Tax aggressiveness in cross-border mergers and acquisitions : the UK and US evidence

(1)

Tax Aggressiveness in Cross-Border Mergers and Acquisitions

The UK and US evidence

University of Amsterdam

Faculty of Economics and Business MSc Finance

Master Specialisation Corporate Finance

Author: Yufan Guo

Student number: 11804904

Thesis supervisor: Dr. Jan Lemmen Finish date: 30 June, 2018

(2)

ACKNOWLEDGEMENTS

First, I would like to thank my thesis supervisor, Dr. Jan Lemmen, for his help during the process of writing this master thesis and I am gratefully indebted to his valuable comments and suggestions on this thesis.

Next, I would like to thank the fellow classmate, Bart van Wijck, in thesis seminar course for his feedbacks on my thesis proposal.

Finally, I must express my special thanks of gratitude to my parents and friends for their unfailing support and encouragement.

Thank you again to all who helped me.

Statement of Originality

This document is written by Yufan Guo, who declares to take full responsibility for its contents.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

ABSTRACT

The paper analyses the tax aggressiveness behaviour in cross-border acquisitions. By using a sample of cross-border acquisitions with acquirers from US and target firms from UK, this paper employs three indicators (ETR, profitability, and leverage) to explore the multinationals’ tax avoidance behaviour. It is found that, for wholly acquired firms, a target firm’s ETR in cross-border acquisitions decreases by 5 percentage points after deals, in comparison with a target firm’s ETR in domestic acquisitions. In addition, a target firm’s profitability increases after acquisition which indicates that multinationals may transfer profits to the low-tax country to reduce their tax burdens. Conversely, the results do not support debt shifting hypothesis as there is no significant change in a target firm’s leverage after transactions.

Keywords:

Cross-border M&A, tax aggressiveness, profit shifting, debt shifting JEL Classification: H26, G34

(4)

1. Introduction

Firms may implement a Mergers & Acquisitions (M&A) strategy to adjust their capital structure by taking in debt and equity of the target firms (Gugler & Konrad, 2002). Harris and Ravenscraft (1991) find that firms are expected to benefit more from engaging in cross-border M&A than in domestic M&A. This finding may reflect reality, in that the volume of cross-nation acquisition has been increasing dramatically, from 23% of total M&A volume in 1998 to 45% in 2007.

Why would gains from cross-border M&As be higher than from domestic M&As? Harris and Ravenscraft (1991) find poor evidence for tax and industry variables while explaining the cross-border effect on shareholder wealth gains. However, Scholes and Wolfson (1990) study the changes in U.S. tax laws that attract foreign buyers to merge with U.S firms and/or purchase of U.S. firms. The U.S. government (federal, state, and local) levies more than one-third portion of a firm’s pre-tax income, as shown in the appendix 1. Because of the considerable tax costs to both corporations and shareholders, managers and shareholders may become tax aggressive.

In this thesis, I study whether such tax aggressiveness is one of the drivers of cross-border M&A. As is widely accepted in the literature, effective tax rate (ETR) is a proxy for tax aggressiveness. A low ETR indicates the possibility that a firm is avoiding taxes (Swenson (1999); Hanlon and Slemrod (2009)). Using the sample of 651 U.K. target firms that were acquired by U.S. firms occurring between 2000 and 2016, I estimate if the target firms in cross-border deals reveal lower ETRs than target firms in domestic acquisitions, which is the main research question of this paper. The difference between domestic acquisitions and cross-nation acquisitions in tax aggressiveness depends on the differences in country corporate tax rates. Because U.S. acquirers may find that it is less costly to their shareholders if they can manipulate their overseas subsidiaries’ earnings to mitigate the large amount of tax costs and/or they can shift debt in order to extract higher tax benefits.

I employ difference-in-difference methodology to compare three indicators (ETR, profitability, and leverage) of tax aggressiveness in this study. Moreover, I gather the financial information for this topic from the Amadeus database as it provides unique

(6)

unconsolidated financial and operating data, which enables the examination of three indicators prior to and after M&A transactions.

Overall, the paper evidences that, for wholly owned subsidiaries, the average effective tax rate of the target firm in cross-border acquisitions is lower than the average effective tax rate in domestic acquisitions after completing the acquisition. Moreover, I find that the profitability of the cross-border target increases more than the profitability of the domestic target, indicating that U.S. multinationals may transfer profits to the UK subsidiary to reduce their tax liabilities. However, the results do not support the argument for debt shifting. Specifically, I do not observe a significant change in target firms’ leverage ratio after acquisitions. The insignificant change may due to the tax unity regimes implemented in the UK.

The paper adds to several strands of existing literature. Firstly, this study contributes to the literature on tax aggressiveness. Extant literature concentrates on the role of tax in mergers and acquisitions. For example, Devos, Kadapakkam, and Krishnamurthy (2009) evaluate the extent to which tax savings are associated with merger gains. Martin et al. (2017) analyze the relationship between target tax aggressiveness and acquisition premiums. Hence, this study fills in the literature gap by testing the two channels that cross-border acquirers could employ to minimize their tax payments.

Secondly, this study also contributes to the debt and profit shifting literature. Schindler and Schjelderup (2012) investigate debt shifting behaviour and ownership structure. Huizinga and Laeven (2008) consider the international profit shifting incentives between affiliates and parent companies. Moreover, Jog and Tang (2001) examine whether the tax reform, measured as the change in the relative tax rate between the United States and Canada, has an influence on leverage of Canadian affiliates of foreign parent firms. In this study, I take the acquisition events into consideration. More specifically, I test whether parent firms change a target firm’s debt level and profitability after acquisition.

Lastly, this study contributes to the existing M&A literature. There exists a growing literature on border acquisition. Fewer studies have examined the difference between cross-border acquisition and domestic acquisition. This paper analyzes the difference by comparing cross-border acquisitions and domestic acquisitions in terms of tax motivation.

(7)

The paper proceeds as follows. Section 2 discusses the previous literature and develops hypotheses. Section 3 describes the data source and selection criteria. Section 4 presents the standard difference-in difference methodology, and in section 5, results are presented. This paper ends with the conclusion.

