CROSS-BORDER MERGERS AND ACQUISITIONS BETWEEN INDUSTRIALIZED AND DEVELOPING COUNTRIES: US AND INDIAN MERGER ACTIVITY

In this paper we study the cross border mergers and acquisition between the US. and Indian fi rms. Our empirical work suggests that US fi rms realize signifi cant losses on the announcement of acquisitions of Indian targets while Indian targets realize signifi cant gains on the announcement of mergers with US acquirers. Publicly-traded Indian fi rms realize insignifi cant returns on their announcement of acquisitions of publicly-traded US fi rms but realize signifi cant positive returns on announcements of acquisitions of privately-held US fi rms and subsidiary fi rm targets. Publicly-traded US targets realize insignifi cant gains when US acquired by Indian fi rms.


Introduction
Much of the current research on cross-border mergers focuses on analyzing the returns of industrialized fi rm's acquisitions of emerging world targets. 1 Little attention has been given to acquisitions of developed country targets by emerging world country fi rms. Also, researchers have focused primarily on the returns to the acquirers and none of the studies to date have investigated the effect of announcement of mergers and acquisitions on the target fi rms in the cross border IJBF mergers and acquisitions involving developed and developing world fi rms. We study both types of cross-border merger activities by examining the Indian and US company returns when US fi rms are acquirers of Indian fi rms and when Indian fi rms are acquirers of US fi rms. 2 We fi nd that there is no country effect when a publicly-traded fi rm acquires a publicly-traded target or subsidiary fi rm target but there is a signifi cant country effect for public-traded acquirers of privately-held targets US (acquirers earn signifi cantly lower returns than Indian acquirers). Our investigation for the target fi rms reveals that acquisition by any US fi rm (public, private or subsidiary fi rm) is benefi cial to the shareholders of the Indian targets whereas the US public targets earn insignifi cant returns on the announcement of their acquisition by Indian fi rms. We also fi nd that the abnormal returns for the targets of publicly-traded acquirers are consistently higher than the abnormal returns of targets of privately-held acquirers. Figure 1 shows the dramatic increase over the last few years in mergers and acquisitions between the US and Indian companies. Prior to 1995 there were few cross-border mergers and acquisitions between the US and Indian fi rms. Post 1995, US acquisitions of Indian targets reached more than 100 fi rms in 2000, then declined rapidly following the dot.com bubble burst. By 2006, US acquisitions of Indian targets were again close to 100 fi rms. Indian acquisitions of US fi rms followed a similar pattern with increases up to the year 2000 (but only about 50 transactions), then a decline following the dot.com bubble burst. By 2006, the number of transactions had risen above the year 2000 level. In Figure 2, we separate the cross-border mergers and acquisitions by types of ownership structure -publicly-traded, privately-held or non-traded subsidiary. The fi gure shows that Indian acquirers of US fi rms are largely publicly-traded fi rms whereas the US targets are typically privately-held fi rms. Mergers and acquisitions between US acquirers and Indian targets are spread across the publicly-traded, privately-held and subsidiary fi rms for both acquirers and targets. Conn, Cosh, Guest and Hughes (2005) assert that there are important theoretical reasons why the acquisition of domestic and cross-border targets may differ and why acquisition of private targets may differ from the acquisition of public targets. 3 2 The economic environment and the business environment of countries play a crucial role in the decision to acquire or being acquired in cross-border merger and acquisitions (La Porta, Lopez, Shleifer & Vishny, 1998, 1999. We cannot get meaningful information by studying all the developed world and the developing world countries together. Doing the merger and acquisition study for all the developing world countries separately in one paper would make the paper too voluminous. Hence we limit the scope of this paper to mergers and acquisitions in India and between India and the US and leave the mergers and acquisition in the remaining developing countries as a topic for future research. 3 We refrain from the discussion as to why cross border mergers and acquisitions and why acquisition of private targets is different from the acquisition of public fi rms. Those issues are discussed at length in section 2 of Conn et al. (2005).  Figure 1 shows an increasing trend in the number of mergers and acquisitions between U.S. and Indian fi rms. Accordingly, we examine mergers and acquisitions between the US and Indian fi rms by separating them into different pairings: publicly-traded acquirers and publicly-traded targets, publicly-traded acquirers and privately-held targets, publicly-traded acquirers and subsidiary fi rms, privately-held acquirers and publicly-traded targets and subsidiary fi rm acquirers and publicly-traded target fi rms.
