WORKING PAPER SERIES

Transcription

1 College of Business Administration University of Rhode Island William A. Orme WORKING PAPER SERIES encouraging creative research Cross Border Mergers and Aquisitions Using ADRs as Considerations James E. Owers, Bing Xuan Lin and Ronald C. Rogers 2005/2006 No. 5 This working paper series is intended to facilitate discussion and encourage the exchange of ideas. Inclusion here does not preclude publication elsewhere. It is the original work of the author(s) and subject to copyright regulations. Office of the Dean College of Business Administration Ballentine Hall 7 Lippitt Road Kingston, RI

2 Cross Border Mergers and Acquisitions Using ADRs as Consideration James E. Owers Department of Finance Robinson College of Business Georgia State University Atlanta, GA (404) Bing-Xuan Lin College of Business University of Rhode Island Kingston, RI (401) and Ronald C. Rogers Moore College of Business University of South Carolina Columbia, SC (803) This draft October 2005 Do not quote without authorization The excellent research assistance of Parijat Cheema, Kaysia Campbell and Girish Tamankar is gratefully acknowledged. Our thanks to the RCB of GSU for making this assistance available. 1

3 Cross Border Mergers and Acquisitions Using ADRs as Consideration Abstract We examine mergers and acquisitions wherein American Depositary receipts (ADRs) are the primary means of consideration. The majority of these transactions involve the acquisition of U.S. Corporations by non-u.s. firms using their ADRs. We show that acquisitions using ADRs involve variations on a theme and outline the institutional details. Next we empirically examine the implications of these variations and demonstrate how different sub-samples differ in terms of the strategy for the acquisition, transactional details, and valuation consequences. We also find that acquirer s performance is better when the target is in a different industry. We relate our findings to the recent debate on the diversification discount. 2

4 Cross Border Mergers and Acquisitions Using ADRs as Consideration I. Introduction Mergers and acquisitions have generated research topics for many years. As these studies have come to examine the impact of more precise attributes of transactions, both cross-border features and the type of consideration (equity or cash) have been researched. In this paper the cross border and type of consideration dimensions are brought together to study mergers and acquisitions in which ADRs are used as a form of payment. ADRs are the securities of companies domiciled in other nations trading in the U.S. with similar institutional arrangements to the stocks of companies with headquarters in the U.S. 1 Although the use of ADRs as a form of payment has received little if any attention in the research literature, there has emerged a pattern of non-u.s. firms with ADRs to use them in making acquisitions. 2 This is an interesting overlap between the processes of mergers and acquisitions and listing abroad from the home country. Most of the transactions in this paper are non-u.s. companies employing their ADRs to purchase U.S. firms but in some instances the acquirer is using ADRs traded in the U.S. to acquire a company in a third country. As such, these transactions are interesting parts of International Finance. The primary roles of this paper are to document this unique type of acquisition activity and to examine whether the use of ADRs affects valuation changes or other attributes associated with these acquisitions. While transactions using ADRs are relatively few compared to all merger and acquisition activity, they nevertheless occur in material volume and incorporate a 1 See, for example Karoli (1998) and Nanda, Owers and Feng (1996). Although the first formal ADR was created in 1927, the essential elements of the instrument can be found in arrangements and transactions during WW I. 2 Given that ADRs are by definition issued by Non-U.S. firms, the terms Non-U.S. and foreign can be used inter-changeably from a U.S. perspective. 3

5 range of institutional and transactional variations. We find that the overall revaluations associated with ADR-consideration acquisitions are similar in profile to those associated with purely domestic acquisitions and international acquisitions not employing ADRs. However, there are notable variations from the overall profile for particular firms and transactions, indicating that these are unique transactions in additional to the somewhat unusual institutional arrangements. ADRs provide an institutionally convenient and cost-effective vehicle for international diversification for investors. For issuing firms they provide access to the large scale U.S. capital markets and added international visibility (from both operations and financial perspectives). 3 It is surprising that the potential benefit of using ADRs as a means of consideration in international acquisitions has received very little (if any) attention in the research literature. In this paper, we examine this subset of mergers and acquisitions and find that targets experience significant increases in value on announcement and acquirers experience returns exhibiting considerable variation. Both merger transactions and the acquisitions of part of targets produce positive returns for acquirers. Interestingly, acquirers returns are higher when targets operate in different industries from acquirers. The remainder of the paper is organized as follows. Section II identifies the relevant literature that provides the context for the study. The hypotheses employed to structure the empirical analysis are spelled out in section III. Details pertaining to the sample, data and methodology employed in the empirical analysis are described in section IV. The findings are presented and interpreted in section V. Section VI summarizes and concludes the paper. 4

6 II. Related Literature There is a voluminous literature pertaining to the topic of this paper and we will address the major papers relevant to the development of the hypotheses of this study under two headings: (1) the literature on mergers and acquisitions and (2) the relevant international finance literature. The Mergers and Acquisitions Literature The transactions examined in this paper are mergers and acquisitions. In most cases they are the acquisition of all of a target firm. In the remainder they are either partial acquisitions of a target (i.e. a toehold ), or completing acquisitions where part of the target was previously owned by the acquirer (often referred to as a wrap-up ). In most cases there is a clear identification of the acquirer and target, with mergers of equals and the formation of a new firm/security comprising few transactions. Findings from earlier studies were summarized by Jensen and Ruback (1983) and Jarrell, Brickley and Netter (1988). The profile of findings from the early studies showed that targets almost universally increase in value significantly whereas acquirers typically experience small declines in values. These transactions have been extensively examined from perspectives such as auction theory and the winner s curse is an established part of the literature. Subsequent studies have added significant refinement to the understanding of these transactions. Arande, Mitchell and Stafford (2001) review some of the more recent findings and evolving perspectives that have come forth since the earlier studies. Most details of mergers and acquisitions have now received attention in both academic and practitioner literature. These include analyzing the economic monetary effects in addition to the percentage abnormal returns 3 While there are 460 Non-U.S. companies listed on the NYSE and in recent years the number of ADRs traded in the U.S. has generally increased, there are both benefits to and costs of such listings and some firms have withdrawn 5