(8)

2. Literature Reviews and Hypotheses Development

2.1 Effective Tax rate

Effective tax rate (ETR) is commonly used as a measure of a firm’s tax aggressiveness. For example, Chen et al. (2010) employ ETR to examine the difference of tax aggressiveness among family firms and non-family firms. Under GAAP, effective tax rate is defined as total tax expense divided by pre-tax income (Dyreng, Hanlon and Maydrew, 2010).

𝐸𝑇𝑅 =

!"# !"#!$%!%

!"#$%$& !"#$%" !"# (!"#) Eq (1)

Erel et al. (2012) examine the incentives that drive cross-border mergers and acquisitions. They measure the average difference of corporate income tax rates among acquirer and target countries as the tax factor. They point out that tax rate is one of determinants in cross-border M&A and they argue that companies located in high corporate income tax rates countries are likely to acquire firms in low corporate income tax rate countries. Scholes and Wolfson (1990) also support this finding by investigating the U.S. tax reform in the 1980s. They argue that the changes in tax laws may modify the taxpayers’ investment and financing plans because foreign investors that face a greater corporate tax rate may value assets relatively more highly than domestic investors. Then, countries with low corporate tax rates become a tax haven compared to other countries that deal with relatively high corporate tax rates. Besides, Erel et al. (2012) also argue that M&As can decrease the joint tax liability of two companies if one firm can exploit a tax shield that the other firm owns but cannot use. Hence, the above findings lead to the first hypothesis:

The target firms’ effective tax rate drops after acquisitions in the context of high corporate tax rate companies acquiring low corporate tax rate companies.

2.2 Profit Shifting

A company that faces a high corporate tax rate may have the incentive to engage in international acquisitions with the aim of reallocating its income to a subsidiary located in a relatively low corporate tax rate country. By shifting its income, the multinational can effectively reduce its overall corporate tax burden. Harris et al. (1993) examine the profitability of US parent companies and detect that parent companies with affiliates in low-tax nations have a significantly lower return than parent companies in high-low-tax countries. Grubert (2013) performs profit shifting analysis for 754 U.S. multinational corporations within the period from 1996 to 2004. He calculates each multinational firm's foreign incomes

(9)

as a share of its total profits, and concludes that changes in foreign profits are negatively associated with the average foreign tax rate. Huizinga and Laeven (2008) further explained that the profit can be transferred via several channels. First, the company can employ intra-firm transaction manipulation. Particularly, the intra-firm could report a lower accounting profit in the high corporate tax country by exaggerating the import prices and lessening the export values. Second, the firm can change its financial structure to raise debt in high corporate tax countries. Therefore, the firm is required to pay large amount of annual interest and consequently minimize the accounting profits. Third, the firm can also redistribute certain costs, such as research and development expenses, to high corporate tax rate countries. Accordingly, it reduces the accounting income in the high-tax countries. Overall, a firm in a high-tax place will transfer the profit to the subsidiary that is located in a low-tax country. Hence, the above findings lead to the second hypothesis:

In the case of UK targets, it is expected that the target’s profitability will increase after acquisition because the tax rate in the UK is lower than in the U.S.

(10)

2.3 Debt Shifting

The company can choose either to use debt finance or equity finance to invest in projects and M&A deals. Huizinga et al. (2008) suggest that firms prefer debt finance to equity finance because interest costs are tax deductible whereas dividends may be subject to double taxation. Firms that operate solely domestically do not face the other countries’ corporate tax systems while cross-border acquisitions deal with the more complex choice of deciding the whole firm’s indebtedness and also determining the proportion of their debts to the parent company and the subsidiaries in other countries. Moore and Ruane (2005) prove this argument by indicating the subsidiaries’ leverage ratios are subject to regional corporate tax rates. Fuest et al. (2010) investigate international debt shifting between developing and developed countries. They find that the effect of the host country tax rate on the leverage ratio of cross-nation subsidiaries in developing countries is positive. Huizinga et al. (2008) further state, because of the tax advantages generated from the debt finance, firms are more willing to set up a higher leverage ratio than would be expected. Hence, they conclude that, for multinational firms, the difference in tax rate among different countries provides the incentive to reallocate their debt internationally.

Moreover, according to Modigliani-Miller theorem, the companies will take much debt as more tax shields can be earned. Hence, the above findings lead to the third hypothesis:

In the case of UK targets, it is expected that the target’s leverage will decrease after acquisition because the higher corporate tax rate in U.S. allows the parent firms to benefit from more tax shields.

Table1: Literature review summary

The table provides literature summary with detailed model information and time period. Authors

(Publication year)

Region Time period Model Results

Erel et al. (2012)

Worldwide 1990-2007 Dependent: number of

cross-border deals scaled by sum of the number of domestic deals and cross-border deals; Independent: difference between exchange rate, difference among real stock return, religion indicator, language indicator, GDP growth, value-weighted Geography, the quality of accounting disclosure, and bilateral trade, valuation and taxes play important roles in motivating mergers and

(11)

market-to-book equity, maximum of bilateral import and export, difference between income tax, disclosure of accounting information quality, geographic proximity, difference between annual GDP divided by the population

acquisition

Scholes and Wolfson (1990)

US 1968-1987 Compare M&A between U.S.

companies and non-U.S. companies The Tax Reform Act of 1986 encourages the cross-border transactions Harris et al. (1993) U.S. manufacturing firms

1984-1988 Dependent: US tax/US assets

Independent: dummy

indicating high-tax or low-tax regions

Controls: R&D, advertising, depreciation, employees, rent, interest (Scaled by worldwide assets)

U.S.

manufacturing companies shift profit from the U.S. to low-tax countries and transferred income from high-tax countries into the U.S.

Grubert (2012) U.S. 1996-2004 Dependent: change in foreign

share of income; change in domestic profit margin; change in foreign profit margin

Independent: change in average effective foreign tax rate, parent R&D/sales, parent advertising/sales, size, change in worldwide profit margin. The difference in tax motivates firms to invest abroad and to shift profit via price manipulation, the location of issuing company debt and other mechanisms. Huizinga and Laeven (2008) European multinationals

1999 Dependent: reported profit

Independent: tax rate, a weighted average of the effective tax rate difference, shifting expense International profit shifting among European firms is significant. The shifting costs are round 0.6% of the tax base of

European multinational firms

(12)

Huizinga et al. (2008)

32 European countries

1994-2003 Dependent: debt ratio

Independent: optimal leverage ratio, the impact of taxation on the optimal leverage ratio that would occur for a purely domestic firm, international tax rate differences on the optimal leverage The optimal leverage ratio is positively related to the corporate tax rate and to the difference among two countries’ corporate tax rate Fuest et al. (2010)

Germany 1996-2007 Dependent: debt-to-capital

ratio

Independent: tax rate, difference in tax rate, GDP, GDP per capita, real interest rate, corruption

The

multinational use the intra-firm loans to finance affiliates due to the tax benefits.

