Cross-border Mergers and Acquisitions: The Role of Exchange Rate Movement

Cross-border Mergers and Acquisitions: The Role of Exchange Rate Movement Anand Shetty, PhD Professor of Finance School of Business Iona College 715 North Ave. New Rochelle, NY, USa 10801 ashetty@iona.edu John Manley, PhD Professor of Finance School of Business Iona College 715 North Ave. New Rochelle, NY, USA 10801 jmanley@iona.edu NyoNyo Kyaw Associate Professor of Finance School of Business Iona College 715 North Ave. New Rochelle, NY, USA 10801 nkyaw@iona.edu EFM Classification codes: 610 & 210

2 Cross-border Mergers and Acquisitions: The Role of Exchange Rate Movement This paper investigates the role of exchange rate movement in cross-border mergers and acquisitions decisions and their outcomes. Past studies on theoretical and empirical relationship between the exchange rate and cross-border investments have reported mixed results. We examine this relationship using 2001-2011 cross-border mergers and acquisitions by 595 U.S. firms. We look at two measures of exchange rate changes the change in the real exchange rate of the U.S. dollar and exchange rate volatility to assess the impact on merger premium. We do not find a significant relationship between the change in the real value of the U.S. dollar and merger premium. The relationship between the volatility measure and merger premium is, however, positive and significant. Our model also examines the effects of bidder specific characteristics such as size, leverage, overseas experience overseas experience in addition to the means of acquisition finance and the level of economic development in the target country. We find a positive and significant effect of the overseas experience and for the choice of MAF (Means of Acquisition Financing), as well as a negative insignificant effect of the level of economic development. 1. Introduction Both domestic and cross-border mergers and acquisitions are motivated by a common goal of increasing value for the shareholders, but the wealth effects of cross-border investments are affected by a number of factors that are not present in domestic mergers and acquisitions. Cultural, geographic, regulatory, and valuation differences, economic development, currency movement, degree of market integration, accounting standards, tax laws are some of the factors that dominate the wealth effects of cross-border mergers. 1 The traditional models on international capital flows operating under the assumptions of perfect capital mobility and perfect capital markets do not provide cost of capital advantage to either the domestic or foreign investor in bidding for an asset (Frootand Stein (1991). In the absence of 1 See Erel et al. 2012 for a detailed discussion of factors.

3 cost of capital advantage, motivations for cross- border investments come from factors ranging from imperfections in the production and factor markets (Kindleberger 1969, Caves 1971 and Hymer1976), industrial organization, and differences in tax and regulatory policies (Scholes and Wolfson 1990). In recent years, the role of exchange rate movement has received substantial attention in the foreign direct investment and international merger literature. Observing the trend in foreign direct investment (FDI) and currency movement in the late 1970s and 80s, Froot and Stein (1991) build a model to explain the link between the FDI flows to exchange rate movement. They argue that information asymmetries about an asset s pay-offs makes it costly for an entrepreneur to finance the asset s acquisition with external sources alone. The more net worth the entrepreneur brings to this investment, the lower will be the cost of capital and the greater will be the relative advantage in purchasing this asset. When the currency of a country appreciates, firms holding most of their wealth in domestic currency denominated assets will experience an increase in their relative wealth position, hence the ability to bid aggressively for foreign assets. Regressing the foreign direct investments in the U.S. from 1973 to 1988 on the real exchange rate of U.S. dollar, they find a significant negative relationship between the FDI in the U.S. and the real value of the dollar in support of the connection between FDI and the exchange rate movement. 2 Blonigen (1997) also provides empirical evidence in support of an inverse relationship between domestic currency value and inward FDI. Blonigen offers a different perspective on the effect of 2 Scholes and Wolfson (1990) confirm this relationship in their observation that the cross-border takeover in the USA slowed down in the early 1980s when the dollar was strong and then surged when the dollar weakened in the latter half of the 1980s.

exchange rate on FDI by separating the price 4 (wealth) effect from the return effect. When the domestic currency appreciates, the domestic firm can aggressively bid for a foreign asset because of the wealth effect, but there is no guarantee that the investment will generate a higher rate of return. When the profits of the acquired firm are translated at the higher valued domestic currency, the returns could be adversely affected. If the acquisition involves a firm-specific asset which is transferable, it can generate profits in other currencies thus avoiding the return problem. Bonigen s emphasis on firm-specific asset acquisition in FDI is designed to focus on the fact that returns on international acquisitions also matter in addition to the price paid for the assets. Other FDI studies which used the exchange rate as a regressor with mixed results include Caves (1989), Ray (1989) and Martin (1991). Since the findings of Foot & Sterns (1991), Blonigen (1997) and others 3 that there exists a connection between currency movement and cross-border capital flow, many papers studying the stockholder benefits from cross-border mergers and acquisitions have introduced exchange rate as one of the explanatory variables. In their study comparing the wealth gains to the stockholders of U.S. targets from domestic takeovers to the benefits from acquisitions by Japanese firms during late 70 s and 1980 s, Harris and Revenscraft (1991) find a strong and positive relationship between the strength of the yen and wealth gains to U.S. targets. Cakici et al. (1996), report a negative and significant exchange rate effect on bidder s abnormal return in univariate regression and no effect in a multivariate regression for foreign acquisitions in the U.S. from 1983 to 1992. Pettaway et al. (1993) find an insignificant positive relationship between a strong yen and the wealth gains to stockholders of Japanese acquirers of U.S. assets in 3 See Klein and Rosengren (1994) and Dewenter (1995)