7 (Dennis and McConnell [1984] and Bradley, Desai and Kim [1988]). The significance of the type of consideration was noted by Asquith, Brunner and Mullins (1987), Huang and Walkling (1987), Travlos (1987), and Heron and Lie (2002) and has now been extensively examined. The findings from this now extensive series of method-of-payment studies are succinctly summarized by Bharadwaj and Shivdansi (2003). They note that It is well known that the method of payment for an acquisition has an important influence on acquirer returns, with acquirers earning returns that are close to zero, on average, in cash tender offers and significantly negative returns in stock tender offers. Recent avenues of empirical examination in determining the distribution of gains between bidders and targets include the relative reputation of financial advisors who provide advice on both the type of merger to pursue and the bargaining strategies. Kale, Kini and Ryan (2003) find the relative reputation of the (respective) advisors to be significant in determining wealth creation and its distribution in mergers and acquisitions. Theoretical work pertaining to mergers and acquisitions has also been extensive. The findings of average negative returns to acquirers led to theories pertaining to agency considerations and free cash flow (Jensen and Meckling [1976], Jensen [1986]), and the role of hubris (Roll [1986]). Behavioral finance theories have also been put forward to structure the examination of mergers and acquisitions. In examining the way in which acquisitions are motivated by stock market conditions, Schleifer and Vishney (2003) position their theory as follows: This theory is in a way the opposite of Roll s (1986) hubris theory of corporate takeovers, in which financial markets are rational, but corporate managers are not. In our theory, managers rationally respond to less-than-rational markets. their ADRs. See for example, Mexican Firms leave NYSE, The Wall Street Journal, January 17,

8 The International Finance literature The directly relevant segments of the international finance literature relating to this paper are studies of ADRs and cross-border acquisitions. We take these in turn. ADRs. ADRs have a long and intriguing history. After the essential features developed during World War I, the first ADR was formally established in The role of ADRs was limited first by the economic conditions of the 1930s and then World War II. During the 1950s ADRs were issued first mainly by European firms. After restrictions were eased, Japanese firms began issuing ADRs. By 1961, 150 firms from 17 countries had ADRs. But growth was limited after the Interest Equalization Tax (IET) was put in place during 1963 to address concerns about an outflow of American capital. After the elimination of the IET in 1974, the ADR markets grew rapidly. As ADRs became numerous and popular as a vehicle for implementing international portfolio diversification strategies, their investment performance came to receive considerable attention in studies such as Foerster and Karolyi (2000) and Nanda, Feng and Owers (2000). The overall profile of their investment performance is generally similar to findings for purely domestic investment vehicles and situations. For example, the investment performance of ADRs that are IPOs has been found to be generally similar to domestic IPOs. Privatizations using ADRs have similar investment performance to purely domestic privatizations. Although ADRs have come to receive considerable attention in the research literature, we are not aware of any other study that has examined the role of ADRs in mergers and acquisitions. Hence this study examines both the institutional attributes of these transactions and the valuation changes associated with them. 7

9 Cross-border acquisitions A number of studies have examined the consequences of U.S. forms acquiring foreign firms. Doukas and Travlos (1989) found generally insignificant valuation consequences for the U.S. acquirers with the exception of when the acquisition was the first major step of the acquirer into the target s country. Lin, Madura and Picou (1994) found material variation in U.S. acquirer valuation reactions according to the domicile of the target. Acquisition of German firms was associated with positive abnormal returns. In contrast, acquisitions of British and Canadian targets were associated with negative abnormal returns. Kang (1993) examined matched-pairs of acquisitions between Japanese bidders and U.S. targets. He found statistically significant wealth gains for both firms. As is appropriate for cross-border acquisitions, he placed his analysis in the context of Direct Foreign Investment (DFI). Specific attributes of bidder firms and exchange rate movements were found to be systematically associated with bidder returns. The contrast between this finding and the profile for purely domestic transactions of target gains and acquirer losses (or essentially zero reaction with cash transactions) is notable. 4 There are apparently different valuation consequences to be found with cross-border acquisitions compared to purely domestic transactions. These differences indicate the potential for the particular type of acquisition examined in this paper to add further insights into the significance of particular details of transactions such as being cross-border and using a somewhat unusual method of payment. 4 See Bruner (2002) for an integrated interpretation of findings of Merger and Acquisition studies from both academic and practitioner perspectives. 8

10 III. Hypotheses The acquisition in 1999 of PacifiCorp by Scottish Power PLC illustrates a number of characteristics associated with acquisition by ADR. The acquisition created the first combination of power utilities in the U.S. and United Kingdom. It was a significant transaction ($6.5 billion) and the merged unit planned to dispose of some of PacifiCorp s properties (including an interest in an Australian power station). Reference to the familiar synergy and cost efficiency motivations for acquisitions was to be found in the chief executive of the acquirer noting that the deal will generate significant cost savings for the merged group. Each PacifiCorp share was to be exchanged for 0.58 ADR of Scottish Power PLC. Each ADR represented 4 shares of Scottish Power. More commonly, an ADR will represent 5 or 10 units of the home country shares. 5 The ADRs of Scottish Power dropped 9.48% (the CAR over [-1,+1]) on announcement of the transactions. They fell further when completion of the acquisition was announced. On the date of announcement, the value of PacifiCorp shares experienced a 4.7% increase. Scottish Power s subsequent experience with PacifiCorp was not as positive as anticipated at the time of the acquisition and in May 2005 it agreed to sell PacificCorp to the MidAmerica unit of Berkshire Hathaway for $5.1 billion in cash plus debt assumption of $4.3 billion. This was Berkshire Hathaway s largest acquisition since 1998 and Warren Buffet noted that it was a transaction in line with the strategy of using some of Berkshire s $40 billion cash for acquisitions in the energy sector. On announcement, Berkshire Hathway A shares increased by 1.8% and those of Scottish Power (its NASDAQ traded ADR) by 5.7%. The focus of this paper is an empirical analysis of mergers and acquisitions wherein ADRs are employed as a form of consideration. As such, the voluminous theoretical and 5 See Muscarella and Vetsuypens (1996) for additional details on ADR/home country share relationships. 9

11 empirical literature pertaining to mergers and acquisitions provides context for this analysis. In this context, we posit the following hypotheses to structure our analysis of these transactions. Similar Overall Profile Hypotheses The market forces and processes pertaining to acquisitions using ADRs are generally similar to those for purely domestic acquisitions. Thus it is hypothesized that the average valuation consequences for the target and acquirers will be consistent with the cumulative profile of studies to date: - H1a: That targets will experience significant positive abnormal returns - H1b: That since equity securities are used by the acquirer, they will incur small negative abnormal returns. The issues associated with these hypotheses will be investigated for the separate subsamples of targets and acquirers and then a matched pair sub-sample. Variations of Returns with Transaction s Attributes There are significant variations in the details of acquisitions using ADRs. These include: - Whether the transacting firms are from developed or developing economies; - Whether the target is a U.S. or foreign firm; - Whether only ADRs are used or a combinations of considerations; - Whether 100% of the target is being acquired (in contrast to a toehold or wrap-up of the remaining part of the target); - Whether the size of the transaction impacts the valuation consequences; and 10