3. Data

3.1 Data Availability

In order to measure the acquisition effect on target firms’ effective tax rate, income and debt, the financial data prior to the acquisition and post-acquisition should be available. Erel et al. (2015) extensively explain that collection of financial data on the U.S. target firms before and after the acquisition is impossible as the targets’ post-acquisition financial statements are reported on the consolidated basis in the United States. By contrast, many European firms are required to release financial information publicly on the unconsolidated basis. Although some targets were private companies or operated as subsidiaries of other firms, their financial data are still reachable via the Amadeus database. However, a potential issue arising from acquisition is that the parent companies may integrate some of targets’ assets with acquirers’ existing assets (Erel et al., 2015).

3.2 Sample Construction

The sample of European acquisitions is gathered from the Zephyr database. In this study, the Zephyr database is more reliable than Thomson SDC since Zephyr database and Amadeus database are both provided by the vendor Bureau Van Dyck. Accordingly, both databases share common firm identification codes (BvD ID number). Consequently, it is more accurately matching acquisition events to financial data by using the common firm identification code.

(13)

I begin with all completed confirmed cross-border M&As with complete dates between January 1, 2000 and December 31, 2016 that are covered by the Zephyr database. I limit the data filter to United Kingdom as the target country while United States of America as the acquirer country. The two intuitions behind the country selection are the following. Firstly, it eliminates the country difference in law since both countries implement common law. Secondly, these two countries fulfil the criteria that the acquirer country has a higher corporate tax rate than the target country. As shown in the appendix 1, the U.S. corporate tax rates are higher than United Kingdom’s corporate tax rates over 2000-2017. Furthermore, I exclude M&A deals in financial and utility industries. Because the financial industry often has a high financial leverage ratio and utility firms often encounter extraordinary market competition and exceptional financial risks (Bouraoui and Li, 2014). I also drop deals that apply to targets involved in restructuring, privatization and leverage buyouts. Finally, I require that the acquirers should completely absorb the target company (i.e. 100% ownership change). The final cross-border M&A sample ends up with 651 observations over 2000-2016. For domestic M&A, I use the same filters except the acquirers are UK firms. These filters are summarized in the table 2.

Table 2: Sample selection summary

The table describes the selection procedure for cross-border M&A and domestic M&A samples. Column 1 shows the criteria for sample selection. Columns 2 and 3 report the number of observations left after each criterion

Search result

(1) Cross-border M&A (2) Domestic M&A

(3)

Acquired by US firms Acquired by UK firms

1. Time period: on and after 01/01/2000

and up to and including

31/12/2016(completed-confirmed)

1,059,062 1,059,062

2. Country (primary addresses): United

Kingdom (GB) (Target) 116,540 116,540

3. Country (primary addresses): United

States of America (US) / United Kingdom (UK) (Acquirer)

8,635 70,769

(14)

6. Remove restructuring, privatization and

leverage buyouts 732 4,407

7. Missing data (e.g. ID number,

acquisition share etc.) 651 3,376

3.3 Sample Characteristics

Firms engaged in cross-border M&A deals are different from firms involved in domestic M&A. Table 3 shows the acquisition information and firms’ descriptive statistics before and after acquisition. Panel A provides more detailed acquisition information. In the case of cross-border deals, 643 out of 651 deals were fully acquired (i.e. 100% share) by US firms, 22 out of 651, acquirers increased the shares they owned in target firms. In total, there were 9 deals, which the offered prices were revised upward. In domestic M&A, 3,283 out of 3,376 deals were 100% acquired by domestic firms. There were 198 deals that that acquirers increased the percentage of shares they owned. 39 deals were revised.

In panel B, the table shows that firms’ characteristics of cross-border M&A are significantly different from those of domestic M&A. Before acquisition, companies in cross-border M&A show higher ETR, higher leverage ratio, larger size, higher ROA, lower capital intensity, lower cost of employment and lower profitability than firms that completed domestic M&A deals. After acquisition, it is observed that firms in cross-border M&A demonstrate similar ETR, higher leverage ratio, smaller size, higher ROA, lower capital intensity, lower cost of employment and lower profitability than those in domestic M&A.

For cross-border M&A firms exhibit, on average post-acquisition, higher ETR, lower leverage ratio, slightly higher size, higher ROA, lower capital intensity, lower costs of employment, and similar profitability compared to the values before acquisition. While in the case of domestic M&A deals, on average, firms have higher ETR, lower leverage ratio, larger size, higher ROA, lower capital intensity, lower costs of employment, and slightly lower profitability compared to the values before acquisition.

(15)

Table 3: Firm characteristics before and after acquisition

This table presents summary statistics for firm characteristics before and after acquisition. In panel A, it shows the summary statistics of deal type. In panel B, it reports the financial statistics for two types acquisition (i.e. cross-border and domestic acquisition). The mean and standard deviation (in parentheses) for each variable are shown separately. Columns 1 and 3 present estimates for cross-border M&A before and after acquisition. Columns 2 and 4 present estimates for domestic M&A before and after acquisition. All financial ratios are winsorized at the 1% level.