In the following fi gure we categorize the mergers and acquisitions by publicly-traded fi rms, by privately-held fi rms, by subsidiaries of large fi rms and by others (government fi rms, joint ventures etc.). Indian acquirers of U.S. fi rms are largely public and private fi rms but the U.S. targets are largely privately held and subsidiaries of large fi rms. Mergers between the U.S. acquirers and Indian targets are spread across publicly-traded, privately-held and subsidiary fi rms for both acquirers and targets.
There is substantial literature on the announcement effects of crossborder acquisitions on acquiring fi rms in developed countries such as the US and the UK but there is limited work that has been done examining mergers and acquisition between the acquirers and the targets of a developing country (such as India) and an industrialized country (such as the US). In cross-border mergers and acquisition studies involving acquirers, Moeller and Schlingemann (2005) have one target Indian fi rm; Francis, Hasan and Sun (2008) have 13 target Indian fi rms; and Chari, Ouimet and Tesar (2010) have 33 target Indian fi rms in their samples. Using a comprehensive sample of 248 fi rms, in this paper we study the announcement effect of mergers and acquisitions on the acquirers of Indian targets. Also, none of the studies done to date have studied the announcement effect of mergers and acquisitions on the target shareholders of the fi rms from the developing world. Conn et al. (2005) stated that despite the increase in the acquisition of cross-border public and non-public targets, nearly all acquisition studies were limited to domestic targets which were publicly traded. They studied the acquisitions of both domestic and cross-border targets that were public and private but they limited their study to the effect of acquisition on UK acquirers only. Utilizing merger and acquisition data from 1995 to 2007 we address the announcement effect on publicly-traded acquirers of publicly-traded, privatelyheld and subsidiary targets, and on publicly-traded targets by privately-held fi rms for US acquirers and Indian targets and Indian acquirers of US targets. We fi nd that Indian target fi rms when acquired by a publicly-traded US fi rm, realize positive abnormal returns around the announcement date whereas acquiring fi rms suffer a loss of market value around the announcement date. Indian targets acquired by privately-held US fi rms enjoy signifi cant gains on the announcement of a merger or acquisition. In the case of an Indian acquisition of a US target, the acquiring Indian fi rm experiences positive abnormal returns around the announcement date. The returns to an Indian acquirer of a publiclytraded US fi rm are statistically insignifi cant whereas Indian acquirers realize signifi cant abnormal returns around the announcement date of the acquisition of privately-held US fi rms and the subsidiaries of US fi rms. US targets realize statistically insignifi cant returns on the announcement of a merger or acquisition by Indian fi rms. International Journal of Banking and Finance, Vol. 8, Iss. 1 [2011], Art. 3 Our results indicate that the shareholders of Indian acquirers of privatelyheld US targets and US subsidiary fi rm targets gain from the announcement of an acquisition whereas the share prices of US acquirers of privately-held Indian targets fall on the announcement of an acquisition. 4 We investigated if this difference in cross-border mergers and acquisitions was due to fi rm characteristics or was due to the country environment. We examined the fi rm characteristics of the acquirers and found that Indian and the US acquirers of privately-held targets have similar high tech status and similar diversifi cation levels. Indian acquirers of US subsidiary targets have substantially greater high tech status and higher diversifi cation levels than the US acquirers of Indian subsidiary targets. For all target categories, the relative size of targets is substantially smaller for US acquisitions. The market value of Indian acquirers is also signifi cantly lower than the market value of US acquirers. Both Indian and US fi rms, acquire majority stakes in target fi rms but the average stake acquired is higher in private and subsidiary fi rm targets than in publicly-traded targets.
To determine if the reaction to the news of an acquisition is due to fi rm specifi c characteristics, we regressed the three-day cumulative abnormal return around announcement (-1 day to +1 day after announcement) on fi rm specifi c characteristics and a dummy variable identifying the country of the acquirer. Our results indicate that there is no infl uence of the country environment when a publicly-traded fi rm acquires a publicly-traded target or subsidiary fi rm target as the dummy variable for the country remains insignifi cant. The dummy variable is signifi cant for public-traded acquirers of privately-held targets indicating that the US acquirers earn signifi cantly lower returns than Indian acquirers.
Our results differ from the recent research of Chari, Ouimet and Tesar (2010) and Gubbi, Aulakh, Ray, Sarkar and Chittor (2010). Chari et al. (2010) studied the returns to the developed market acquirers of emerging market fi rms. They reported a positive and signifi cant abnormal return for the developed country acquirer of targets from emerging markets whereas we found a negative but insignifi cant abnormal return for the US acquirers of Indian targets. They found that private targets were associated with signifi cantly higher announcement returns for acquirers from developed world countries whereas we found that US acquirers of private Indian targets earned signifi cantly negative abnormal returns. Gubbi et al. (2010) found that the Indian acquirers of foreign targets (both developing and developed country fi rms) gained on the announcement of their mergers and acquisitions. However, when they regressed the abnormal returns on explanatory variables, the coeffi cient on private targets was positive and insignifi cant in one model (as in their Table 4) and negative and insignifi cant in the other (in Table 5). We found that Indian acquirers of targets earned signifi cant, positive abnormal returns on the announcement of US acquisitions of targets but this gain was limited to the acquisition of private targets only.