12 - The significance of the acquirer and target having the same SIC code. The significance of these variations in institutional transactional details will be investigated. The null hypothesis is that these attributes will affect return outcomes for both targets and bidders. IV. Data & Methodology The source of the sample of transactions came from the SDC database where it listed the means of consideration between 1985 and From the original listing of 107 acquisition transactions involving ADRs, we identify 107 targets and 98 acquirers wherein ADRs are employed in the transaction. The original listing was then examined using the customary filters for confounded events, sufficiency of transaction details, and data availability requirements necessary to apply the methodology. We confirmed SDC dates with reference to financial press sources such as The Wall Street Journal and Lexis/Nexus. The transactions were tracked separately for targets and acquirers (respectively) that were amenable to empirical examination. There are 55 targets and 78 acquirers meeting the criteria for empirical analysis. Of the separate sub-samples, there were 41 transactions where both the target and acquirers were examined and this is referred to as the match set sub-sample of transactions. Table 1 enumerates on these and other sample attributes. Table 1 shows the wide variation in number of transactions per year. This varies from 0 in 1986 to 29 in It also reflects the wide value range of ADR acquisitions from a low of $4.5 million in 1989 (one transaction) to an average of $5.5 billion in 1995 (with 4 transactions). The ADR returns data are from the CRSP daily returns file. The returns data requirements necessary to run the market model over an interval from 160 to 31 days before the 11

13 transactions was a significant source of attrition between the original (unfiltered) sample listing and the final empirical sample. Reaction of the equity securities for transactions in the sample is examined using established event-time methods. Parameters of the market model are estimated over the interval extending from day -160 to day -31 relative to the announcement on day 0. Abnormal returns are estimated for each day in the event interval (days -30, +30) and for various identified intervals around the announcement (day 0). For each security j, the market model is used to calculate an abnormal return (AR) for event day t as follows: AR jt = R jt - (a j + β j R mt ) (1) where R jt is the rate of return on security j for event day t, and R mt is the rate of return on the Center for Research in Security Prices (CRSP) value-weighted index on event day t. The coefficients α j and β j are the ordinary least squares estimates of the intercept and slope, respectively, of the market model regression. The cumulative abnormal return (CAR) from day T 1j to day T 2j is defined as: CAR = j T 2j Σ AR (2) t = T 1j j We cumulate over various intervals around the announcement date. For a sample of N securities, the mean CAR is defined as: CAR = Ν Σ CAR / N (3) j=1 j The expected value of CAR is zero in the absence of abnormal performance. 12

14 The test statistic described by Dodd and Warner (1983) is the mean standardized cumulative abnormal return. To compute this statistic, the abnormal return AR jt is standardized by its estimated standard deviation s jt. The value of S jt is: S 2 1 = S j (1= D j + ( R - R D j )2/ ( R t = 1 2 jt mt m Σ where mt - R m ) 2 ) 2 S j D j R mt R m R mt = residual variance for security j from the market model regression = number of observations during the estimation period = rate of return on the market index for date of the event period = mean rate of return on the market index during the estimation period = rate of return on the market of day t of the estimation period SAR jt = AR jt / s jt (4) employed. The SARs are cumulated and both Z and t test statistics for the sample of N securities are V. Findings and Interpretation The findings are presented first for the separate sub-samples of targets and acquirers. The matched-pairs sub-sample is then examined. Given the substantial cross-sectional variation in the individual company results, we subsequently undertake a cross-sectional analysis in pursuit of the attributes that are most significant for variation from the overall profiles as calibrated in the averages. 13

15 Targets The findings for the 52 target firms for which there were sufficient data to apply the methodology are reported in Table 2. It comes as no surprise that targets experience significant positive abnormal returns when a bid is initially made. Hypothesis 1A is strongly supported. Some potentially intriguing aspects of the findings are NOT explained by the international nature of these transactions. The day +1 AR is notable. Since the returns data are from CRSP they are generated using U.S. data that comes at the end of the trading day. As the global markets function, Asian markets begin the trading for a specified calendar date, following by European and then North American. Thus the notable abnormal returns on day +1 are not explained by date line considerations, making this of considerable interest and something for which we do not have a convincing explanation or even a credible conjecture. Given the speed of transmission of information that applied for all of the sample period, we find it somewhat surprising that the consequences for the target firm are impounded over an interval that includes the calendar day after the formal announcement date. As noted above, with the exception of the Day +1 abnormal return, the positive target abnormal returns reflect the silhouette from studies of domestic and cross border transactions. However, there are some contrasts with the findings of previous merger and acquisition studies in regards to the consequences for target firms. First, the percentage magnitudes are somewhat smaller. This overall sub-sample includes transactions in which less than 100% of the target is acquired and is a potential factor in the lower abnormal returns that is addressed below. Second, there is a significant accumulation of CAR beginning at approximately day 12. There is a clear economically material response to the announcements for the target firms, including some notable pre-announcement CAR accumulation. Table 2 Panel B reports 14

16 that there are notable pre-announcement patterns on days 7 and -3. However, a closer examination of both the number of positive/negative individual firm ARs and the difference between the Z and t statistics suggests these are driven by a relatively small number of specific firms. The notion that a small number of firms generate the pre-announcement abnormal returns of note is further supported by the tests of significance of average CAR accumulation reported in Table 2 Panel C. The CAR accumulation over (-10,-2) of 1.41% (encompassing 28 positive and 24 negative individual firms results) is not statistically significant at even the 10% level. In Table 2 Panel C, we also see that the ARs over (-1, 0) and (0, 0) are both positive and significant at the 1% level. These results are in line with the findings in prior researches where target firms experience price appreciation upon deal announcements. Figure I portrays the CAR pattern over the 41 day interval ending 10 days after the announcement. It profiles the notable but statistically insignificant CAR accumulation from day 10 to day 2 and stabilizes with a CAR of approximately 20%. Examination of the individual daily ARs for the sample target firms comports with the above statistical findings. For a small number of firms there are notable ARs before the transactions are announced. While interesting, they are not a major focus of this paper and remain a potential subject for examination in a subsequent study. Acquirers By definition, these are acquisitions wherein at least part of the consideration is equity securities in the form of ADRs. The extensive research (both theoretical and empirical) has established the rationale for and realization of negative equity value reactions to acquisitions 15