Panel A: Acquisition information

Cross-border M&A Domestic M&A

Wholly owned (acquired 100% share)

643 3,283

Increase share 22 198

Increase offer 9 39

Panel B: Financial statistics

Before Acquisition After Acquisition

Cross-border M&A (1) Domestic M&A (2) Cross-border M&A (3) Domestic M&A (4) ETR -0.188 (0.501) -0.239 (0.415) -0.139 (0.473) -0.149 (0.433) Leverage 0.807 (1.406) 0.637 (0.393) 0.745 (1.872) 0.619 (0.575) Size 16.953 (1.741) 16.777 (1.660) 17.032 (1.635) 17.225 (1.899) ROA 6.815 (20.391) 5.438 (15.361) 7.009 (23.005) 6.250 (17.410) Capital intensity 0.373 (0.269) 0.409 (0.282) 0.280 (0.258) 0.378 (0.304) Profitability 0.015 (0.452) 0.022 (0.220) 0.015 (0.465) 0.021 (0.349) Costs of employment 0.374 (0.332) 0.434 (0.382) 0.354 (0.380) 0.360 (0.444)

4. Methodology

4.1 Effective tax rate

To test the first hypothesis, I use the difference-in-difference approach to mitigate the omitted variable bias effect. The estimation equation is used as follows:

(16)

Where

𝐸𝑇𝑅

𝑖𝑡 represents the effective tax rate for company i at period t,

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡 equals to

one if the target firm was acquired by the US firm (i.e. cross-border acquisition), and

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡

equals to zero if the target firm engaged in domestic acquisition deals, which is considered as the control group.

𝑃𝑜𝑠𝑡

𝑖𝑡 is a dummy equal to one after the acquisition. The interaction

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡

×𝑃𝑜𝑠𝑡

𝑖𝑡 measures the expected difference in mean of effective tax rate between

cross-border acquisition and domestic acquisition over time. Hence

𝛿

is the coefficient of interest in this study. The null hypothesis is

𝛿

=0, which could illustrate that the effective tax rate is indifferent between cross-border M&A and domestic M&A after acquisition. Conversely, if tax rate difference between UK and US is one of the factors that drive cross-border M&A, I expected a negative value,

𝛿

< 0. A negative

𝛿

means that firms in cross-border transactions are likely to have a lower ETR than firms in completed domestic M&A deals.

In addition, in order to control for other effects, I include several control variables proposed by Richardson, Taylor and Lanis (2013) that impact the tax aggressiveness. They argue that firm size, leverage, capital intensity, R&D intensity, the market-to-book ratio and return on assets are associated with tax aggressiveness. Furthermore, the institutional ownership may also affect the firm performance. In order to overcome this issue, the sample is divided into three subsamples. The first subsample contains firms wholly owned by acquirers (100% acquisition). The second subsample consists of events where parent firms purchase the target firm’s share for the first time (i.e. excluding events where the acquirers increase a holding share. The third subsample involves 100% acquisition while excluding events where acquirers increase the percentage of holding shares. Concerning data availability, for most targets, R&D expenses were missing either because they were not operating in R&D intensive industries or they were not asked to disclose R&D costs. Furthermore, the majority of targets are private and/or subsidiaries of other companies. It is not possible to obtain the market value of these firms and the market-to-book ratio. Consequently, in this study, I include firm size (SIZE), leverage (LEV), capital intensity (CAP), and return on assets (ROA) as the control variables. I use the natural log of total assets as a measurement of SIZE. In Tran’s (1998) study of book-tax income gap causes (as cited in Richardson and Lanis (2007)), he discovers that larger firms have vast resources that allow them to engage in tax-planning investment and arrange

(17)

firms’ activities to reach the optimal tax saving. Therefore, larger firms can gain more from tax-planning investments than small firms. It is expected a negative relationship between ETR and firm size.

Firm’s financing and investment choices could influence ETRs. Specifically, companies with higher leverage are likely to have lower ETRs because of the tax-deductible interest costs. In this study, leverage (LEV) is defined as the sum of long-term debt and current liabilities divided by total assets. Additionally, Stickney and McGee (1982) claim that firms’ investment plan can also affect ETRs. The larger the value of the assets the firm invests, the larger depreciation would be, and larger tax savings can be obtained from depreciation. Hence, the effective tax rate is lower. In this case, we use capital intensity (CAP) to measure the asset value, which is measured as fixed assets divided by total assets.

Return on assets (ROA) is positively related to ETRs since higher return on assets will induce firms to pay more tax, which then lead to higher ETRs. ROA is calculated as pre-tax profits divided by total assets.

Before running the regression, I conduct the correlation test to avoid the multicollinearity issue in this study. As shown in the appendix 2, leverage is negatively correlated with firm size, ROA and capital intensity. As leverage increases, the firm size, ROA and capital intensity drop. The firm size is positively correlated with capital intensity and negatively correlated with ROA, which indicates that the larger the firm size is, the higher capital intensity and lower ROA are. Moreover, ROA is negatively correlated with capital intensity. The higher ROA is, the capital intensity tends to decrease. Overall, the table suggests that there is no multicollinearity problem in this regression.

4.2 Profit shifting

To study whether profit shifting is one of the motivations that drives cross-border acquisition, I estimate the following equation.

(18)

Where

𝑃𝑟𝑜𝑓𝑖𝑡

𝑖𝑡 represents the ratio of net income over total assets for company i at period t,

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡 equals to one if the target firm was acquired by the US firm (i.e. cross-border

acquisition), and

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡 equals to zero if the target firm engaged in domestic acquisition

deals, which is considered as the control group.

𝑃𝑜𝑠𝑡

𝑖𝑡 is a dummy equal to one after the

acquisition. The interaction

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡

×𝑃𝑜𝑠𝑡

𝑖𝑡 measures the expected difference in profit

mean between cross-border acquisition and domestic acquisition over time. Hence

𝛿

=0, which could illustrate that the profitability is indifferent between cross-border M&A and domestic M&A after acquisition. Conversely, if tax rate difference between UK and US is one of the factors that drive cross-border M&A, I expected a positive value,

𝛿

>0. A positive

𝛿

means that firms in cross-border transactions are likely to have a higher profitability than firms in completed domestic M&A deals.

In this model, I include control variables namely British GDP growth rate, the difference between U.S. and UK corporate tax rate, leverage, costs of employment and capital intensity. To begin with the macroeconomic situation, real GDP growth rate is considered as the measure of the country performance. Firms in the period of better economic development should have higher profit. Second, the difference between US and UK corporate tax rate is expected to have positive impact on the firm’s profit as the higher difference in the tax rate, the more likely that U.S. multinationals would shift profit to UK. Third, the leverage is likely to have a negative association with the firm’s profit. The firm is likely to pay more interest on debt when the firm has a higher leverage ratio. Fourth, costs of employment can also influence the profit. Particularly, the firm pays more to employees, the lower the profit retained. Last but not least, the firm could use its revenue to purchase fixed assets such as equipment and/or land. Hence, capital intensity is anticipated to be negatively correlated with profit.