The size of our sample allows us to avoid the grouping of targets and acquirers from different countries. Whereas, the developed world countries are somewhat economically and culturally similar to each other there are large differences across the countries of emerging markets. Studying cross-border mergers and acquisitions between representative developing and industrialized countries should provide less noisy information.
The rest of the paper is organized as follows: Section 2 reviews related literature and section 3 discusses the data sources and methodology. Section 4 presents the empirical results. Section 5 provides some of the probable cases for the gains/losses of the shareholders of the acquirers and the targets, and section 6 concludes.

Literature Review
We summarize the literature on cross-border mergers and acquisitions in Table  1; US targets gain signifi cantly when acquired by foreign fi rms, foreign acquirers of US fi rms gain signifi cantly whereas the US acquirers of foreign fi rms show mixed results. The fi rms in these studies are predominantly from industrialized countries. Markides and Ittner (1994) examined 276 U.S. international acquisitions made from 1975 to 1988 and found the two-day cumulative abnormal return [CAR (-1, 0)] for acquiring fi rms to be 0.32% (statistically signifi cant). Most of the acquisitions in Markides and Ittner's study came from the manufacturing and fi nance, and insurance industries and are from predominantly developed countries. Markides and Oyon (1998) used a sample of 236 US acquisitions consisting of 47 Canadian targets and 189 European targets (32 from France, 81 from UK, 27 from Germany, 13 from Italy, 14 from Spain, 15 from Holland and 7 from Belgium) and found that US acquisitions in Europe generated signifi cant returns while returns for US acquisitions in UK and Canada were not signifi cant. Seth, Song and Pettit (2002) investigated 100 cross-border acquisitions between foreign acquirers and US targets during the time period [1981][1982][1983][1984][1985][1986][1987][1988][1989][1990]. They found that the cumulative abnormal return [CAR (-10, 10)]: 10 days before and 10 days after the fi rst bid by the ultimately successful bidders) to the foreign bidder was 0.11 per cent. 5 The acquirers of the 100 cross-border acquisitions in this study were mainly from industrialized countries -Great Britain (52), Japan (10), Canada (10), Australia (8), West Germany (3), and Switzerland (2). Conn et al. (2005) studied 4,000 UK domestic and cross-border public and private acquisitions. For the acquirers of publicly traded targets they found signifi cant losses for domestic mergers and acquisitions but insignifi cant losses for cross-border mergers and acquisitions. They found signifi cant gains for acquirers of privately-held fi rms for both domestic and cross-border targets.  Cakici, Hessel and Tandon (1996) 195 foreign firms that acquired US firms and 112 US acquisitions of foreign firms Moeller and Schlingemann (2005) studied the announcement effect on US acquirers for a sample of 383 cross-border transactions and 4047 domestic takeover transactions during 1985 to 1995. They found a three-day (-1, +1) market adjusted return of 0.307 per cent for cross-border acquirers and 1.173 per cent for domestic acquirers. In the Moeller and Schlingemann (2005) study, the UK is the most frequent target country (31%), followed by Canada (21%), France (9%) and Germany (9%). In their sample there is only one target fi rm from India.
In a recent study, Francis, Hasan and Sun (2008) used a sample of 1,491 foreign acquisitions by US fi rms and 7,692 US domestic acquisitions. They found that over the full sample period of 1990-2003, acquirers in domestic M&As experienced an average abnormal stock return of 1.49 per cent whereas the acquirers in cross-border M&As experienced an average abnormal return of 0.96%. Among their 1,491 transactions, 1,275 (85.50%) were from integrated fi nancial markets, with the remaining 215 (14.50%) from segmented fi nancial markets. There were only 13 Indian fi rms acquired by US fi rms in this sample.
Most merger and acquisition studies have focused on publicly-traded fi rms within the domestic US market or those between the US and other industrialized countries. Little research has examined the returns to shareholders from crossborder mergers and acquisitions between the US and developing country fi rms. 6 Chari, Ouimet and Tesar (2010) fi lled this gap and examined the returns to shareholders of developed country fi rms that undertook acquisitions in emerging markets. They found that when developed country acquirers gained control of emerging-market targets, they experienced positive and signifi cant abnormal returns of 1.16 per cent, on average, over a three-day event window. Though Chari et al. (2010) studied the acquisition of targets from India by developed world fi rms there were only 33 target fi rms from India. In this paper we have a sample of 248 fi rms from the acquiring targets in India. Chari et al. (2010) did not study the effect of announcement of mergers and acquisitions on target fi rms. Also, they did not study the acquirers of targets from developed world fi rms (like US) by developing world fi rms (like India).