the 1980 s. Kang (1993) finds a positive and 5 significant relationship between Japanese bidder gains and appreciation of the yen during Japanese acquisitions of U.S. firms in late 1970 s and the 1980 s. HalilKiymaz (2004) reports a positive but not significant effect of exchange rate variable on the bidder s wealth gains in his study covering cross-border acquisitions of U.S. financial institutions during 1989-1999 period. Moeller et al. (2005) find no significant effect of the strong dollar on the gains to U.S. bidders during 1985-95 period. In a comprehensive study of the determinants of cross-border mergers and acquisitions spanning a period from 1990 to 2007, Erol et al. (2012) observe that currency movement is a major factor determining the pattern of cross-border mergers. They find the firms from countries whose currencies have appreciated are more likely to acquire firms from countries whose currencies have depreciated. Sonenshine et. al. (2014) report that firms pay more for target firms as the exchange rate of the home country of the target firm appreciates. They also find that acquirers with high intangible asset intensity pay relatively more for targets than those with low intangible asset intensity when target firm currency appreciates. In this paper, we extend the examination the effects of exchange rate movements on the wealth gains in 595 foreign acquisitions of U.S. firms during 2001-11 period. This period has seen a significant amount of fluctuation in the value of U.S. dollar. The real trade-weighted U.S. dollar index declined to mid and low eighties during the financial crisis period after hovering around high nineties during the first half of the sample period. We also examine how gains to the bidder are affected by (i) bidder characteristics such as size, leverage and foreign operating experience, (ii) the means of acquisition financing, and (iii) the level of economic development of target s country. For the exchange rate variable, we construct two measures of real exchange rate

change. The first measure is constructed as the 6 change in the real exchange rate of the U.S. dollar relative to the target currency during the year of the announcement of the acquisition and it is similar to the measure used by Froot and Stein (1991) and Sonenshine et al. (2014). The second measure is obtained by using the two-step procedure used by Harris and Ravenscraft (1991). The real exchange rate prevailing in the year of the announcement of the acquisition is subtracted from the average real exchange rate for the entire sample period (2001-11) and the difference is divided by the average real exchange rate of the same period. A positive value indicates that the U.S. dollar is stronger relative to the foreign currency and a negative value indicates the opposite. Cakici (1996) and Kang (1993) use this method. Following Blonigen (1997) observation that while a strong currency may help an acquirer bid aggressively for a foreign frim and pay a higher price, there is no guarantee that the investment will generate a higher rate of return. If the currency remains strong, the expected future cash flows from the subsidiary would have a lower discounted value. The merger return, therefore, depends not only on the current value of the currency, but also on its future changes. The higher the variability of the currency, the higher is the uncertainty shout the expected future cash flows, hence the uncertainty about the return on the investment. The measurement of the impact of exchange rate movement on the merger outcome should involve both the changes in the level and the variability of the exchange rate. We, therefore, introduce a measure to capture the effect of variability (risk) which is represented by the standard deviation of the average of weekly percent changes in the nominal exchange rate during the year of the announcement year. Halil Kiymaz (2004) also uses a currency variability measure and finds a significant relationship between exchange rate volatility and wealth gains of the acquiring firm.

7 The paper is organized as follows. Section 2 presents data and methodology. Section 3 presents the analyses of empirical results regarding wealth gains. Section 4 outlines the econometric methodology and describes the factors included in the regression model. Section 5 provides a summary of findings of regression analysis. Section 6 has the summary and the conclusions. 2. Data and Methodology 2.1.Data Selection The sample consists of U.S. bidders of foreign targets in international mergers and acquisitions during the 2001-11 period. The sample is obtained from Bloomberg s merger data base. The original sample consisted 750 firms from 44 countries. The sample is restricted by the inclusion of only those firms (1) not involved with another merger at the time of the announcement, (2) not involved with any significant event within two months of the merger, (3) not acquired or turned private during the months measured, and (4) with a history of performance for at least five years prior to the announcement to provide necessary data for the cross-sectional variables to determine the expected mean excess returns. This sample is reduced to 595 when all the necessary data on dependent and explanatory variables involved in the analysis are collected. Table 1 reports the mergers and acquisitions by (i) year, (ii) developed and developing countries, and (iii) pre-crisis (2001-2007) and post-crisis (2008-2011) periods. The stock s abnormal return data is obtained from the CRSP data base supplied by the University of Chicago Booth School of Business. The financial statements of bidder firms are obtained from Compustat. The data on exchange rates are obtained from two sources, weekly nominal exchange rate data from the Pacific Exchange Rate Service provided by the University of British Columbia s Sauder School

of Business and the annual real exchange rate 8 data from the U.S. Department of Agriculture s Economic Research Service. The information on whether the target firm is from developed or developing country is obtained from the U. N. Department of Economics and Social Affairs. Table 1. Mergers by year, developed and developing target countries Ye a r T ota l Numbe r of M&As Developed Developing Period 2001 34 27 7 2002 34 30 4 2003 28 24 4 2004 63 55 8 Pre-Crisis Total: 387 2005 58 45 13 2006 79 58 21 2007 91 75 16 2008 84 69 15 2009 42 32 10 Post-Crisis Total: 208 2010 78 58 20 2011 4 4 0 T ota l 595 477 118 595 2.2. Methodology We measure the daily abnormal return to the acquirer s stockholders over two symmetric event periods, 4. The two periods are t = -10 to t+10 and t= -1 to t= +1 with t = 0 being the announcement date. The cumulative abnormal return (CARi) for each firm i is obtained by summing the daily abnormal returns (ARit) over the event period: CARi CARi 20 t=-20arit for 21-day window (1) 1 t=-1arit for 3-day window (2) A daily average abnormal return (AAR) for each day (t) is calculated by averaging the ARs across the firms: AARt 67 i=1ari)/n (3) 4 Lauterbach, Malitz and Long (1991), in a working paper, demonstrate a symmetric event period is less arbitrary and therefore more appropriate for event studies.

