Revisiting Shareholder Value Creation via International Joint Ventures: Examining Interactions Among Firm- and Context-Specific Variables

Canadian Journal of Administrative Sciences, Jun 2004 by Merchant, Hemant

Following Garcia-Canal (1996), this study defined task complexity (COMPLEX) as the number of distinct types of value chain activities (R&D, manufacturing, marketing) undertaken simultaneously. Thus, COMPLEX assumed a minimum value of one and a maximum value of three. Data for coding COMPLEX were obtained from Dow Jones, which explicitly identified all value chain activities that would be undertaken by an IJV. The reported information was of a factual nature, and so could be unambiguously coded.

Per Das et al. (1998) and many others, this study measured firm size (FIRMSIZE) in terms of logarithm of American parents' market value over the two-day announcement period. An American firm's market value was defined as the product of its stock price times the number of outstanding shares. The data needed to operationalize this variable were obtained from Center for Research on security Prices tapes compiled by the University of Chicago.

Following Lummer and McConnell (1990), this study defined partner type (PARTNER) as a dummy variable, such that PARTNER = 1 if the non-American partner was a firm; PARTNER = O if the non-American partner was a state-owned enterprise. The data needed to categorize the type of non-American partner were obtained from the Dow Jones text, which contained information about the firms' IJV formations. Usually half a single-spaced page, this text almost always explicitly noted if the non-American partner was a government-owned entity.

Per Bleeke and Ernst (1991), the study coded American parents' relative competitive position (COMPPOS) on a three-point scale (1 = Strong to 3 = Weak). This coding involved content analysis (Weber, 1990) of competitive information obtained from Value Line for the quarter immediately preceding that in which a firm announced its IJV participation. The 400-word analyst report for each firm was analyzed for negative and positive indicators of competitive pressures. In most cases, the tone and content of the reports were more or less consistent, so it was not too difficult to assess the magnitude of competitive pressures. In cases where the tone and content were of a mixed nature, the observation was dropped from analysis.

This study measured ownership position (OWNERSHIP) as the percentage of an IJVs total equity held by the American parent. The data needed to code this variable were obtained from the Dow Jones text that contained information about equity distribution between partners. These data were reported very frequently. The study assumed a 50-50 equity ownership between partners when the ownership percentages were not reported. This protocol agrees with reports that a 50-50 ownership position is the most dominant form of equity ownership in IJVs (e.g., Blodgett, 1991).

As done by Saxton (1997), this study defined a venture's decision-making structure as a dichotomous variable so that DECISION = 1 if both partners shared the decision-making process; DECISION - 0 otherwise. Information needed to code this variable was reported in the text of IJV formation announcement in Dow Jones. In most cases, the text explicitly noted if only one partner was responsible for decision-making; this reporting practice was observed to be consistent across the sampled years. This study discarded observations with ambiguous information pertaining to the above variable.