17 with equity consideration. That context generated our hypothesis 1B that acquirers will experience negative abnormal returns associated with the transaction. That is what our empirical analysis finds and hypothesis 1B is supported. Table 3 documents and figure II profiles significant negative average acquirer reaction to the announcements of these transactions. The average CARs for the 78 acquirers over the event intervals are negative and the number of negative responses does dominate. For example, on day 0 there are 47 negative acquirer ARs. But that still leaves 31 positive. The average negative reaction should thus be interpreted with caution. Clearly there are negative reactions that on average are greater then the positive responses and the sign test in Table 3 Panel B attests to the significance of this difference. However, the lack of overwhelming predominance of negative acquirer CARs indicates material variation in outcomes for acquirers. The difference between the Z and t teststatistics also indicates substantial variation in the abnormal return outcomes. This aspect of the acquirer value changes associated with the transaction will be visited in greater detail in the analysis of the match-pair sub-sample. The results reported in Table 3 and depicted in Figure II show acquirer negative abnormal returns consistent with findings for domestic acquisitions wherein equity is the method of payment. The cumulative abnormal returns over (-10, 0) are 1.64%. 0.62% of this accumulates over (-1,0) and 52 of the 78 acquirers experience negative two-day abnormal returns. The CAR (-1,0) is significant at the 1% level of significance as calibrated by the Z score but is not significant according to the t statistic. Match Pairs of Transactions 16

18 As examined above, the sub-sample of targets is such that 52 firms had attributes (nonconfounded events, data availability) enabling the event analysis to be performed. In the case of acquirers, the corresponding number is 78. The intersection of these two sets was 41 transactions in which both transacting firms were in the empirical analysis. We now look in more detail at those 41 transactions to glean further insights into the nature and consequences of these transactions. Table 4 reports the abnormal returns for the 41 match-pairs target firms. Mirroring the previous format, Panel A reports daily abnormal returns, the CAR from day 30 to day +10 and the percent positive/negative. Panel B reports selected interval CARs and associated test statistics. As a comparison of tables 2 and 4 shows, there is not a notable difference in either economic magnitudes or statistical significance between the sub-sample of 55 separate targets and the 41 matched-pair targets. This is not surprising since the intersection is 75% of the separate target sub-sample. In a manner similar to that for the separate target sub samples, there is not a marked difference between the 41 match-sample acquirers when compared to the separate sub-sample of 78 acquirers. Not surprisingly there is somewhat more difference than in the case of the targets given that the intersection is only approximately 53%. Nevertheless there are not marked indications that the match-pair sub-sample is materially different from the separate sub-samples. For subsequent analysis we thus focus on the 41 match-pair sub-sample transactions. What is more interesting about the match-pairs sub-sample is reported in Table 6. Here the pairs of CAR accumulation over the various intervals identified in the table are classified according to whether 17

19 the pair-wise targets and acquirers in specific transactions individually experienced positive or negative CARs. Notably for the immediate interval around the announcement, while 25 (of the 41) transactions generated the typical profile for equity consideration acquisitions of target positive and acquirer negative accumulation, for 13 of these transactions both firms experienced positive revaluations. Generally similar patterns are observed over the longer intervals around the announcement. Over (-5, +1) the corresponding numbers are 26 and 10. For ( 10, +1), they are 23 and 14. This set of outcomes indicates a generally more positive set of shared experiences for the equity claimants of both targets and acquirers than is typically the case when equity is the means of consideration in domestic acquisitions. It fits with the profile of more positive acquirer experiences in cross border acquisitions such as the findings in Kang (1993) that was referenced in the literature review section. There, studies of cross-border acquisitions noted circumstances (such as the domicile of acquirers and targets and acquirer firms attributes) associated with positive acquirer outcomes as calibrated by equity value reaction. Our findings suggest a generally more positive acquirer outcome when using ADRs than in domestic transactions. Cross Sectional Variations Hypotheses As previously noted, there are substantial variations in the institutional details of the overall sample of acquisitions employing ADRs examined in this paper. The significance and consequences of these variations have been partially examined in the preceding analysis of various sub-samples. We now undertake a cross- sectional regression in order to further investigate the significance of different details of the transactions. 18

20 Hypothesis #2 addresses variations of return outcomes with transaction attributes. Among the major significant transactional details are: - Variable 1: Whether the transacting firms are from developed or developing economies; - Variable 2: Whether the transaction is a traditional acquisition or a merger; - Variable 3: the significance of the acquirer already having an ownership interest in the target (measured by the % already owned [i.e. a wrap-up ]) - Variable 4: the significance of the acquirer purchasing part of the target rather than all the outstanding equity interest ( measured by the % not acquired [i.e. a toe-hold ]) - Variable 5: Whether the size of the transaction impacts the valuation consequences; and - Variable 6: The significance of the acquirer and target having the same SIC code. In order to examine whether these variations have a material impact of the valuation outcomes as hypothesized, we undertook cross-sectional regression analyses. The 3-day Car (-1, +1) is the dependent variable and it is regressed on independent variables representing these 6 attributes identified as major institutional variations across transactions in the sample. In some cases, the subset of the samples with a particular attribute was very small. For example, mopup transactions comprised only 5 of the 55 target acquisitions (4 of the 41 matched pair subsample). This examination provided substantially more interesting insights in the case of acquirers rather than targets. Interestingly, for target firms, none of these attributes had a significant impact. For targets being acquired with the use of ADRs as a means of payments, none of the six institutional attributes of the transaction had a statistically significant on the 3-day announcement window CAR. Both the regression results and a comparison of the event study 19