Research by Gubbi, Aulakh, Ray, Sarkar and Chittor (2010) studied 425 foreign mergers and acquisitions by Indian fi rms from January 2000 to December 2007. They found positive abnormal returns for the acquiring fi rm shareholders when combining all target fi rm countries (developing and emerging) into one group. They investigated if the abnormal returns for the acquirers of developed world targets were different than those for emerging world target and found that acquirers had statistically higher abnormal returns when the target fi rms were located in advanced economies. They also investigated if the returns for the acquirers of private targets were different from the returns for public targets but found no signifi cant results. In their Table 4, they performed a cross-sectional regression on the abnormal return for the acquirers around announcement date and reported a negative and insignifi cant 'private' variable which was one for private targets and zero for public targets. In their Table 5, Gubbi et al. (2010) reported a positive an insignifi cant 'private' variable for a different cross sectional regression model. Hence their fi ndings on the abnormal returns to the acquirers for different public/private stat were ambiguous. Also, Gubbi et al. (2010) did not study the announcement effect of mergers and acquisitions on the target fi rms.
Cross-border mergers and acquisitions can be benefi cial for the acquirers resulting in positive abnormal returns on the announcement date; they can also result in losses for the acquirers. Researchers have given several reasons for the gains from cross-border mergers and acquisitions. Caves (1971Caves ( , 1998, Yeung (1991, 1992) and Williamson (1979) point to the internalization benefi ts in the cross-border mergers and acquisitions. Firms extract above normal returns from cross-border investments by internalizing the host country market imperfections when their fi rm specifi c assets cannot fi nd comparable values elsewhere. Ayban and Ficici (2009) note that the resulting rents derived from internalization are expected to be capitalized into a higher value of the fi rm. Baldwin and Caves (1991) point out that cross-border mergers and acquisitions may result in gains from diversifi cations when businesses seek synergies arising from intangible and information-based assets like brand names, technical knowledge and R&D expertise. According to Kogut (1983) cross-border acquisitions may increase the operational fl exibility of the fi rm by giving it the opportunity to exploit market conditions. Conn et al. (2005) state that in crossborder merger, geographical diversifi cation by direct investments in overseas subsidiary permits fi rms to expand the boundary of the fi rm. This will result in an increase in revenues for the fi rm.
Researchers give several reasons for the loss from cross-border mergers and acquisitions. Conn et al. (2005) argue that overseas targets are more diffi cult to value accurately because of imperfect information. They point out that there are diffi culties of managing the post-merger process when cultural differences make integration and acculturation, a diffi cult, time-consuming and expensive process. The bigger the cultural gap, the bigger the relative size of the target the worse the problem may be. Aybar and Ficici (2009) point out that differences in natural culture, customer preferences, business practices and institutional forces may jeopardize the potential gains of cross-border mergers and acquisitions. Hitt, Hoskissons and Ireland (2001), Hitt, Ireland, Camp and Sexton (2001) and Kissin and Herrera (1990) point out that complication in target assessments, misidentifi cation of asset complementarities, informational asymmetries and high premiums paid for target may have adverse effects on the value of acquiring fi rms.
The above factors infl uence the developed and developing country fi rms differently. Furthermore, when a fi rm from a developed country acquires a developing country fi rm, it gains from the cheap labour thus reducing its operational cost whereas when a fi rm from a developing country acquires a developed world fi rm, it accesses the technological know-how thus increasing the growth potential for the fi rm. In their studies, Chari et al. (2010) have grouped countries into developed markets and emerging markets and Gubbi et al. (2010) have clubbed developed and developing world target countries together. Whereas, the developed world countries are economically and culturally similar to each other and can be grouped together, there exist noticeable differences across the countries of emerging markets. While the countries in the developed world are all democracies, there are different government structures in the emerging world countries. Whereas the countries in the developed world predominantly follow the same religion and have similar cultures, there are differences in the cultural and religious practices of the emerging world population. Whereas the judicial system in the developed world countries is strong, it is not so strong in all the emerging world countries.
