporate finance, and takeovers. The American Economic Review, 76, 323-329.
Jensen, M. C., and Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3, 305-360.
Johnson, N. J. (1978). Modified t tests and confidence intervals for asymmetrical populations. Journal of the American Statistical Association, 73, 536-544.
Johnson, S. A., Ryan, H. E., and Tian, Y. S. (2009). Managerial incentives and corporate fraud: The sources of incentives matter. Review of Finance, 13, 115-145.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R. (2000). Investor protection and corporate governance. Journal of Financial Economics, 58, 3-27.
36
Lui, F. T. (1985). An equilibrium queuing model of bribery. Journal of Political Economy, 93, 760-781.
Lyon, J. D., Barber, B. M., and Tsai, C. L. (1999). Improved methods for tests of long‐run abnormal stock returns. The Journal of Finance, 54, 165-201.
Masulis, R. W., Wang, C., and Xie, F. (2007). Corporate governance and acquirer returns. The Journal of Finance, 62, 1851-1889.
Mauro, P. (1995). Corruption and growth. The Quarterly Journal of Economics, 110, 681-712.
Petersen, M. A. (2009). Estimating standard errors in finance panel data sets: Comparing approaches. The Review of Financial Studies, 22, 435-480.
Rhodes–Kropf, M., Robinson, D. T., and Viswanathan, S. (2005). Valuation waves and merger activity: The empirical evidence. Journal of Financial Economics, 77, 561-603.
Roll, R. (1986). The hubris hypothesis of corporate takeovers. Journal of Business, 197-216.
Salama, F. M., and Putnam, K. (2013). The impact of corporate governance on the financial outcomes of global diversification. The International Journal of Accounting, 48, 364-389.
Shleifer, A., & Vishny, R. W. (2003). Stock market driven acquisitions. Journal of financial Economics, 70, 295-311.
Shleifer, A., & Vishny, R. W. (1993). Corruption. The Quarterly Journal of Economics, 108, 599-617.
Serdar Dinc, I., and Erel, I. (2013). Economic nationalism in mergers and acquisitions. The Journal of Finance, 68, 2471-2514.
Wei, S. J. (2000). How taxing is corruption on international investors? Review of Economics and Statistics, 82, 1-11.