In order to ensure no multicollinearity problem in this equation, I perform the similar correlation test that is presented in section 4.1. The results show that GDP growth rate has low degrees of correlation with leverage, firm size, capital intensity and costs of employment. Furthermore, the difference in tax rates also has relatively low extent of correlation with the other three variables. Additionally, costs of employment is positively related to leverage and

(19)

negatively correlated with capital intensity. The higher costs of employment will induce higher leverage and lower capital intensity. In general, no significant high correlation detected implies that no multicollinearity problem should be considered in this case.

4.3 Debt shifting

To capture a firm’s debt shifting behaviour, I employ the same difference-in-difference equation as stated in the following:

𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒

𝑖𝑡

= 𝛼 + 𝛽𝑇𝑟𝑒𝑎𝑡

𝑖𝑡

+ 𝛾𝑃𝑜𝑠𝑡

𝑖𝑡

+ 𝛿𝑇𝑟𝑒𝑎𝑡

𝑖𝑡

×𝑃𝑜𝑠𝑡

𝑖𝑡

+ 𝜃𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑠 + 𝜀

𝑖𝑡

Eq(4)

Where

𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒

𝑖𝑡 represents book leverage for company i at period t,

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡 equals to

one if the target firm was acquired by the US firm (i.e. cross-border acquisition), and

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡

equals to zero if the target firm engaged in domestic acquisition deals, which considered as the control group.

𝑃𝑜𝑠𝑡

𝑖𝑡 is a dummy equal to one after the acquisition. The interaction

𝑇𝑟𝑒𝑎𝑡

𝑖𝑡

×𝑃𝑜𝑠𝑡

𝑖𝑡 measures the expected difference in mean of leverage ratio between

cross-border acquisition and domestic acquisition over time. Hence

𝛿

=0, which could illustrate that the leverage ratio is indifferent between cross-border M&A and domestic M&A after acquisition. Conversely, if tax rate difference between UK and US is one of the factors that drive cross-border M&A, I expected a negative value,

𝛿

<0. A negative

𝛿

means that, in the case of UK targets and US acquirers, firms in cross-border transactions are likely to have a lower leverage than firms completed domestic M&A deals.

In determining the firm’s leverage ratio, the firm size, profitability and capital intensity would be considered as control variables. Gupta (1969, as cited in Nunkoo and Boateng (2010)) suggests that it costs more for small-sized firms to issue new equity than for large firms. Therefore, small firms are likely to have more leverage than large firms. Furthermore, small-sized firms’ earning volatility is another barrier to issue equity. In addition, profitability has a positive impact on leverage since more profitable firms would expect larger tax shelter and hence greater leverage capacity. Lastly, Bradley et al. (1984) indicate that firms can borrow

(20)

money at lower interest costs if the tangible assets are secured as collateral. Firms that invest heavily in these types of assets would have higher leverage ratio.

Similar to the other two tests discussed above, the correlation test is used. Leverage is negatively correlated with the firm size, capital intensity and profitability. In addition, the firm size has a positive correlation with capital intensity and a negative correlation with profitability. Lastly, the profitability is negatively related to capital intensity. Hence, the regression does not have the multicollinearity issue give the relatively low correlations among independent variables.

5. Results

The following part presents the results of this paper. Firstly, I answer whether target firms in cross-border acquisitions have lower ETRs after acquisition. The last two analyses deal with two channels that acquirers could mitigate ETRs, namely through profit shifting and debt shifting.

5.1 Effective tax rate

The estimates of equation (2) are reported in Table 4. In column (1), the estimates present the results from the model that has no additional control variables. In column (2), I include four variables discussed in the previous section for the whole sample. In column (3), I exclude the acquisition events that were not fully acquired by bidders while keeping four variables as control variables. In column (4), I remove acquisitions that involve increasing the share from a certain percentage to another percentage. In column (5), I eliminate the companies that were not entirely owned by acquirers and the events that acquirers intend to increase the target firms’ shares. In the following estimations, standard errors are corrected for clustering at the firm level.

The estimates in the following five models indicate similar results. Both Treat and Post variables exhibit statistically significant at the 5% level. It means that, on average, the target firms engaged in cross-border M&A has higher ETRs before the acquisition than target firms acquired by domestic firms. Moreover, the coefficient of Post indicates that, on average in the domestic M&A deals (the control group), the effective tax rate is around 10% higher after

(21)

acquisition than before acquisition. The estimates of the interaction

𝛿

shown in Table 4 are all negative. However, only in the case of 100% acquisition (column 3), it shows a statistically significant

𝛿

at the 10% level. The coefficient of the interaction in column 3 illustrates that, on average, the ETR of target firms in cross-border acquisition is 5% lower after completing the acquisition compared to the target firms in domestic acquisition deals. To some extent, the finding is not consistent with the results from Erel et al. (2012), in which they argue that the larger difference in corporate tax rate, the more likely to attract foreign investment. The potential reason of the inconsistency is the different measurement for the tax. They use the difference in tax rate while, in this paper, I use the effective tax rate. Such noticeable differences between partially and wholly owned subsidiaries are also observed by Desai et al. (2004), which study the elements of partial ownership of the foreign subsidiaries of U.S multinationals. Hence, another potential reason for the difference could be expensive coordination costs arising from minority owners. First, costly frictions may arise among minority owners and parent firms because the parent firms intend to shift profits away from affiliates with minority owners. Second, multinational firms desire to locate their production worldwide and this may create another conflict with minority owners.

Furthermore, the models provide no evidence for leverage and firm size. The insignificant finding for firm size is similar to the finding from Stickney and McGee (1982), which they detect that firm size is less important in analysing diversity in effective tax rates. They suggest that the variation in capital intensity may clarify the size effect given the relatively high correlation between the firm size and the capital intensity, as shown in the appendix 2. The coefficients of leverage show insignificant negative signs, which is contradictory to outcomes from Stickney and McGee (1982). This may be due to the different definition used for leverage, where they define leverage as long-term debt divided by stockholders’ equity. In addition, the models support ROA and capital intensity arguments. Both ROA and capital intensity present statistically significant coefficients at the 1% level.