21 results for the respective partitions indicated no significant difference based on these transaction attributes. We thus reject the null hypothesis that these attributes significantly influence return outcomes for targets. This lack of systematic impact of some attributes is notable. In particular, given the significance of the SIC matching being of consequence to acquirers (as discussed below), its lack of significance for the target is notable. While it is shown below that it is a relative plus for acquirers to purchase unrelated targets, this attribute does not significantly affect the returns experienced by the target firm. In contrast to the lack of significance for targets, for acquirers using ADRs there are some significant influences on the return outcomes according to these transaction attributes. In this context, we present in Table 7 and further discuss the regression results for acquirers only. Attribute #4: Toe-hold Acquisitions The most significant institutional variation relates to attribute #4 wherein the acquirer is purchasing only a toehold and not all the target. This has a positive impact on acquirer outcomes. Intuitively this is consistent with the notion that since acquisitions (employing equity consideration) typically have a negative valuation consequence for the acquirer, purchasing only a party of the target is a relative plus. A sample transaction that falls into this category is the widely discussed toehold that UK based British Telecom acquired in MCI in Attribute #2: Mergers In transactions that are effectively mergers (attribute #2) the consequences for the acquirer are mitigated in terms of negative CAR. We are cautious about classifying one firm an acquirer and the other a target when the transactions is effectively a merger but accepted the SDC categorization in these transactions. We note that there are only two transactions in our 20

22 match-pair sample that appeared to be mergers in contrast to the dominant profile of transactions with a clear delineation of target and acquirer roles. Attribute #6 Acquisitions of Unrelated Targets: We find the significance of attribute #6 to be of considerable interest. This is whether the SIC codes of the target and acquirer are two-digit matches. This SIC match criterion has a marginally important (5.7% level of significance) impact of acquirer CAR and it is negative. The acquirers with targets that do NOT match their Line of Business (at the two-digit SIC level) have a less negative/more positive consequence associated with the acquisition. This is notable in light of the ongoing analysis of whether there is a diversification discount. The research into the valuation effects of diversification is now very extensive. There are numerous studies identifying potential value enhancing effects of diversification. Weston (1970) focused on synergies, Berger and Ofek (1995) examined tax consequences, and Stein (1997) considered internal capital markets. There is also an extensive literature supportive of a diversification discount. Lang and Stultz (1994) address the multi-segment firm discount. The positive re-valuations associated by undoing of diversification mergers by divestitures examined in works such as Bhagat, Schleifer and Vishney (1990) and Kaplan and Weisback (1992) lend further support to the diversification discount. Scharfstein and Stein (2000) captured a profile of thinking when they noted that it has become almost axiomatic among researchers in finance and strategy that a policy of corporate diversification is typically value reducing. Yet Akbulent and Matsuska (2003) study announcement returns associated with mergers from 1950 to 2002 and generally challenge the notion that diversification destroys value. Recent work reviewing both the existence and interpretation of the diversification discount include Campa and Kedia (2002) and Villolonga 21

23 (2004). In addition to stock market evidence, some studies have been using micro-data to examine the productivity implications of diversification. While Lichtenberg (1992) and Maksimovic and Phillips (2002) conclude that diversification depresses productivity, Schoar (2002) concludes that at the individual plant level, diversification is associated with increased productivity. However, the increased productivity associated with newly acquired plants is offset by a reduction in overall firm productivity. The issue of the diversification discount is not a major theme of this paper and the empirical finding that diversification reduces the negative effect on acquirers is the main comment we want to make. However, in light of the diversification controversy and inconsistent conclusions from the studies looking at micro productivity data underlying stock market valuations, we find stock market evidence of diversification on average being positively regarded in cross border acquisitions using ADRs to be a potentially very fruitful area for subsequent detailed research. VI. Summary and Conclusions In this paper we examine a type of merger and acquisition transaction that does not appear to have previously been examined in the research literature. These are cross-border acquisitions wherein the means of consideration is primarily ADRs. Both cross-border mergers and acquisitions and ADRs have separately received considerable attention in the research literature but we are not aware of any previous study of transactions wherein they come together in a particular institutional arrangement. We identified an overall sample of 107 such transactions over the interval from 1985 to Of the 78 acquirers and 52 targets that met data and transactional criteria and could be 22

24 examined separately, there was an overlapping sample of 41 matched pair transactions wherein the data and transaction criteria for both firms were met. Overall, the findings of these transactions comport with the findings for purely domestic transactions. Namely, targets experience significant increases in value on announcement and, since these are acquisitions using equity, acquirers on average experience negative returns. However, as the statistical tests suggested and individual acquirer outcomes indicated directly, the acquirer returns experienced considerable variations and suggested a closer examination. The matched-pair analysis showed that a surprisingly large portion of the transactions were associated with positive returns for both acquirers and targets. For the interval (-1,+1), 13 of the 41 matched pair transactions had positive returns for both the acquirer and target. The analysis addressed several attributes on which the transactions varied. These included the developed/developing domicile of the firms, whether the transaction was classified by SDC as an acquisition or merger, whether the acquirer was purchasing all or part of the target, the size of the transaction, and whether the SIC codes of the acquirer and target matched (at the two digit level). None of these attributes significantly impacted the target returns. However, there were some significant impacts for acquirer returns. Both merger transactions and the acquisition of a part of a target were positives for acquirer returns. Also positive for acquirer returns was when the target line of business did NOT match the acquirer s as calibrated by the two-digit SIC code. This is an interesting finding in that it provides evidence contrary to the widely accepted diversification discount that is financial market based and not productivity based as in some recent studies questioning the existence of a diversification discount. 23

25 This paper is a first investigation of a type of transactions not previously examined in the research literature. As such, several lines of subsequent investigation remain. These include a closer examination of the diversification discount/premium attribute of industrial organization and corporate restructuring. Do cross-border transactions wherein ADRs are employed have the potential to provide further insights into this issue that was in the not so distant past considered to be settled only to have significant questions arise from recent micro-data productivity studies? Other avenues for further examination emanate from Kang s (1993) lines of investigation in his study of Japanese acquisitions of U.S. firms. These include an examination of the firm-specific attributes of acquirers and targets and the role of country pairing so as to capture any exchange rate influences. The pre-announcement CAR accumulation for some targets also appears to provide potential for systematic study of how varying cross-border sanctions for insider trading affect the use of strategies based on awareness of upcoming acquisitions. 24

26 Table 1: Sample Description: The data includes cross-border mergers and acquisitions using ADRs for the sample period of Transaction activity sourced from the SDC database. The number of transactions is the aggregate before filters for data requirements and elimination of confounded transactions. Mean Value of Calander Year No. of Transactions No. of Acquirers No. of Targets No. of Matched Pairs Total Value of Transactions ($ millions) Transactions ($ millions) , , , , , , , , , , , , , , , , , , , , , , , Total $334, yr Avg: $2,