9 3. Empirical Results and Discussion Table 2 reports the summary statistics of CARs in the two event windows. Table 2: Summary Statistics Variable N Mean Std Dev Minimum Maximum 21 Day CAR 595 0.0160 0.0953-0.3895 0.3385 3 Day CAR 595 0.0022 0.0494-0.2462 0.2759 DIO 595 41.6546 22.7378 0.0000 100.0020 dev 595 0.8017 0.3991 0.0000 1.0000 LRdif 595 0.0160 0.1017-0.5145 0.6443 SizeRatio 595 0.7980 0.2655 0.1076 2.0610 MAF 595 0.8437 0.3635 0.0000 1.0000 EX1 595-2.5553 6.6913-17.6000 24.3000 EX1New 595 0.0367 0.1320-0.4370 0.8700 EX2R 595-0.5755 15.7467-384.0000 0.9710 Table 3 reports average cumulative abnormal returns of 595 acquiring firms with t-statistics for both event windows. Panel A contains the information on wealth gains during the pre-crisis (2001-2007) and the post-crisis period (2008-11). Panel B contains the information on average wealth gains from acquisitions in developed and developing countries. [Table 3 reports cumulative abnormal stock returns as well as the t-statistics. Out of the 20 event days, there is only one daily average abnormal stock return that is significant (at the 90% level) and it is negative. However, two-thirds of the AARs have positive signs, and the other third have negative signs. In addition, there is no indication of a pattern of significance or signs in the daily average abnormal returns. These results are consistent with results reported in previous studies. 5 ] 5 Dodd (1980) and Asquith (1981) reports weak positive abnormal returns for days t-1 and t0; Eger (1983) reports significant positive abnormal returns for days t-1 and to; whereas Travlos (1987) finds significant negative abnormal returns for days t-1 and t0. In contrast, we do not find significant returns on any one of these three days: t-1, t0, or t1.

Table 3. 10 Panel A. Average cumulative abnormal returns (CARs) for the full period and two sub-periods Window Full period (2001-2011) Post-Crisis: 2008-2011 Pre-Crisis: 2001-2007 Mean Diff (-1, +1) Mean/Mean Diff 0.0022-0.0035 0.0052-0.0087 t-stats 1.0812-0.9548 2.1662-2.0539 (-10, +10) p-value 0.2801 0.3408 0.0309** 0.0404** Mean/Mean Diff 0.0160 0.0214 0.0130 0.0084 t-stats 4.0869 3.2257 2.6980 1.0250 p-value 0.0000*** 0.0015*** 0.0073*** 0.3058 Panel B. Average cumulative abnormal returns (CARs) for acquisitions in the developed and developing countries Full period Window (2001-2011) Developed Developing Mean Diff (-1, +1) Mean/Mean Diff 0.0022 0.0019 0.0032-0.0013 t-stats 1.0822 0.8743 0.6513-2.0539 (-10, +10) p-value 0.2801 0.3824 0.5161 0.0404** Mean/Mean Diff 0.0160 0.0136 0.0255-0.0118 t-stats 4.0869 3.1528 2.7966-1.0250 p-value 0.0000*** 0.0017*** 0.0060*** 0.3058 Panel A of Table 3 shows that the average cumulative abnormal return for the entire sample is positive for both windows and it is 0.016% and significant for (-10, +10) window. It is consistent with the general perception that there are benefits to be derived from overseas expansion as a result of international diversification, economies of scale, market imperfections and international network. Kiymaz (2004) also finds positive CARs for U.S. bidders in both event windows. These results are different from the findings of Cakici et al.(1996) and Doukas and Travlos (1988). Cakici et al. (1996) report a negative and insignificant CARs for U.S. acquirers. Doukas and Travlos (1988) report positive and insignificant abnormal returns to U.S.

11 bidders. Cakici et al. (1996), Kang (1993), and Pettway et al. (1993) report that foreign acquirers of U.S firms enjoy significant wealth gains. Panel A also presents CARs by sub-periods: post-crisis and pre-crisis. We bifurcated the sample this way because of the market turmoil witnessed after 2007 as a result of sub-prime mortgage problem. During the post-crisis period (2008-2011), the U.S dollar was more volatile and suffered some decline compared to pre-crisis period (2001-2007). During the sub-period 2001-2007, the average cumulative abnormal return is positive and significant over both event windows. The same is not true about the sub-period 2008-2011. It is negative and not significant for (-1, +1) window and positive and significant for (-10, +10) window. The difference between the two sub-periods is significant for (-1, +1) window. It is significant to note that 65% of the sample is concentrated in the first sub-period. Economic stability seems to spawn more merger activities and they are also more rewarding. Panel B of Table 3 presents CARs for acquisitions in developed and developing countries. We report a positive average cumulative abnormal returns of 0.0136% and 0.0.0255% for developed and developing countries respectively for (-10, +10) window and both are statistically significant. The yield from developing countries is higher, although the difference is not statistically significant. Over the (-1, +1) period, the average yield from developing countries is still higher but neither is statistically significant and the difference, however, is significant. Kiymaz (2004) reports positive and significant wealth gains from acquisitions in Latin American countries and positive but insignificant wealth gains from acquisitions in Europe and East Asia. Sonenshine (2014) finds the average deal premium paid for deals in developing and developed