(22)

Table 4: The impact of cross-border acquisition on target firms’ effective tax rate

The table presents difference-in-difference regression estimates with effective tax rate (ETR) as the dependent variable. Column (1) has no control variables. Column (2) adds 4 control variables, namely leverage, firm size, ROA, capital intensity, for all M&A. Column (3) adds 4 control variables for wholly owned M&A (i.e. 100% acquisition). Column (4) adds 4 control variables while excluding increasing share acquisitions. Column (5) adds 4 control variables for wholly owned acquisition while excluding increasing share acquisition events. Standard errors are shown in parentheses and are clustered at firm level. *, **, *** denote statistically significant at the 10%, 5%, and 1% level, respectively.

(1) (2) (3) (4) (5)

Sample All All Wholly

owned Exclude increasing share acquisitions Wholly owned and exclude increasing share acquisitions Treat 0.051** (0.198) 0.058** (0.024) 0.065*** (0.024) 0.062** (0.024) 0.063*** (0.024) Post 0.090*** (0.0137) 0.091*** (0.017) 0.095*** (0.017) 0.095*** (0.017) 0.097*** (0.017) Treat×Post -0.041 (0.275) -0.045 (0.033) -0.054* (0.034) -0.043 (0.034) -0.049 (0.035) Leverage -0.013 (0.006) -0.018 (0.017) -0.016 (0.017) -0.019 (0.017) Size -0.001 (0.005) -0.0002 (0.005) -0.001 (0.005) 0.001 (0.005) ROA -0.002*** (0.0003) -0.002*** (0.0003) -0.002*** (0.0003) -0.002*** (0.0003) Capital intensity 0.068** (0.030) 0.071** (0.030) 0.080*** (0.030) 0.082*** (0.030) Constant -0.239*** (0.011) -0.234*** (0.078) -0.239*** (0.080) -0.268*** (0.081) -0.260*** (0.083) No. of observations 5,745 5,436 5,232 4,922 4,792 R2 0.0086 0.0200 0.0217 0.0233 0.0238

Controls No Yes Yes Yes Yes

5.2 Profit shifting

I present the estimates of equation (3) in Table 5. The estimates presented in column (1) have no additional control variables. In column (2), I include five variables, which are leverage, costs of employment, UK corporate tax rate, UK real GDP growth rate and capital intensity, for the whole sample. In column (3), I exclude the acquisition events that were not fully acquired by bidders while keeping the mentioned five variables as control variables. In column (4), I remove acquisitions that involve increasing the share from a certain percentage

(23)

to another percentage. In column (5), I eliminate the companies which were not entirely owned by acquirers and the events that acquirers intend to increase the target firms’ shares. In the following estimations, standard errors are corrected for clustering at the firm level.

The estimates from column (1) to column (2) show indifferent findings. The coefficients on Treat interacted with Post are all positive, indicating that the profitability of cross-border target firms increases more than the profitability of domestic target firms. The difference in changes of profitability between cross-border deals and domestic deals are significant at 10% significance level. The results demonstrate US acquirers may shift profits to UK. The findings emphasize the important roles of leverage, costs of employment, difference in tax rates, and capital intensity on profitability. These variables are statistically significant at 1%, 5%, 10%, and 1% level, respectively. Higher leverage, cost of employment and capital intensity will lower the firm’s profitability. The results further confirm that the larger difference in US and UK corporate tax rate, the higher profitability after acquisition. Nevertheless, in all subsamples, the results do not reveal the evidence for domestic real GDP growth.

Table 5: The impact of cross-border acquisition on target firms’ profit

The table presents difference-in-difference regression estimates with profit as the dependent variable. Column (1) has no control variables. Column (2) adds 4 control variables, namely leverage, employee costs, UK corporate tax rate, real GDP growth, capital intensity, for all M&A. Column (3) adds 5 control variables for wholly owned M&A (i.e. 100% acquisition). Column (4) adds 5 control variables while excluding increasing share acquisitions. Column (5) adds 5 control variables for wholly owned acquisition while excluding increasing share acquisition events. Standard errors are shown in parentheses and are clustered at firm level. *, **, *** denote statistically significant at the 10%, 5%, and 1% level, respectively.

(1) (2) (3) (4) (5)

Sample All All Wholly

owned Exclude increasing share acquisitions Wholly owned and exclude increasing share acquisitions Treat -0.037** (0.015) -0.1627* (0.096) -0.183* (0.099) -0.183 (0.104) -0.197 (0.103) Post 0.001 (0.011) 0.0002 (0.010) 0.0002 (0.010) -0.0002 (0.010) -0.00009 (0.010) Treat×Post 0.031 (0.021) 0.072* (0.027) 0.077* (0.045) 0.082* (0.046) 0.081* (0.046) Leverage -0.179*** (0.020) -0.180*** (0.021) -0.179*** (0.020) -0.180*** (0.021) Costs of -0.052** -0.062** -0.054** -0.064**

(24)

Difference in tax rate 1.170* (0.693) 1.329* (0.707) 1.280** (0.748) 1.393* (0.738) GDP growth 0.051 (0.373) 0.083 (0.3381) 0.070 (0.404) 0.145 (0.361) Capital intensity -0.105*** (0.017) -0.104*** (0.017) -0.107*** (0.018) -0.105*** (0.018) Constant -0.022*** (0.008) 0.200*** (0.019) 0.203*** (0.019) 0.205*** (0.019) 0.206*** (0.019) No. of observations 5,745 5,436 5,232 4,922 4,792 R2 _0.0012 _0.2660 _0.2781 _0.2743 _0.2860

5.3 Debt shifting

The estimates of equation (4) are shown in Table 6. Column (1) displays estimates that have no additional control variables. Column (2) examines the difference-in-difference estimates on leverage while including firm size, profitability, and capital intensity as control variables. Columns (3) through (5) examine subsamples of 100% acquisitions, acquisitions without increasing share, and 100% acquisition while excluding acquisition without increasing share. In the following estimations, standard errors are corrected for clustering at the firm level. The results show that cross-border acquisitions do not decrease leverage ratio of target firms compared with firms that completed domestic acquisition given the statistically insignificant coefficient on Treat interacted with Post. This finding is not consistent with arguments from Huizinga et al. (2008). The potential reason could be explained by Ruf (2011). He argued that multinationals could form a tax unity that includes all subsidiaries in the host country. Hence, the individual subsidiary debt may be consolidated at the level of the holding company (As shown in the appendix 4). In addition, the following estimates prove that firm size and profitability are critical determinants of firms’ leverage ratio. This finding is consistent with the results of Nunkoo and Boateng (2010). Moreover, column (2) to column (4) fail to provide evidence for capital intensity. Hence, I do not find any evidence that US parent firms intend to decrease the leverage ratio after acquisition using the domestic target firms as the control group.