27 j j Table 2A: Event Study Results for Target Firms For each security j, the market model is used to calculate an abnormal return (AR) for event day t as follows: AR jt = R jt - ( a + β Rmt ) where Rjt is the rate of return on security j for event day t, and R mt is the valueweighted index return on event day t. The coefficients α j and β j are the ordinary least squares estimates of the intercept and slope, respectively, of the market model regression. The cumulative abnormal return (CAR) from day T 2j Ν T 1j to day T 2j is defined as: CAR j = Σ AR j. The mean CAR is defined as: CAR= Σ CAR j / N for a sample of N t =T 1j j = 1 securities. Abnormal returns are estimated for each day in the event interval (days -30, +30) and for various identified intervals around the announcement (day 0). The total number of observation is 52 firms. Panel A: Mean AR, CAR and Number of Firms with Positive/Negative AR ± Days Mean AR (%) CAR (%) Positive : Negative : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :27 26

28 Panel B: AR is the abnormal return. Number positive/negative reports the number of observations with positive/negative AR at the specific date. The test of the null hypothesis that the abnormal return follows the work of Dodd and Warner (1983) and employs the z-statistic. The t-statistic the cross-sectional standard deviation test t, similar to the one used by Brown and Warner (1985); Generalized sign test indicating whether the proportion of positive to negative abnormal returns is significantly different from one. Day AR Number positive/negative Z - statistic T - statistic Generalized sign Z % 26: % 20: % 22: % 29: ** 1.829* 1.451* % 21: % 23: % 32: * % 21: * % 32: * % 33: *** 1.303* 2.564** % 45: *** *** 5.905*** Panel C: Cumulative Average Abnormal Return for Different Event Windows CAAR is the cumulative average abnormal return. Number positive/negative reports the number of observations with positive/negative CAAR at the specific window. The test of the null hypothesis that the abnormal return follows the work of Dodd and Warner (1983) and employs the z-statistic. The t-statistic the cross-sectional standard deviation test t, similar to the one used by Brown and Warner (1985); Generalized sign test indicating whether the proportion of positive to negative abnormal returns is significantly different from one. Interval (Day) CAAR Number positive/negative Z - statistic T - statistic Generalized sign Z (-10, -2) 1.14% 28: (-1, 0) 14.15% 46: *** *** 183*** (0, 0) 13.39% 45: *** *** 905*** ***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively using a 1-tail test. 27

29 j j Table 3: Event Study Results for Acquiring Firms For each security j, the market model is used to calculate an abnormal return (AR) for event day t as follows: AR jt = R jt - ( a + β Rmt ) where Rjt is the rate of return on security j for event day t, and R mt is the valueweighted index return on event day t. The coefficients α j and β j are the ordinary least squares estimates of the intercept and slope, respectively, of the market model regression. The cumulative abnormal return (CAR) from day T 2j Ν T 1j to day T 2j is defined as: CAR j = Σ AR j. The mean CAR is defined as: CAR= Σ CAR j / N for a sample of N t =T 1j j = 1 securities. Abnormal returns are estimated for each day in the event interval (days -30, +30) and for various identified intervals around the announcement (day 0). The total number of observation is 78 firms. Panel A: Mean AR, CAR and Number of Firms with Positive/Negative AR ± Days Mean AR (%) CAR (%) Positive : Negative % -0.09% 33: % 0.78% 49: % 1.22% 38: % 0.57% 31: % 0.74% 42: % 0.55% 30: % 0.55% 32: % 0.85% 33: % 0.95% 36: % 0.93% 36: % 0.67% 33: % 0.12% 29: % 0.53% 36: % 0.56% 35: % 0.47% 43: % 0.36% 31: % -0.03% 33: % -0.14% 37: % -0.08% 37: % 0.65% 39: % 0.40% 32: % 0.84% 33: % 0.51% 38: % 0.35% 36: % 0.40% 38: % 0.19% 32: % 0.12% 40: % -0.40% 35: % -0.62% 35: % -0.57% 40: % -1.24% 31: % -1.25% 38: % -1.34% 34: % -1.52% 33: % -1.39% 36: % -1.40% 37: % -1.88% 33: % -2.04% 42: % -1.75% 40: % -1.57% 37: % -1.41% 32:46 28

30 Panel B: AR is the abnormal return. Number positive/negative reports the number of observations with positive/negative AR at the specific date. The test of the null hypothesis that the abnormal return follows the work of Dodd and Warner (1983) and employs the z-statistic. The t-statistic the cross-sectional standard deviation test t, similar to the one used by Brown and Warner (1985); Generalized sign test indicating whether the proportion of positive to negative abnormal returns is significantly different from one. Generalized Day AR Positive/negative Z - statistic T statistic sign Z % 32: % 33: % 38: % 36: % 38: % 32: % 40: % 35: % 35: % 40: % 31: *** * * Panel C: Cumulative Average Abnormal Return for Different Event Windows CAAR is the cumulative average abnormal return. Number positive/negative reports the number of observations with positive/negative CAAR at the specific window. The test of the null hypothesis that the abnormal return follows the work of Dodd and Warner (1983) and employs the z-statistic. The t-statistic the cross-sectional standard deviation test t, similar to the one used by Brown and Warner (1985); Generalized sign test indicating whether the proportion of positive to negative abnormal returns is significantly different from one. Interval (Day) AR CAAR Positive/ negative Z - statistic T statistic Generalized sign Z (-10, -2) -1.28% -0.75% 33: (-1, 0) -0.62% -1.28% 26: ** ** (0, 0) -0.67% -1.74% 31: *** * * ***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively using a 1-tail test. 29