countries is statistically identical. Doukas and 12 most when the acquisitions are in the developing countries. Travlos (1988) report that the firms benefit the 4. Explanatory Variables and the Regression Model The wealth gains from international mergers and acquisitions are affected by numerous factors. Imperfections in product and capital markets, relative size of the target, leverage, the means of financing, prior overseas experience, level of economic development of the target country, type of assets acquired, industry characteristics and taxation are some of the widely employed factors in international merger studies. The exchange rate movement has also figured prominently in some recent studies. In this study, the exchange rate variable has received a primary focus. We also include in the regression model a few other variables that other studies have found to be important in determining wealth gains in cross-border mergers. The explanatory variables of the regression model are defined and explained below. 4.1 Exchange rate As described in the introduction, we employ three measures of exchange rate variable. The first is the percent change in the real exchange rate of U.S. dollar expressed in terms of the foreign currency during the year of announcement (EX1), and it is similar to the exchange rate measure used by Froot and Stein (1991). The second measure is the proportionate deviation of the real value of the U.S. dollar in the merger year from its average value over the entire sample period relative to the foreign currency (EX2). A positive (negative) value of EX2 indicates that the U.S. dollar is strong (weak) against the foreign currency. These two measures represent the change in the wealth position of the acquiring firm relative to the target and cost of capital advantage when the firm s currency appreciates as pointed out by Froot and Stein (1991). The third measure is

constructed to represent the effect of variation 13 in the nominal exchange on the bidder s return. This measure is an average of weekly per cent change in the nominal exchange rate in the year preceding the announcement year (EX3). The exchange rate in this case is measured as the value of the foreign currency relative to the U.S. dollar. The higher the volatility, the higher the uncertainty about the cash flows, hence the wealth gains to the stockholders. An additional risk measure is developed using first the standard deviation of monthly exchange rate changes and second of weekly exchange rate changes. These are then regressed for the longer 21-day period and the short 3-day period. The standard deviation of weekly changes in the exchange rate during the 3-day CAR period is highly significant and positive. 4.2. Economic Development The differences in the level of economic development of the countries where targets are located may present some interesting results for wealth gains from cross-border mergers. To examine this, we divide the target countries into developed and developing using the United Nations classifications for the level of economic development. Using (0, 1) dummy, developed countries are coded as 0, and developing countries as 1(DEV). When a target is in a developing country, the bidder has an opportunity to generate greater revenue using its managerial expertise to exploit market imperfections. Therefore, mergers involving developing countries are expected to have positive impact on bidder s wealth. Bidder returns could also depend on other factors such as agency problem, asymmetric information, corporate governance, and if the bidder is required to pay a higher premium to get the ownership control (Moeller et al. (2005). The return could be lower or higher depending on the above factors. Kiymaz (2004) finds significant positive wealth gains to the bidders when the target firms are in the developing countries. Sonenshine et al. (2014) find a significant negative coefficient for targets in developing

countries, but it turns positive and significant 14 when an interactive variable that combines the percentage ownership with developing country dummy is used. Doukas and Travlos (1988) also report a positive and significant relationship between wealth gains and acquisitions in the developing countries. 4.3 International Experience The acquirers with existing foreign operations are in a better position to capitalize on the market opportunities presented in foreign countries and benefit from multinational network and international diversification. We measure the degree of international experience of the acquirer (DIO) as a percent of total sales generated from overseas operations. Doukas and Travlos (1988) find that U.S. bidders who expand into the foreign market first time and to less developed countries earn positive abnormal returns and those with prior experience of operating in foreign markets earn zero or insignificant abnormal returns. Cakici et. al. (1993) find a negative or insignificant effect on bidder abnormal returns from the overseas exposure. Sonenshine et. al. (2014) tests the effect of sales ratio (target/acquirer s sale) on international merger outcome and find an insignificant relationship between the two. 4.4 Leverage Leverage measures any change in leverage after the merger. An increase in the leverage of the firm rearranges the capital structure claiming the tax shelter benefits for the stockholders. In this study, the leverage (LR) is constructed by subtracting the pre-merger leverage ratio (LTD/TA) from the post-merger leverage ratio (LTD/TA), where LTD is the book value of long term debt

and TA is the value of total assets of the firm 15 as reported on the same balance sheet. 6 The higher the resulting leverage ratio, the greater should be the gain to stocks. As such, the sign for LR should be positive. Kang (1993) report a positive and statistically significant relationship between bidder leverage and merger gains. This finding is consistent with Jensen s (1986) free cash flow hypothesis. 4.4.Means of Acquisition Financing Many studies including some referenced in this study investigate the impact of the means of payment (MAF) on shareholder wealth. It is a (0, 1) dummy variable where it is designated as 0 if the method of payment is an exchange of stock and 1 if it is a cash exchange. The MAF is expected to be positively related to stock values. In a world of asymmetric information, Myers and Majluf (1984) argue that acquirers prefer cash payment if it believes its stock is undervalued. Thus, a cash payment is good news concerning the bidding firm s true value, causing an MAF information-signaling effect. Studies investigating the effects of means of financing find that cash offers generally have positive wealth effects for both acquirers and targets. Travlos (1987) finds that cash-financed domestic mergers earn insignificant abnormal returns to the bidder. Harris and Ravenscraf (1991) find that U.S. targets in Japanese takeovers gain when cash is used in the takeovers. Pettway et. al. (1993) report positive effect on wealth gains to both buyers and sellers when cash is used, but in either case it is not statistically significant. Kiymaz (2004) finds statistically significant negative effect on buyers and positive effect on the U.S. seller when cash is used in international mergers involving financial institutions. 6 See Choi andphilippatos (1983);Shrieves and Pashley (1984); and Lubatkin and O Neill (1987)