The fi nancial markets in all the developed world countries can be considered to be at least weak form effi cient; the fi nancial markets for many emerging world countries are not even weak form effi cient. Due to the difference within the emerging world countries, a fundamental factor may play an important role in one emerging world country but may not be important for mergers in other emerging world countries. It may also be that because of the differences across the emerging world countries effect due to fundamental factors cancel out when we club the data from different countries together. Studying the cross-border mergers and acquisitions between two countries at a time especially when the merger and acquisition is between the developed world and the emerging world countries would hence provide more meaningful information. Also combining all targets in one group may cancel the effects of their different organizational form hence one should be cautious in deriving any inference from the papers studying all targets together and more meaningful information can be derived by splitting the targets into public and private.
In this paper, we examine the gains and losses to shareholders of both targets and acquirers, from mergers and acquisitions between the US and the developing country of India by separating the targets into their organizational forms (public, private and subsidiary fi rms). The overall purpose of our analysis is to bring new evidence to view on shareholder wealth from the cross-border (especially when the acquirer is from an emerging world country) merger and acquisition activity.

Data, Variables and Methodology
This study focuses on cross-border acquisitions between the US and India announced over the period January 1995 -August 2007. 7 We extract our merger and acquisition sample from SDC's (Securities Data Corporation, a database from Thomson Financial) Mergers and Acquisitions database during the sample period of January 1995 to August 2007. We use CRSP for daily returns and daily index returns for US fi rms and DataStream for the stock prices of Indian fi rms. We use the BSE200 (Bombay Stock Exchange) and the BSE500 from DataStream for the Indian market index. The BSE 500 is a more compensative index than BSE200, but it was fi rst introduced on February 1, 1999. Hence, we use BSE200 as the Indian market index during 1995-1998 and BSE500 as the Indian market index after 1998. This table shows the number of acquiring and target fi rms in our initial sample and the fi nal sample for the mergers and acquisitions between the US and the Indian fi rms.
The initial data set consists of 676 US acquisitions of Indian fi rms and 230 Indian acquisitions of US fi rms. This sample includes all the bidding and targets fi rms irrespective of whether or not the merger was successful. Firms without complete daily return or stock price information are excluded from the sample. As reported in Table 2, this reduces our initial sample to 248 US acquiring fi rms and 98 Indian target fi rms, and 16 US target fi rms and 128 Indian acquiring fi rms. In our sample all of the subsidiary targets are non-traded fi rms.
For the cross-sectional regression we collect the data on fi rm characteristics from SDC. There are several Indian fi rms with incomplete or missing data on SDC. We use DataStream to obtain the data on these fi rms. Excluding relative size data and the data on percentage owned after transaction, we have complete data on 99 Indian acquirers of US targets and 129 US acquirers of Indian targets. When we include the percentage owned after transaction our sample has 62 Indian acquirers of targets and 69 acquirers of Indian targets. When we include We compute the announcement date abnormal return on a three-day window of -1 day to 1 day after the announcement. We use the standard eventstudy methodology to analyse the impact of the acquisitions announcement on shareholders' wealth for both the acquiring and the target fi rms. We set the event date as the announcement date of the acquisition as reported in the SDC Platinum database.
The abnormal return for stock j on day t ( ) is computed using the marketadjusted returns model: ( 1) where is defi ned as the raw return of the common stock of the jth fi rm on day t, and is the expected return on stock j. To calculate the expected return on stock j, we use the market model: ( 2) where is the daily market return using a value-weighted index, and for these fi rms are computed through a regression of the fi rm returns on the market returns during the time period of 120 days to 5 days before announcement. For US fi rms, the daily value-weighted return from CRSP is used as the proxy for the US market return; while for Indian fi rms, the BSE200 and BSE500 indices from DataStream are used as the proxy for the Indian market returns. 8 The average abnormal return for day t is calculated as: ( 3) where N is the number of fi rms in the sample. Over an interval of three days beginning with day -1 and ending with +1, the cumulative average abnormal return is:

AAR CAAR
To test the null hypothesis that cumulative average abnormal returns from -1 day to +1 day, is zero, we compute the t statistic (tstat) as: (5) where is the average abnormal return on day t, is the sample variance of the average abnormal return for day t, and n is the number of fi rms in the sample.
Our objective is to test if the cumulative average abnormal returns for acquirers or targets are signifi cantly different from zero. We perform the following hypothesis test: where is the cumulative average abnormal return. We use the t-test statistics of equation (5) to determine the statistical signifi cance of the results.

Results
In this section, we analyse the shareholder wealth effects for both the acquiring and the target fi rms. We report the cumulative abnormal returns for acquirers and targets during the three-day event window of -1 day to 1 day after the announcement.