31 j j Table 4: Event Study Results for Matched-Pair Target Firms For each security j, the market model is used to calculate an abnormal return (AR) for event day t as follows: AR jt = R jt - ( a + β Rmt ) where Rjt is the rate of return on security j for event day t, and R mt is the valueweighted index return on event day t. The coefficients α j and β j are the ordinary least squares estimates of the intercept and slope, respectively, of the market model regression. The cumulative abnormal return (CAR) from day T 2j Ν T 1j to day T 2j is defined as: CAR j = Σ AR j. The mean CAR is defined as: CAR= Σ CAR j / N for a sample of N t =T 1j j = 1 securities. Abnormal returns are estimated for each day in the event interval (days -30, +30) and for various identified intervals around the announcement (day 0). The total number of observation is 41 firms. Panel A: Mean AR, CAR and Number of Firms with Positive/Negative AR ± Days Mean AR CAR Positive : Negative % -0.33% 22: % -0.91% 20: % -0.60% 17: % -0.69% 24: % -1.32% 15: % -1.78% 21: % -2.57% 13: % -2.93% 17: % -2.49% 18: % -2.94% 20: % -2.10% 26: % -3.07% 18: % -2.15% 20: % -0.44% 21: % -0.97% 17: % -0.65% 17: % -0.70% 20: % -1.22% 13: % -0.53% 26: % 0.78% 22: % 1.56% 23: % 1.29% 17: % 0.70% 19: % 1.83% 23: % 2.09% 19: % 1.91% 17: % 2.15% 25: % 2.24% 17: % 2.81% 25: % 3.37% 27: % 16.75% 36: % 19.77% 30: % 19.46% 21: % 18.62% 18: % 20.23% 22: % 19.64% 18: % 19.28% 15: % 19.61% 19: % 19.96% 21: % 19.21% 14: % 18.87% 19:22 30

32 Panel B: Cumulative Average Abnormal Return for Different Event Windows CAAR is the cumulative average abnormal return. Number positive/negative reports the number of observations with positive/negative CAAR at the specific window. The test of the null hypothesis that the abnormal return follows the work of Dodd and Warner (1983) and employs the z-statistic. The t-statistic the cross-sectional standard deviation test t, similar to the one used by Brown and Warner (1985); Generalized sign test indicating whether the proportion of positive to negative abnormal returns is significantly different from one. Interval (Day) N CAAR Positive/ negative Z - statistic T statistic Generalized sign Z (-30, -2) % 24: * (-1, 0) % 37: *** *** 5.694*** (+1, +30) % 28: ** 1.622* 2.873** ***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively using a 1-tail test. 31

33 j j Table 5: Event Study Results for Matched-Pair Acquiring Firms For each security j, the market model is used to calculate an abnormal return (AR) for event day t as follows: AR jt = R jt - ( a + β Rmt ) where Rjt is the rate of return on security j for event day t, and R mt is the valueweighted index return on event day t. The coefficients α j and β j are the ordinary least squares estimates of the intercept and slope, respectively, of the market model regression. The cumulative abnormal return (CAR) from day T 2j Ν T 1j to day T 2j is defined as: CAR j = Σ AR j. The mean CAR is defined as: CAR= Σ CAR j / N for a sample of N t =T 1j j = 1 securities. Abnormal returns are estimated for each day in the event interval (days -30, +30) and for various identified intervals around the announcement (day 0). The total number of observation is 41 firms. Panel A: Mean AR, CAR and Number of Firms with Positive/Negative AR ± Days Mean AR CAR Positive : Negative % -0.11% 20: % 0.58% 30: % 0.75% 18: % 0.43% 17: % 0.35% 19: % 0.33% 16: % 0.38% 15: % 0.49% 15: % 0.04% 16: % -0.10% 19: % -0.39% 17: % -0.46% 16: % -0.25% 21: % 0.49% 22: % 0.55% 21: % 0.19% 14: % -0.52% 14: % -0.30% 19: % 0.31% 24: % 0.81% 20: % 0.74% 18: % 0.57% 15: % 0.70% 24: % 0.85% 20: % 0.96% 23: % 1.08% 20: % 0.72% 19: % 0.19% 14: % -0.26% 16: % -0.22% 22: % -2.22% 12: % -2.23% 21: % -2.30% 19: % -2.35% 21: % -2.48% 20: % -2.42% 20: % -2.57% 18: % -2.46% 25: % -1.95% 22: % -2.31% 17: % -1.90% 21:20 32

34 Panel B: Cumulative Average Abnormal Return for Different Event Windows CAAR is the cumulative average abnormal return. Number positive/negative reports the number of observations with positive/negative CAAR at the specific window. The test of the null hypothesis that the abnormal return follows the work of Dodd and Warner (1983) and employs the z-statistic. The t-statistic the cross-sectional standard deviation test t, similar to the one used by Brown and Warner (1985); Generalized sign test indicating whether the proportion of positive to negative abnormal returns is significantly different from one. Interval (Day) N CAAR Positive/ negative Z - statistic T - statistic Generalized sign Z (-30, -2) % 17: (-1, 0) % 7: *** *** *** (+1, +30) % 24: * ***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively using a 1-tail test. 33

35 Table 6 Cross-patterns of CARs for match-pair sub-sample of 41 transacting firms by differing intervals. (Target sign, Acquirer sign) Interval (days) ( +, + ) (, ) ( +, ) (, + ) -30 to to to to

36 Table 7 Regression Results for Acquirer Firms in 41 matched pair transactions Acquisition transactions with ADRs as a form of payment. OLS regression Coefficient (p-value) Developing or non-developing economy (0.183) Traditional acquisition or merger 0.124*** (0.006) Percent ownership interest in target (0.936) Percent of target equity not acquired 0.144*** (0.003) Transaction size (0.652) Same SIC between acquirer and target * (0.058) Intercept (0.326) Number of observations 41 Adjusted R F-statistic ***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively using a 1-tail test. 35

37 Figure I Average target CARs from -30 days through +10 days of the transaction announcement. N = 52 target firms CAR (%) No. of days plus or minus announcement date 36

38 Figure II Average Acquirer CARs from -30 days through +10 days of the transaction announcement. N = 52 acquirer firms CAR (%) No. of days plus or minus announcement date 37