A new MAF measure is introduced (not yet 16 incorporated at the time of this writing) to the regression. A research paper in 2015 found Cash to be insignificant, but Shared to be significant. We have defined a dummy for the Shared measure. Cash is Cash, Cash & Debt, Debt = 506 mergers; => MAFDUM1 Equity is Stock = 20 mergers; => MAFDUM3 Shared is Cash & St(ock), Cash or Stock, Cash, Stock & Debt = 72 mergers. Eckbo (1990) found CAR higher for Shared than either Cash or Equity as the MAF. Our current regression results (11/14/2017 & 3/2/2017) reflect only MAFDUM5 (Cash only). We will re-run the regression with the new measure of MAF (as well as the new measure for exchange rate risk) with results before the scheduled conference. 4.5 Size A size variable is used as a control for the size of the bidder and the target either as a relative or stand-alone measure. In this study, we represent the size variable as the ratio of pre-merger total asset to post-merger total assets of the bidder. Past studies have mixed findings on the effect of size on bidder gains. Kimyaz (2004) reports that the gains to the acquirer increases with the increase in its total assets size relative to that of the target in pre-merger year. Cakiciet. al. (1996) find a statistically insignificant negative relationship between bidder abnormal returns and the size of the equity value of target relative to that of the bidder. Kang (1993) reports a positive but not significant effect of bidder size (measured by the log of market value of assets) on bidder s abnormal returns. 4.6 Regression model To gain insights into the effects the above factors on merger benefits, the following cross-section regression model is estimated:

17 CARi = b0 + b1devi + b2dioi + b3ex1i + b4ex2i + b5ex3 + b6sizei + b7lri + b8mafi+ε for all firms i = 1 through n. CARi measures the cumulative abnormal return to the common stock of firm i. We perform several univariate and multivariate regressions. To avoid possible heteroscedasticity, we use heteroscedasticity-consistent standard error (robust) estimates. The regression results are reported in Table 4. (Insert Table 4 here) Table 4 Panel A. Full Period 3-Day CAR 21-Day CAR mafdum5-0.0065-0.0035-0.004-0.0077-0.0037-0.0049 [0.403] [0.604] [0.555] [0.615] [0.788] [0.728] ex1-0.0007** -0.0006** 0.0006 0.0007 [0.016] [0.039] [0.287] [0.178] ex1new 0.004 0.0051 0.0257 0.0303 [0.774] [0.703] [0.412] [0.330] ex2r -0.0001*** -0.0001*** -0.0001*** -0.0002*** -0.0002*** -0.0002*** [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] dio 0.0002** 0.0002** 0.0002 0.0002 [0.029] [0.030] [0.306] [0.336] dev1-0.0006-0.0007 0.0105 0.0115 [0.902] [0.896] [0.288] [0.251] lrdif -0.0102-0.0122-0.0035-0.0025 [0.680] [0.624] [0.931] [0.951] sizeratio -0.0217** -0.0233** -0.0413** -0.0395** [0.022] [0.015] [0.021] [0.026] Constant 0.0076 0.0004 0.002 0.0021 0.0108 0.0138 0.0223 0.0175*** 0.0150*** 0.0158*** 0.0443** 0.0411** [0.317] [0.862] [0.381] [0.325] [0.265] [0.165] [0.132] [0.000] [0.001] [0.000] [0.026] [0.039] Observations 595 595 595 595 595 595 595 595 595 595 595 595 R-squared 0.002 0.009 0 0.001 0.041 0.034 0.001 0.002 0.001 0.001 0.022 0.021

Panel B. Pre-Crisis 18 3-Day CAR 21-Day CAR mafdum5-0.0078-0.0065-0.0062-0.0025 0.0013 0.0011 [0.391] [0.382] [0.422] [0.893] [0.939] [0.946] ex1-0.0009** -0.0006-0.0005 0.0001 [0.019] [0.169] [0.573] [0.915] ex1new 0.0105-0.0003 0.0167 0.0014 [0.483] [0.987] [0.601] [0.968] ex2r 0.0249** 0.0126 0.0231* 0.0228 0.0235 0.0213 [0.028] [0.338] [0.059] [0.397] [0.450] [0.467] dio 0.0004*** 0.0004*** 0.0002 0.0002 [0.010] [0.009] [0.486] [0.485] dev1 0.0027 0.0021 0.0088 0.0088 [0.654] [0.726] [0.456] [0.450] lrdif -0.0339-0.0353-0.0279-0.0278 [0.278] [0.258] [0.565] [0.571] sizeratio -0.0147-0.0152-0.0520** -0.0519** [0.171] [0.156] [0.016] [0.016] Constant 0.0115 0.0017 0.0051* 0.0037 0.0027 0.0048 0.015 0.0112* 0.0129** 0.0116** 0.0429* 0.0427* [0.193] [0.577] [0.072] [0.174] [0.821] [0.698] [0.406] [0.069] [0.017] [0.036] [0.085] [0.085] Observations 387 387 387 387 387 387 387 387 387 387 387 387 R-squared 0.004 0.011 0.001 0.01 0.065 0.061 0 0.001 0.001 0.002 0.026 0.026 Panel C. Post-Crisis 3-Day CAR 21-Day CAR mafdum5 0.0003 0.008 0.0077-0.0251-0.0239-0.021 [0.982] [0.575] [0.585] [0.361] [0.381] [0.446] ex1-0.0003-0.0003 0.0013* 0.0016** [0.476] [0.503] [0.078] [0.037] ex1new 0.0323 0.0316 0.0272 0.0331 [0.390] [0.393] [0.709] [0.654] ex2r -0.0001*** -0.0001*** -0.0001*** -0.0002*** -0.0002*** -0.0002*** [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] dio 0 0 0.0002 0.0002 [0.921] [0.993] [0.479] [0.560] dev1-0.0021 0.0002 0.0224 0.0188 [0.797] [0.984] [0.175] [0.254] lrdif 0.0495 0.0483 0.0733 0.0721 [0.215] [0.224] [0.334] [0.348] sizeratio -0.0325* -0.0329* -0.0175-0.0165 [0.068] [0.064] [0.599] [0.619] Constant -0.0037-0.0035-0.0063-0.0037 0.0163 0.0143 0.0434 0.0216*** 0.0190** 0.0211*** 0.0428 0.0382 [0.767] [0.337] [0.231] [0.317] [0.308] [0.376] [0.105] [0.001] [0.047] [0.002] [0.179] [0.242] Observations 208 208 208 208 208 208 208 208 208 208 208 208 R-squared 0 0.002 0.003 0.003 0.041 0.042 0.008 0.011 0.001 0.003 0.046 0.03