(25)

Table 6: The impact of cross-border acquisition on target firms’ leverage ratio

The table presents difference-in-difference regression estimates with leverage as the dependent variable. Column (1) has no control variables. Column (2) adds 4 control variables, namely firm size, profitability, capital intensity, for all M&A. Column (3) adds 4 control variables for wholly owned M&A (i.e. 100% acquisition). Column (4) adds 4 control variables while excluding increasing share acquisitions. Column (5) adds 4 control variables for wholly owned acquisition while excluding increasing share acquisition events. Standard errors are shown in parentheses and are clustered at firm level. *, **, *** denote statistically significant at the 10%, 5%, and 1% level, respectively.

(1) (2) (3) (4) (5)

Sample All All Wholly owned Exclude

increasing share acquisitions Wholly owned and exclude increasing share acquisitions Treat 0.170*** (0.041) (0.067) 0.091 (0.067) 0.097 (0.070) 0.087 (0.070) 0.088 Post -0.018 (0.028) 0.017 (0.027) 0.018 (0.027) 0.025 (0.028) 0.026 (0.028) Treat×Post -0.044 (0.057) -0.013 (0.128) -0.008 (0.132) 0.007 (0.138) 0.006 (0.137) Size -0.057*** (0.01 -0.054*** (0.011) -0.060*** (0.012) -0.056*** (0.137) Profitability -1.377*** (0.350) -1.420*** (0.361) -1.417*** (0.362) -1.460*** (0.374) Capital intensity -0.098 (0.070) -0.091 (0.072) -0.071 (0.076) -0.069 (0.077) Constant 0.637*** (0.022) 1.662*** (0.194) 1.604*** (0.188) 1.710*** (0.205) 1.649*** (0.199) No. of observations 5,745 5,436 5,232 4,922 4,792 R2 _0.0052 _0.2740 _0.2830 _0.2805 _0.2891

6. Conclusion

This paper studies whether tax avoidance is one of the determinants of cross-border M&A. Managers and shareholders may intend to implement cross-border acquisitions to minimize the firms’ tax payments and therefore increase the firms’ total returns. Due to financial disclosure requirements among European countries, it is feasible to compose a sample of UK acquisitions containing financial information on target firms before and after acquisition. I use this sample to test if US firms may behave tax aggressively to acquire UK firms in order to lower their tax burdens by using difference-in-difference methods. Specifically, I test if the target firms acquired by US firms would exhibit lower ETRs than target firms acquired by domestic firms after acquisitions. Furthermore, I examine whether the US parent firms would

(26)

transfer their profit to UK affiliates to reduce their tax burden. Alternatively, US parent firms can also diminish the target firms’ debt level as they can raise new debt in US to generate higher tax shields.

First, this paper documents that, on average for a 100% acquisition, a target firm’s ETR in a cross-nation acquisition declines by 5 percentage points after completion compared to a target firm’s ETR in domestic acquisitions. However, the results do not reveal evidence for other subsamples. The potential underlying reason could be the upscale coordination costs stemming from other minority owners. For example, the parent firms may plan to relocate profits or production away from minority owners. The conflicts between minority owners and parent firms could further lower affiliates’ profits. Hence, effective tax rate is indifferent between cross-border acquisitions and domestic acquisitions in other subsamples.

Second, this paper captures the fact that a target firm’s profitability in cross-border deals would increase on average by 8% post acquisitions, in comparison with a target firm in domestic deals. This finding suggests that US firms transfer profits to UK subsidiaries in order to mitigate the tax burdens. However, the results fail to present evidence for debt shifting hypothesis. In general, the results do not demonstrate significant changes in a target firm’s leverage after deals. This may due to the UK tax unity regimes introduced in 2006. The debt level of subsidiary is consolidated at the country level.

(27)

Reference:

Bouraoui, T. & Li, T. (2014) The impact of adjustment in capital structure in Mergers & Acquisition on US acquirers’ business performance, The Journal of Applied Business Research, 30(1), 27-41.

Bradley, M., Jarrell, G. A., & Kim, E. H. (1984). On the existence of an optimal capital structure: Theory and evidence. The Journal of Finance, 39(3), 857-878.

Chen, S., Chen, X., Cheng, Q. & Shevlin, T. (2010). Are family firms more tax aggressive than non-family firms? Journal of Financial Economics, 95 (1), 41 – 61.

Desai, M. A., Foley, C. F., & Hines Jr, J. R. (2004). The costs of shared ownership: Evidence from international joint ventures. Journal of Financial Economics, 73(2), 323-374.

Devos, E., Kadapakkam, P. R., & Krishnamurthy, S. (2009). How do mergers create value? A comparison of taxes, market power, and efficiency improvements as explanations for synergies. The Review of Financial Studies, 22(3), 1179-1211.

Dyreng, S. D., Hanlon, M. & Maydew, E. L. (2010). The effects of executives on corporate tax avoidance. Accounting Review, 85 (4), 1163 – 1189.

Erel, I., Jang, Y., & Weisbach, M. S. (2015). Do acquisitions relieve target firms’ financial constraints?. The Journal of Finance, 70(1), 289-328.

Erel, I., Liao, R. C., & Weisbach, M. S. (2012). Determinants of cross‐border mergers and acquisitions. The Journal of Finance, 67(3), 1045-1082.

Fuest, C., Hebous, S., & Riedel, N. (2011). International debt shifting and multinational firms in developing economies. Economics Letters, 113(2), 135-138.

Grubert, H. (2013). Foreign taxes and the growing share of US multinational company income abroad: Profits, not sales, are being globalized. National Tax Journal, 65(2), 247–281.

(28)

Gugler, K., & Konrad, K. A. (2002). Merger target selection and financial structure. University of Vienna and Wissenschaftszentrum Berlin (WZB).

Gupta, M. C. (1969) The effect of size, growth, and industry on the financial structure of manufacturing companies, Journal of Finance, 24, 517–29.