US Acquirers and Indian Targets
In Panel A of Table 3 we present the CAARs for US acquirers of Indian publiclytraded, privately-held and non-traded subsidiary fi rms. Though the sign on the cumulative abnormal returns around the announcement date is negative, the cumulative abnormal returns for acquiring US fi rms are not signifi cant. When we split the sample 9 into the US acquirers of publicly-traded, privately-held and subsidiary fi rm Indian target companies, we fi nd the abnormal returns on the announcements of US acquisitions of publicly-traded and subsidiary fi rm targets are statistically insignifi cant. Shareholders of US fi rms who acquire privatelyheld Indian fi rms realize statistically signifi cant abnormal returns of -1.1 per cent on the three day window (-1 to +1).  Panel A presents the cumulative average abnormal returns for publiclytraded US fi rms that announced the acquisition of Indian target fi rms. The overall results in column 2 shows negative abnormal returns for the shareholders of the acquiring fi rms. When we separate them into different groups the results show losses for the acquirers of publicly-traded, privately-held and subsidiary fi rm targets. In Panel B we present the cumulative average abnormal returns around the announcement date for Indian target fi rms. Overall results in column 2 show that shareholders of target fi rms earn signifi cant returns around the announcements of mergers and acquisitions. These results hold even when we separate these targets into being acquired by publicly-traded (public), privatelyheld (private) or subsidiary fi rms.
In Panel B of Table 3 we present the CAARs for publicly-traded Indian targets acquired by publicly-traded US fi rms, privately-held US fi rms and subsidiaries of US fi rms. We fi nd a signifi cant gain of 4.08 per cent for the set of Indian targets as a whole. When we look at the abnormal returns by the type of US acquirer -publicly-traded, privately-held or US subsidiary fi rms we fi nd that the Indian targets realize signifi cant gains of approximately 4 per cent across all types of acquirers for the three-day event window. The above results indicate that the acquisition by any US fi rm (public, private or subsidiary fi rm) is benefi cial to the shareholders of the Indian targets. The shareholders of US acquirers of Indian targets suffer losses around the announcement of mergers and acquisitions.

Indian Acquirers and US Targets
In Panel A of Table 4 we show the CAARs for Indian acquirers of US targets. The acquisition announcement by Indian fi rms is associated with a 2.71 per cent stock price increases for the Indian acquiring fi rms. When we decompose the US targets into publicly-traded and privately-held fi rms, we fi nd that the Indian fi rms do not realize statistically signifi cant gains for acquisitions of publiclytraded US targets. Indian fi rms realize a statistically signifi cant gain of 3.09 per cent for the acquisition of privately-held US fi rms and a gain of 2.26 per cent for their acquisition of US subsidiary fi rms. In Panel B of Table 4 we report the CAARs of the US targets in the crossborder acquisitions. Our sample size is very small. For the sixteen publiclytraded US fi rms acquired by all types of Indian fi rms, the US targets experienced large positive abnormal returns but these were statistically insignifi cant. When we decompose the Indian acquirers into privately-held and publicly-traded companies, we fi nd large cumulative abnormal returns for the US targets but these are not statistically signifi cant.
These results indicate that Indian acquiring fi rms realize gains on the announcement of their acquisition of US fi rms although the gains are concentrated in acquisitions of privately-held or subsidiaries of US fi rms. No signifi cant gains are earned in the acquisition of publicly-traded US companies. Panel A presents the cumulative average abnormal returns for publiclytraded Indian fi rms which announced the acquisition of U.S. targets. The overall results in column 2 show positive abnormal returns for the shareholders of the acquiring fi rms. When we separate them into different groups, publiclytraded (public), privately-held (private) and subsidiary fi rms the results indicate substantial gains for the acquirers of the privately-held targets and subsidiary fi rm targets.
Panel B presents the cumulative average abnormal returns around the announcement date for the US target fi rms. The shareholders of the target fi rms gain from the announcement of mergers and acquisitions but this gain is statistically insignifi cant.

Public, Private and Subsidiary fi rm Acquirers and Public Targets
Obviously, the shareholders of public targets will agree to sell their perpetual dividend stream and future capital gains when they believe that the future prospects of the company's growth are bleaker than the buyer's beliefs. They would also agree to sell if they get a very lucrative offer from a buyer and there are possibilities for further negotiations of the prize. In either of these cases the merger is good news for the shareholders and should result in positive abnormal returns. Bargeron, Schlingemann, Stulz and Zutter (2008) fi nd that the premium paid for acquisitions of public targets is signifi cantly lower when the acquirer is a private fi rm instead of a public fi rm. They propose a managerial discretion theory of takeovers where managers may gain from the acquisitions that do not benefi t shareholders.
Due to the private benefi ts of acquisitions for managers, they pay more for target fi rms than shareholders would. Bargeron et al. (2008) provide evidence consistent with the managerial discretion theory of takeovers by showing that the difference in target shareholder gains between acquisitions by privately-held and by publicly-traded fi rms fall as the proportion of managerial ownership of the publicly-traded bidder increases.