5. Regression Results 19 The regression results for the full sample are presented in Panel A of Table 4. It contains the results of six separate regressions for both (-1, +1) and (-10, +10) windows. Probability values of the coefficients are provided in the parenthesis. The first regression equation uses the means of financing variable (MAF) to explain its wealth effects. Next three equations use exchange rate variable, each one representing different measure of the exchange rate variable as described in the previous section to determine the wealth effects of exchange rate movement. The fifth equation includes all variables except EX2 and the sixth equation includes all variables except EX1. The coefficient of the MAF is positive but not significant in both univariate and multivariate regressions for (-1, +1) window. It is negative and not significant for (-10, +10) window. The result is not surprising given that past studies have reported both positive and negative wealth effects. Pettway et. al (1993) report positive effect on wealth gains to both buyers and sellers when cash is used, but in either case it is not statistically significant. Kiymaz (2004) reports a negative and significant coefficient for cash payment for U.S, bidders. The coefficient of the exchange rate variable EX1 is positive, but not significant, in both univariate and multivariate regressions for (-10, +10) window, indicating that strong dollar has a positive effect on the wealth grains to the U.S. bidders. It is, however, negative and significant in in both univariate and multivariate regressions for the (-1, +1) window. Sonnenshine (2014) who uses a similar measure reports that acquirers pay more for foreign targets when their

20 currency is strong. In his regression model, he finds the coefficient to be positive but not significant indicating essentially no effect on the wealth gains of acquirers. We find the coefficient of the exchange rate variable EX2positive but not significant in both univariate and multivariate regressions for both windows. The construction of this measure is the same as the method employed by Harris and Ravenscraft (1991), Cakici et al. (1996), Kang (1993), Pettaway et al.(1993), and Kiymaz (2004). Our findings arein line with the findings of Pettaway et al. (1993), Cakici et al (1996), and Kiymaz (2004). All the three report no significant relation between bidder s wealth gains and exchange rate change. Kang (1993), however, finds a positive and significant impact on wealth gains to Japanese bidders when yen increased in value against the U.S. dollar. Harris and Ravenscraft (1991) find a positive and strong effect on target s wealth gains when the acquirer currency is strong. The coefficient of EX3 variable, which represents the exchange rate volatility, is negative and significant at 1 percent level in both univariate and multivariate regressions and for windows. This confirms the possibility of higher currency volatility having a negative impact on bidder s wealth gains. Kiymaz who employs EX1 and EX3 measures in the same regression in a multivariate setting also finds similar results as ours. We find support for a positive and significant effect of prior overseas experience on the merger benefits.harris and Ravenscraft (1991), Cakici et al. (1996) and Sonenshine et al. (2014) find no relation between wealth gains and international experience.

Our findings for economic development 21 variable (DEV) and leverage LEV) are negative and insignificant and not as expected for the (-1, +1) window. The results are not different for (-10, +10) window. Other studies had different results. Kiymaz (2004) anddoukas and Travlos (1988) find positive and significant impact on wealth gains when targets are in developing countries. Kang(1993) reports that bidder gains increasewith leverage ratio. The coefficient of SIZE variable is negative and significant over both event windows. This indicates that the smaller the size of the acquirer relative to other US firms involved in international mergers, the greater is the benefit to the acquiring firm. This is an interesting finding, leading to the observation that either the acquirers find the US domestic market limited for growth potential due perhaps to a dominant player or that the overseas opportunities offer greater reward for the investment dollar. Past findings vary. Cakici et al. (1996) report negative and insignificant impact onwealth gains when the size is measured as a relative size of target to bidder..kiymaz (2004) and Kang (1993) report positive and significant effects. The difference in the findings could, perhaps, be explained by the way the variable is constructed. 6. Summary and Conclusions This paper examines the role of exchange rate movement in cross-border investment decisions and its impact on the return outcome along with other determinants of return outcomes. After reviewing the findings of recent empirical studies on cross-border mergers and acquisitions that have included the exchange rate as an explanatory variable, we investigate the impact of the exchange rate movement on shareholder wealth gains from cross-border mergers and acquisitions by the U.S. firms during 2001-2011. We find that the U.S. acquirers experienced