Gupta, S., & Newberry, K. (1997). Determinants of the variability in corporate effective tax rates: Evidence from longitudinal data. Journal of Accounting and Public Policy, 16(1), 1-34. Hanlon, M., & Slemrod, J. (2009). What does tax aggressiveness signal? Evidence from stock price reactions to news about tax shelter involvement. Journal of Public Economics, 93(1-2), 126-141.

Harris, D., Morck, R., & Slemrod, J. B. (1993). Income shifting in US multinational corporations. In Studies in international taxation (pp. 277-308). University of Chicago Press. Harris, R. S., & Ravenscraft, D. (1991). The role of acquisitions in foreign direct investment: Evidence from the US stock market. The Journal of Finance, 46(3), 825-844.

Huizinga, H. & Laeven, L. (2008). International profit shifting within multinationals: A multi-country perspective. Journal of Public Economics, 92 (5/6), 1164 – 1182.

Huizinga, H., Laeven, L., & Nicodeme, G. (2008). Capital structure and international debt shifting. Journal of Financial Economics, 88 (1), 80 – 118.

Jog, V., & Tang, J. (2001). Tax reforms, debt shifting and tax revenues: multinational corporations in Canada. International Tax and Public Finance, 8(1), 5-25.

Martin, X., Tang, Y., Wang, C., & Zou, H. (2017). Does Target Tax Aggressiveness Matter in Corporate Takeovers?. SSRN Working Paper.

Moore, P. J., & Ruane, F. (2005). Taxation and the financial structure of foreign direct investment. Trinity College Dublin, IIIS Discussion Paper 88.

(29)

Nunkoo, P. K., & Boateng, A. (2010). The empirical determinants of target capital structure and adjustment to long-run target: evidence from Canadian firms. Applied Economics Letters, 17(10), 983-990.

Richardson, G., & Lanis, R. (2007). Determinants of the variability in corporate effective tax rates and tax reform: Evidence from Australia. Journal of Accounting and Public Policy, 26(6), 689-704

Richardson, G., Taylor, G., & Lanis, R. (2013). The impact of board of director oversight characteristics on corporate tax aggressiveness: An empirical analysis. Journal of Accounting and Public Policy, 32(3), 68-88.

Rossi, S., & Volpin, P. F. (2004). Cross-country determinants of mergers and acquisitions. Journal of Financial Economics, 74(2), 277–304.

Ruf, M. (2011). Why is the Response of Multinationals' Capital-Structure Choice to Tax Incentives That Low? Some Possible Explanations. FinanzArchiv: Public Finance Analysis, 67(2), 123-144.

Schindler, D., & Schjelderup, G. (2012). Debt shifting and ownership structure. European Economic Review, 56(4), 635-647.

Scholes, M. & Wolfson, M. (1990). The effects of changes in tax laws on corporate reorganization activity, The Journal of Business, 63(1), S141-S164.

Stickney, C., & McGee, V. (1982). Effective corporate tax rates: the effect of size, capital intensity, leverage and other factors, Journal of Accounting and Public Polity, 1(2), 125-152. Swenson, C. (1999). Increasing stock market value by reducing effective tax rates. Tax Notes, 83 (June 7), 1503-1505.

(30)

Appendix 1: USA and GBR Tax Rates over 2000-2017

Table 7:

Figure 1:

Source: Oxford University Centre for Business Taxation (

https://www.sbs.ox.ac.uk/faculty-research/tax/publications/data ) 0% 5% 10% 15% 20% 25% 30% 35% 40% USA GBR USA GBR 2000 35% 30% 2001 35% 30% 2002 35% 30% 2003 35% 30% 2004 35% 30% 2005 35% 30% 2006 35% 30% 2007 35% 30% 2008 35% 30% 2009 35% 28% 2010 35% 28% 2011 35% 28% 2012 35% 26% 2013 35% 24% 2014 35% 23% 2015 35% 21% 2016 35% 21% 2017 35% 20%

(31)

Appendix 2: Correlation among variables

ETR Leverage Size ROA CINT Costs of

employment Profitability Growth rate Difference in tax rates ETR 1.0000 Leverage 0.0211 1.0000 Size 0.0323 -0.1199 1.0000 ROA -0.0994 -0.3752 -0.1210 1.0000 CINT 0.0541 -0.0061 0.3766 -0.2216 1.0000 Costs of employment -0.0453 0.1971 -0.4352 0.0080 -0.2463 1.0000 Profitability -0.0687 -0.3761 -0.1045 0.9551 -0.2012 -0.0016 1.0000 Growth rate -0.0144 0.0070 0.0760 -0.0050 0.0275 -0.0201 -0.0095 1.0000 Difference in tax 0.0195 0.0066 -0.0096 0.0254 -0.0576 -0.0297 0.0215 0.0126 1.0000

Given the relatively low correlations between variables shown in the above table, the multicollinearity issue is not the case in this study.

(32)

Appendix 3: Definition of variables

Variable Description

ETR Tax expenses /earning before taxes (Source:

Amadeus database)

Leverage Long-term debt (LTDB)+ Current liabilities

(CULI)/ Total assets (Source: Amadeus

database)

Size Natural logarithm of total assets (Source:

Amadeus database)

ROA EBIT/Total assets (Source: Amadeus database)

CINT Fixed assets (FIAS)/Total assets (Source:

Amadeus database)

Costs of employment Costs of employment (STAF)/Total assets

(Source: Amadeus database)

Profitability Net income/ Total assets (Source: Amadeus

database)

UK GDP Growth rate Annual percentage real GDP growth rate in local

currencies (Source: World Bank)

Difference in tax rates The difference between UK and US corporate tax

rates (Source: Oxford University Centre for Business Taxation)

(33)

Appendix 4: Tax Unities

Figure 2: US parent firms Holding company 1

Sub 1 Sub 2 Sub 3

UK

Holding company 2

Under a profit shifting agreement between holding companies and affiliate companies, multinationals could form a tax unity including all UK subsidiaries. Hence, the taxable profit of all UK subsidiaries is consolidated at the level of the holding company 1. Consequently, it may fail to capture multinationals tax planning activities for a specific affiliate

(34)

Tax aggressiveness in cross-border mergers and acquisitions : the UK and US evidence