Our results are consistent with Bargeron et al. (2007) in that we fi nd the abnormal returns for the targets of publicly-traded acquirers to be consistently higher than the abnormal returns of targets of privately-held acquirers; (a) the abnormal returns for the Indian targets of publicly-traded US acquirers are 4.49 per cent (signifi cant at 1% level) and the abnormal returns for the Indian targets of privately-held US acquirers are 3.55 per cent (signifi cant at 5% level), (b) the abnormal returns of the US targets of publicly-traded Indian acquirers are 18.50 per cent and the abnormal returns of the US targets of privately-held Indian acquirers are 3.89 per cent, though both of these are statistically insignifi cant.

Public-Traded Acquirer and Publicly Traded, Private and Subsidiary fi rm Targets
In section 4 we found that shareholders of US acquirers of Indian targets suffered losses on and around the announcement of mergers and acquisitions of Indian companies. However, shareholders of Indian fi rms gained on the announcement of their acquisition of privately-held US fi rms and subsidiaries of US fi rms. We investigate if this opposite reaction is due to fi rm specifi c characteristics by regressing the three-day cumulative abnormal return around announcement (-1 day to +1 day after announcement) on the following fi rm specifi c characteristics 10 : • Relative size = Value of deal/market value of acquirer. 11 Moeller and Stulz (2004) fi nd that for domestic acquisitions large fi rms loose when they acquire small fi rms whereas small fi rms gain when they acquire large fi rms. Moeller and Stulz (2004) fi nd that small acquirers acquiring large fi rms have 2 per cent higher announcement returns. • Acquirer size = Market value of acquirer. Mitchell and Stafford (2000) fi nd that announcement returns are positively related to the size of the acquirer for domestic acquisitions. • High-Tech = "1" if both target and acquirer are defi ned as high tech by the SDC Thomson data source and "0" otherwise. Conn et al. (2005) fi nd that fi rms (both acquiring and target) in the same high-tech industry have a positive infl uence on the returns of the acquirer. • Related = "1" if target and acquirer have the same four-digit SIC code and "0" otherwise. This variable measures if the merger is within the industry or if the fi rms are diversifying. Megginson, Morgan and Nail (2004) fi nd that for domestic acquisitions gains are higher in related acquisitions. • Value = "1" if acquirer's market to book value of equity is in quintile one and "0" otherwise. • Glamour = "1" if acquirer's market to book value of equity is in quintile fi ve and "0" otherwise. These defi nitions of value and glamour variables are from Conn et al. (2005). • TQ = Tobin's q of the acquirer. We estimate Tobin's q as the market value of fi rm divided by the book value. • Downer= Dummy equals "1" if the majority stake is acquired (greater than 50%) and "0" otherwise. Chari et al. (2010) fi nd that acquirers from the developed world gain when they announce acquisition of majority stake in fi rms from the emerging world. • DAQC = "1" if the acquirer is from A and "0" if the acquirer is from India.
We provide the summary statistics by cross-border relationship for the above variables in Table 5. We fi nd that the Indian and the US acquirers of private targets have similar high-tech status levels. The Indian acquirers of publiclytraded US targets and US subsidiary fi rm targets are more often classifi ed as high-tech status as compared to US acquirers of publicly-traded Indian fi rms and Indian subsidiary fi rm targets. Irrespective of the target fi rm status, Indian acquirers appear to engage in company diversifi cation more than US acquirers. The relative size of the targets is substantially smaller for US acquisitions. The market value of Indian acquirers of US targets is also signifi cantly lower than the market value of US acquirers of Indian targets. Though the average percentage owned after the transaction by Indian acquiring fi rms is higher than the percentage owned after the transaction by acquirers, both Indian and acquirers, acquire majority stakes in the target fi rms. For both the US and Indian acquirers, the average stake acquired in the private and subsidiary fi rm targets is higher than the average stake acquired in publicly-traded targets. In Panel A we present the summary statistics for the Indian acquirers of US targets and in Panel B we present the summary statistics for the US acquirers of Indian targets. A comparison of Panel A and Panel B shows that the relative size of US targets is substantially higher than the relative size of Indian targets and the US acquirers have higher market value than the Indian acquirers. The Indian acquirers of the US targets diversify their business more than the US acquirers of the Indian targets.