22 positive wealth gains during the sample period studied. The wealth gains are statistically significant during 2001-2007, and these gains are higher when the target companies are in developing countries compared to the targets in developed countries. While the wealth gains to acquirers are generally positive, the evidence supporting the effects of individual determinants are mixed. The determinants considered include the exchange rate, the means of acquisition financing, overseas experience, the level of economic development of target countries, leverage and size. We find a change in the real exchange rate (EX1)is negatively and significantly related to the wealth gains of acquirers over the narrow window (-1, +1) and, its effect is positive, but not significant, over the wider window (-10, +10). The effect of EX2 variableis positive and not significant over both narrow and wider windows. We, however, find a significant negative relationship between the currency volatility and acquirers wealth gains. The regression results show that there is nosignificant impact of cash as the medium of exchange, acquisitions in developing countries orthe leverage on acquirer wealth gains.the size of the acquirer has a significant negative effect. The evidence on acquirer s overseas experience is positive and strong, even though only over the narrow window (-1, +1). The findings of this study regarding the role of the exchange rate movementsare significant and worth emphasizing. While the acquirers can bid aggressively for foreign firms when their currency is strong due to the wealth gainsand cost of capital advantage resulting from information asymmetries in the global capital markets, the evidence on its ability to deliver strong returns by itself is weak at the minimum. This supports the observation of Blonigen

23 (1997) that there is no guarantee that the returns from cross-border investment will be positive as the result of the currency being strong at the time of announcement unless the acquisition involves a firm-specific asset which is transferable. The currency volatility, rather than a change in the level of the exchange rate, appears to have an expectednegative and significant effect on the wealth gains to stockholders. The following is a portion of the Summary & Conclusions from the older versions of the paper. I left it if we can use any part of it in the current version. The older version did not a separate section for the discussion of regression results. [Results indicate that experience in the targeted nation, the change in comparative risk on the acquirer, both of the exchange rate explanatory variables, the size of the acquirer, and the level of foreign experience of the acquirer are all important factors in determining the merger benefit for a US firm targeting a foreign firm. The previous international experience of the acquirer is suggested as having a positive effect on the market s expectation of the acquirer s ability to claim benefits from an international merger. Within this study, we have measured an acquirer s previous experience with both a measure of the firm s previous broad international experience (EXPER) and with a measure of the firm s previous investment (therefore, experience) in the targeted nation for the current merger (INVEST). Both measures, INVEST and EXPER, indicate that the market responds with a positive expectation due to the acquirer s previous international experience, and most strongly to the acquirer s previous investment in the merger partner s nation. The exchange rate variables EX1 (the measure of the change in value of the US dollar versus the currency of the targeted merger) and EX 2 (the measure of exchange rate variability of

the targeted nation s currency) are both 24 significant and have the expected sign. As such, these results confirm the expectation that an appreciating dollar expands the opportunities for US firms to invest in foreign assets. This is consistent with the findings of Harris and Ravenscraft (1991) and Tandon (1996). But previous studies have not also studied the impact of the variability in the value of the targeted currency. The results reported herein also confirm the expectation that stock investors are concerned regarding acquisitions in countries with greater exchange rate uncertainty. Since SIZE, also significant but with the opposite sign, indicates that the greater benefit belongs to the acquirer with the smaller relative size versus the size of other US firms involved with international mergers, this study reports that the foreign merger market offers greater benefits to the smaller US firm than does the current domestic market for expansion. The remaining variables are found not to be significant. Since DEV (the measure indicating whether the targeted nation is classified as developed or developing by the United Nations) is not significant, there is no indication that there is a preference for expansion for US firms into either developed or developing nations. As such, there is no indication whether the positive opportunities (associated with bringing efficiencies from internal capital markets or associated with benefits from bringing developed nations efficiencies into a developing market) or the negative effects (associated with an increase in risk associated with the greater uncertainties associated with investment in developing nations or the increase in costs associated with greater monitoring expenses and limitations placed on operations, such as limits on ownership or on capital repatriation) are dominant on the benefits of international expansion. LR (the financial leverage ratio which measures the change in the acquirer s debt level) does not support the expectation that a change in leverage will enable the acquirer to claim

additional benefits from the merger. 25 Therefore, there is no support for the expectation that the development level of the target nation, LR, or MAT are useful tools for creating additional value by US firms in international mergers. The empirical findings of this study support the presence of exchange rate influences on the merger premium. It is reported here that changes in the real exchange rate do have an impact on foreign acquisitions through the wealth effect these changes generate; and, that greater exchange rate volatility is negatively associated with benefits to the acquirer s stock values. In addition, the acquirer s stock values are positively affected by the acquirer having greater international experience, experience in the target nation, an increase in market risk derived from the merger, and greater benefits to the smaller acquirer. ]