39 References: Akbulent & Matsusaka, 2003, 50 Years of Diversification Announcements, Working Paper, University of Southern California Asquith, P., R. Brunner, and, D. Mullins, 1987, Merger returns and the form of financing, Proceedings of the Seminar on the Analysis of Security Prices 34, 1, Andrade, Gregor, Mark Mitchell and Erik Stafford, 2001, New evidence and perspectives on mergers, Journal of Economic Perspectives I5, 2 Berger P.G., and E. Ofek, 1995, Diversification s effect on firm value, Journal of Financial Economics 37, 1, Jan, Bhagat, S., A. Shleifer, and R.W. Vishny, 1990, Hostile takeovers in the 1980s: The return to corporate specialization, Brookings Papers on Economic Activity, Special Issue, Bharadwaj, A. and A. Shivdansi, 2003, Valuation effects of bank financing in acquisitions, Journal of Financial Economics 67, Bradley, M., A. Desai, and E. Kim, 1988, "Synergistic Gains from Corporate Acquisitions and their Division between the Stockholders of Target and Acquiring Finns," Journal of Financial Economics 21, Bruner, R.F Does M&A Pay? A review of the evidence for the decision-maker, Journal of Applied Finance (Spring/Summer 2002) 12: Campa, J. M., and S. Kedia, 2002, Explaining the diversification discount, The Journal of Finance 57, Dennis, Debra K. and McConnell, J., Corporate Mergers and Security Returns, Journal of Financial Economics, 16, June 1986, Dodd, P. and J.B. Warner, 1983, On corporate governance a study of proxy contests, Journal of Financial Economics 11, Doukas, J. and N.G. Travlos, 1988, The effect of corporate multi-nationalism on shareholder wealth: evidence from international acquisitions, The Journal of Finance 43, Forester. R and Karolyi. A; "The E ect of Market Segmentation and Investor Recognition on Asset Prices: Evidence from Foreign Sticks Listing in the United States"; Journal of Finance, 1999, p Heron, R., and E. Lie, 2002, Operating performance and the method of payment in 38

40 takeovers, Journal of Financial and Quantitative Analysis 37, Huang, Y, and R Walkling, 1987, Target abnormal returns associated with acquisition announcements: payment, acquisition form, and managerial resistance, Journal of Financial Economics, 19, Jarrel, G., J. Brickley, and J. Netter, 1988, The market for corporate control: The empirical evidence since 1980, Journal of Economic Perspectives 2, 2, Jensen, M., 1986, Agency costs of free cash flow, corporate finance, and takeovers, American Economic Review 76, 2, Jensen, M.C., and W.H. Meckling, 1976, Theory of the firm: managerial behavior, agency costs and ownership structure, Journal of Financial Economics 3, 4, Jensen, M., and R. Ruback, 1983, The market for corporate control: The scientific evidence, Journal of Financial Economics 11, 1-4, Kale, J., O. Kini, and H.E. Ryan, Financial advisors and shareholder wealth gains in corporate takeovers, Journal of Financial and Quantitative Analysis 38, Kang, J. 1993, The international market for corporate control: Mergers and acquisitions of US firms by Japanese firms, Journal of Financial Economics, vol. 34, pp Kaplan, S., and M. Weisback, 1992, The success of acquisitions: Evidence from divestitures, The Journal of Finance 47, 1, Karoli., G. A., 1998, Why Do Companies List Shares Abroad?: A Survey of the Evidence and Its Managerial Implications, Financial Markets, Institutions & Instruments 7, 1, Lang, L. and R. Stulz, 1994, Tobin s Q, corporate diversification and the firm performance, Journal of Political Economy 102, Lichtenberg, F.R., 1992, Industrial De-diversification and its Consequences for Productivity, Journal of Economic Behavior and Organization 18, Lin, J.W., J. Madura, and A. Picou, 1994, The wealth effects of international acquisitions and the impact of the EEC integration, Global Finance Journal 5, 1, Maksimovic, V., and G. Phillips, 2002, Do conglomerate firms allocate resources inefficiently across industries? Theory and evidence, The Journal of Finance, 57, 2, Muscarella, C., and M. vetsuypens, 1996, Stock Splits: Signaling or Liquidity? The Case 39

41 of ADR Solo-Splits, Journmal of Financial Economics 42, Nanda, S., J.E. Owers and C. Feng, 1996, Institutional and investment performance attributes of American Depositary Receipts, Empirical Issues in Raising Equity Capital (Elsevier Science B.V., North Holland, Amsterdam). Nanda, S., C. Feng and J.E. Owers, 2000, The International Attributes and Return Performance of Newly-Listed ADRs, International Journal of Business, 5, 1, 1-25 Roll, R., 1986, The hubris hypothesis of corporate takeovers, Journal of Business, 59, Scharfstein, D., and J. Stein, 2000, The dark side of internal capital markets: Divisional rent-seeking and inefficient investment, The Journal of Finance, 55, 6, Schleifer, A. and R.W. Vishny, 2003, Stock market driven acquisitions, Journal of Financial Economics, 70, Schoar, A., 2002, Effects of corporate diversification on productivity, The Journal of Finance 57, 6, Stein, Jeremy, 1997, Internal Capital Markets and the Competition for Corporate Resources, The Journal of Finance 52, , Travols, N., 1987, Corporate Takeover Bids, Method of Payment, and Bidding Firms' Stock Returns", The Journal of Finance, September 1987, pp Villalonga, B., 2004, Does Diversification Cause the Diversification Discount?, Financial Management, 33, Weston, J.F., 1970, The nature and significance of conglomerate firms, St. John s LawReview 44,

42 Founded in 1892, the University of Rhode Island is one of eight land, urban, and sea grant universities in the United States. The 1,200-acre rural campus is less than ten miles from Narragansett Bay and highlights its traditions of natural resource, marine and urban related research. There are over 14,000 undergraduate and graduate students enrolled in seven degreegranting colleges representing 48 states and the District of Columbia. More than 500 international students represent 59 different countries. Eighteen percent of the freshman class graduated in the top ten percent of their high school classes. The teaching and research faculty numbers over 600 and the University offers 101 undergraduate programs and 86 advanced degree programs. URI students have received Rhodes, Fulbright, Truman, Goldwater, and Udall scholarships. There are over 80,000 active alumnae. The University of Rhode Island started to offer undergraduate business administration courses in In 1962, the MBA program was introduced and the PhD program began in the mid 1980s. The College of Business Administration is accredited by The AACSB International - The Association to Advance Collegiate Schools of Business in The College of Business enrolls over 1400 undergraduate students and more than 300 graduate students. Mission Our responsibility is to provide strong academic programs that instill excellence, confidence and strong leadership skills in our graduates. Our aim is to (1) promote critical and independent thinking, (2) foster personal responsibility and (3) develop students whose performance and commitment mark them as leaders contributing to the business community and society. The College will serve as a center for business scholarship, creative research and outreach activities to the citizens and institutions of the State of Rhode Island as well as the regional, national and international communities. The creation of this working paper series has been funded by an endowment established by William A. Orme, URI College of Business Administration, Class of 1949 and former head of the General Electric Foundation. This working paper series is intended to permit faculty members to obtain feedback on research activities before the research is submitted to academic and professional journals and professional associations for presentations. An award is presented annually for the most outstanding paper submitted. Ballentine Hall Quadrangle Univ. of Rhode Island Kingston, Rhode Island