We regress the CAAR for the acquirers of public, private and subsidiary targets on these variables to investigate whether there are abnormal return differences to the acquirer's home country after controlling for fi rm characteristics. The data for relative size and percentage owned after transaction is available for fewer fi rms hence for each of the target types we fi rst run the regression without the variables, 'relative size' and 'Downer' as the lack of data on these variables substantially reduces the sample size. We use the following three models to test if the CAARs are country dependent: Mode (6) Model II: Model III: (8) where LMV is the log of market value of the acquirer. EXC in the above equation is the exchange rate variable 12 which measures the effect of changes in the exchange rate on mergers and acquisitions. We follow Cakici et al. (1996) who estimate the effects of exchange rate variable on bidding shareholders' wealth as a result of foreign acquisitions of the US fi rms, and compare these to a control sample of foreign acquisitions by US fi rms.
We pool the country data and run the above regression models for the acquirers of publicly-traded, privately-held and subsidiary targets separately. The results are reported in Table 6. For the acquirers of publicly-traded and subsidiary targets we fi nd an insignifi cant DAQC variable suggesting that there is no country difference in the abnormal returns of the Indian and the US acquirers of the publicly-traded and subsidiary targets.
In the following table we present the regression on the CAAR (-1 day to +1 day after the announcement) as a dependent variable for the acquirers of public, private and subsidiary targets. After controlling for fi rm characteristics we fi nd that the dummy variable for country (DAQC=1 for US acquirers and DAQC=0 for Indian acquirers) is negative and signifi cant for the acquirers of private targets suggesting a lower abnormal return for the US acquirers as compared to the Indian acquirers of private targets.
The coeffi cient for the sample of publicly-traded acquirers of privatelyheld targets is negative and signifi cant suggesting a lower CAAR for the US acquirers of privately-held Indian targets as compared to the Indian acquirers of privately-held US targets. 12 We compute the exchange rate variable using the method adopted by Cakici et al. (1996). We take the Indian currency's average exchange rate (per dollar) for the sample period 1995-2007 and subtract the Indian currency's exchange rate for the year of acquisition. Then we divide this difference by the average exchange rate.

Conclusion
Prior empirical research on mergers and acquisitions is predominantly related to the developed markets of the US and Europe. There is limited work on mergers and acquisitions when targets from the developed world are acquired by companies from the developing world. In the cross-border mergers and acquisitions between developed and developing world fi rms, researchers have studied the effect of the announcement of mergers and acquisitions on the acquirer fi rms only. None of the studies to date have investigated the effect of announcement of mergers and acquisitions on the target fi rms. In this paper we studied the cross-border mergers and acquisitions between the US and India for both acquirers as well as target fi rms.
We used data on mergers and acquisitions from January 1995 to August 2007 and found that the mergers and acquisitions with US acquirers and Indian targets resulted in signifi cant losses for the acquirers and signifi cant gains for the targets. Mergers and acquisitions between Indian acquirers and US targets resulted in signifi cant gains for the acquirers and insignifi cant gains for the targets.
We further examined the abnormal returns by decomposing the sample of fi rms into publicly-traded, privately-held and non-traded subsidiary fi rms. We found that the US acquirers of publicly-traded Indian fi rms realize insignifi cant losses while publicly-traded Indian targets acquired by US fi rms earned signifi cant returns on the announcement of a merger or acquisition. Indian acquirers of publicly-traded US fi rms earned insignifi cant gains/losses and US targets of publicly-traded Indian acquirers earned insignifi cant positive abnormal returns on the announcement of mergers and acquisitions. These results are similar to the results we found in the existing literature. Variable  I  II  II  I  II  II  I  II  In the cross-border acquisition of private targets, publicly-traded US acquirers suffered losses whereas publicly-traded Indian acquirers realized gains. The Indian acquirers of US subsidiary fi rm targets realized signifi cant gains while the US acquirers of Indian subsidiary fi rm targets realized insignifi cant gains/losses. We investigated these cross-border merger and acquisition results to determine if the results were due to fi rm characteristics or country dependent by regressing the cumulative abnormal returns on the fi rm characteristics and a dummy variable for the country of the acquirer.
For the private targets, we found the country dummy variable to be signifi cant after controlling for fi rm characteristics indicating a lower abnormal return for US acquirers of privately-held Indian fi rms as compared to the Indian acquirers of privately-held US fi rms. The dummy variable is not signifi cant for the acquirers of publicly-traded fi rms and for subsidiary fi rm targets indicating that fi rm characteristics and not the country environments govern the returns on these mergers and acquisitions.
Author information: Professors Gordon Karels, Edward Lawrence and Jin Yu are staff member of the fi nance faculty respectively at the University of Nebraska-Lincoln, Florida International University and St. Cloud State University, all in the USA. They may be contacted at: gkarels@unl.edu; edward. lawrence@fi u.edu; iyu@stcloundstate.edu.