26 References Asquith, P. (1983) "Merger Bids, Uncertainty, and Stockholder Returns," Journal of Financial Economics, April. Asquith, P. and E. Han Kim (1982) "The Impact of Merger Bids on the Participating Firms' Security Holders," Journal of Finance, December. Asquith, P. and David W. Mullins (1986) "Equity Issues and Offering Dilution," Journal of Financial Economics, January/February. Blonigen, A. (1997) Firm specific assets and the link between exchange rates and foreign direct investment, American Economic Review 87, 447 65 Bradley, M., A. Desai and E.H. Kim (1983) "The Rationale Behind Interfirm Tender Offers: Information or Synergy," Journal of Financial Economics. Brick, I.E., L.J. Haber and D.G. Weaver (1982) "Financial Motives in Conglomerate Mergers: An Empirical Test," Mergers and Acquisitions: Current Problems in Prospective, edited by M. Keenan and L.J. White, Heath Lexington Press. Cakici, Nusret, Chris Hessel and Kishore Tandon (1996) Foreign acquisitions in the United States: Effect on shareholder wealth of foreign acquiring firms Journal of Banking & Finance, 20, 307-329 Caves, Richard E.(1971) International corporations: The industrial economics of foreign investment,economica 38, l-27. Cebenoyan, A.S., G.J. Papaioannou and N.G. Travlos (1992) Foreign Takeover Activity in the U.S. and Wealth Effects on Target Firm Shareholders, Financial Management, Autumn. Choi, D. And G.C. Philippatos (1983) "An Examination of Merger Synergism," Journal of Financial Research, Fall. Crawford, Dean (1987) "The Structure of Corporate Mergers: Accounting, Tax and Form-of- Payment Choices," Thesis, University of Rochester, Rochester, New York. Dann, L. and W. H. Mikkelson (1984) "Convertible Debt Issuance, Capital Structure Change and Financing-related Information," Journal of Financial Economics, vol. 13. Dewenter, Kathryn L. (1995) "Does the Market React Differently to Domestic and Foreign Takover Announcements? Evidence from the U.S. Chemical and Retail Industries," Journal of Financial Economics, 37. Dewenter, Kathryn L., 1995, Do exchange rate changes drive foreign direct investment? Journal of Business 68, 405 Doukas, John and Nicholaos G. Travlos (1988) The Effect of Corporate Multinationalism on Shareholders Wealth: Evidence from International Acquisitions, The Journal of Finance, Vol. 43, No. 5, pp. 1161-1175. Eckbo, E. B. (1986) "Valuation Effects of Corporate Debt Offerings," Journal of Financial Economics, vol. 15. Eger, C.E. (1983) "An Empirical Test of the Redistribution Effect in Pure Exchange Mergers," Journal of Financial and Quantitative Analysis, December. Errel, Isil, Rose C. Liao and Michael S. Weisbach (2012) Determinants of Cross-Border Mergers and Acquisitions, The Journal of Finance, Vol. 67, No.3 (June), pp. 1045-1082. Froot, Kenneth A. and Jeremy C. Stein (1991) Exchange Rates and Foreign Direct Investment: An Imperfect Capital Markets Approach The Quarterly Journal of Economics, vol. 106, No. 4, pp. 1191-1217. Fishman, M.J. (1989) "Pre-emptive Bidding and the Role of the Medium of Exchange in

27 Acquisitions," Journal of Finance. Georgopoulos, George T (2008), Cross-Border Mergers and Acquisitions: Does the Exchange Rate Matter? Some Evidence for Canada The Canadian Journal of Economics Vol. 41, No. 2, pp. 450-474. Halpern, P.J. (1973) "Empirical Estimates of the Amount and Distribution of Gains to Companies in Merger," Journal of Business, October. Hansen, R.G. (1987) "A Theory for the Choice of the Exchange Method in Mergers and Acquisitions," Journal of Business. Harris, R.S. and D. Ravenscraft (1991) The Role of Acquisitions in Foreign Direct Investment: Evidence From the U.S. Stock Market, Journal of Finance, vol. 46, No. 3.. Hayn, C. (1989) "Tax Attributes as Determinants of Shareholder Gains in Corporate Acquisitions," Journal of Financial Economics. Healy, P., K. Palepu and R. Ruback (1992) "Does Corporate Performance Improve after Merger," Journal of Financial Economics, April. Hymer, Stephen H. (1976), The international operations of national firms: A study of direct foreigninvestment (MIT Press, Cambridge, MA Jarrell, G.A., J.A. Brickley and J.M. Netter (1988) The Market for Corporate Control: The Empirical Evidence Since 1980, Journal of Economic Perspectives, 2. Jensen, M. and Meckling (1976) "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure," Journal of Financial Economics, October. Jensen, M.C. and R.S. Ruback (1983) The Market for Corporate Control: The Scientific Evidence, Journal of Financial Economics, 11. Kang, J.K. (1993) The International Market for Corporate Control: Mergers and Acquisitions of U.S. Firms by Japanese Firms, Journal of Financial Economics, 34. Kim, McConnell and Greenwood (1977) "Capital Structure Rearrangements and Me First Rules in an Efficient Capital Market," Journal of Finance, June. Kindleberger, Charles (1969) American Business Abroad: Six Lectures on Direct Investment Yale University Press. Kiymaz, Halil (2004) Cross-border acquisitions of US financial institutions: Impact of macroeconomic factors Journal of Banking & Finance, 28, pp. 1413-1439 Klein, Michael W., and Eric S. Rosengren, 1994, The real exchange rate and foreign direct investment in the United States: Relative wealth vs. relative wage effects, Journal of International Economics 36, 373-390. Lauterbach, Malitz and Long (1991) "Model Specification, Information Asymmetry and Antitakeover Defenses," Working Paper. Levy, H. and M. Sarnat (1973) "Diversification, Portfolio Analysis and the Uneasy Case for Conglomerate Merger," Journal of Finance, September. Lewellen, W.G. (1971) "A Pure Financial Rationale for Conglomerate Merger," Journal of Finance, May. Lubatkin, M. and H. O'Neill (1987) "Merger Strategies and Capital Market Risk," Academy of Management Journal. Malatesta, P.H. (1983) "The Wealth Effect of Merger Activity and the Objective Function of the Merging Firms," Journal of Financial Economics. Malitz, Ileen B. (1989) "A Re-Examination of the Wealth Expropriation Hypothesis: The Case of Captive Finance Subsidiaries," Journal of Finance, September. Manley, J.F. (1995) Cross Sectional Determinants of the Merger Premium, Thesis, Rutgers,