Printer Friendly

Ownership strategy for a foreign affiliate: an empirical investigation of Japanese firms

As a multinational corporation (MNC) establishes its affiliate in a foreign market, it has two options in determining the affiliate's ownership structure; full ownership (wholly-owned subsidiary) and shared ownership (equity joint venture). Generally, the choice of ownership structure for an affiliate depends on the MNC's strategy and on the costs of alternative ways of implementing the strategy (Gomes-Casseres 1989). Since the choice usually involves a two step process, namely, the determination of the MNC's preference, and the negotiation process with a suitable host government, the realized outcome depends on the interaction between these two factors (Gomes-Casseres 1989, 1990).

Transaction cost arguments have been successfully applied in previous studies investigating the MNC preference for ownership structure for its foreign affiliate (Anderson and Gatignon 1986, Gatignon and Anderson 1988, Hennart 1988, 1991, Kogut and Singh 1988a, 1988b, Kim and Hwang 1992). They posit that the choice between full and shared ownership depends on the relative costs and benefits of the two alternative ownership structures. However, though most transaction cost-based studies recognize the impact of host government restrictions on such choices, they have rarely investigated the impact explicitly in an empirical setting. In reality, such restrictions can induce an MNC to select a shared ownership structure in instance where transaction cost analysis would suggest a full ownership structure. Gomes-Casseres (1989, 1990) has complemented the transaction cost-based studies by empirically investigating the impact of host government restrictions with a model that integrates both the transaction cost and the bargaining power approaches.

The present study builds upon prior research by examining the choice of foreign ownership structure by Japanese manufacturing MNCs in 36 foreign countries over a period of 23 years. This study offers the first large sample study of the factors which affect the choice of foreign ownership structure by non-U.S. MNCs in a global setting, unlike previous studies which have investigated the ownership policies of U.S. MNCs in foreign markets (Gatignon and Anderson 1988, Gomes-Casseres 1989, 1990, Kim and Hwang 1992) or those of foreign (non-U.S.) MNCs in the U.S. (Kogut and Singh 1988a, Hennart 1991).

This study also represents the first attempt to test whether cultural differences influence the choice of foreign ownership structure by MNCs from a single home country. Cultural variables have recently been shown to be important in the foreign establishment mode decisions by firms (Kogut and Singh 1988b, Cho and Padmanabhan 1992).

Finally, this study incorporates both investing firm-related and host country-related factors, for example, host government policy regarding the ownership structures and the degree of cultural similarity between the home and host countries, to determine what influence (if any) these variables have on the choice of ownership structure by MNCs. Conceptually, by attempting to incorporate variations among host countries, but not variations across home countries, we hope to better understand the factors that determine the MNC's choice of ownership structure for foreign affiliates.

The next section reviews the relevant literature on the choice of MNC's foreign ownership structure. This is followed by a presentation of hypotheses to be tested. The data, methodology, and variables used in the study are described in detail, followed by a discussion of the empirical results and concluding remarks.

Full Ownership versus Shared Ownership

The choice between the two alternative ownership structures basically involves trade-offs related to the level of resource commitment (Stopford and Wells 1972), the degree of control (Caves 1982, Davidson 1982, Root 1987), the specification and assumption of risks and returns, and the degree of global rationalization (Franko 1971, Stopford and Wells 1972, Hladik 1985, Kim and Hwang 1992). However, the choice is also influenced by the needs and relative bargaining power of host governments (Farge and Wells 1982, Grieco 1982, Lecraw 1984, Gomes-Casseres 1989, 1990). A number of investing firm and host country characteristics are suggested and/or identified as important determinants of the choice within the framework of the transaction cost theory, the eclectic model of international production, and/or the bargaining power model. Recent works by Anderson and Gatignon (1986), Gatignon and Anderson (1988), and Hennart (1991) provide excellent reviews of ownership choice explanations based on the transaction cost theory.

Firm-related Determinants

It is reasoned that a larger investing firm is more likely to possess the necessary financial resources for full ownership of its foreign operations and is better positioned for a (more resource-demanding) full ownership structure than a smaller firm. Analogously, a foreign operation that requires larger resources (relative to the resource availability of the parent) is more likely to be structured as jointly-owned. Empirical evidence partially supports this contention. For instance, Kogut and Singh (1988b) found that the size of foreign affiliate was positively and significantly related to shared ownership of foreign affiliates. However, Kogut and Singh (1988b) and Hennart (1991) found that the size of investing firm was not significantly related to full ownership structure for foreign firms investing in the U.S.

A firm entering into a product area not produced by the parent may find that the necessary product-specific capability (such as technology, manufacturing know-how, distribution, etc.) is possessed by another firm (and/or is costly to create internally), and is therefore most efficiently obtained through a joint venture (Stopford and Wells 1972, Hennart 1991). Stopford and Wells (1972) and Hennart (1991) found that foreign affiliates that manufactured products different from those of the parent were more likely to be joint ventures. However, Gomes-Casseres (1989) found that the relatedness of foreign affiliate's products to those of the parent (he used this as a proxy for experience) had no effect on the probability of establishing a joint venture.

The extent to which international business/host country experience influences the ownership choices is not well established. On the one hand, a firm with greater international/host country experience is more likely to be able to bear the risks and management responsibility associated with full ownership of foreign operations and, thus, may find it less compelling to form a joint venture to share the risks. On the other hand, greater international business/host country experience may enable the firm to effectively deal with the costs and uncertainty associated with accepting equity partners and to become more willing to choose shared ownership. Empirical evidence in the literature seems to support the former scenario. For example, Gatignon and Anderson (1988) found that experienced MNCs were less likely to establish joint ventures in foreign markets. Gomes-Casseres (1989, 1990) and Hennart (1991) confirmed that familiarity with host country was more likely to lead MNCs to the full ownership of their foreign affiliates.

A firm with high R & D intensity may prefer to have complete control over its proprietary know-how in order to preserve and/or best exploit the know-how, given imperfections in the external markets for technology (e.g., Buckley and Casson 1976, Rugman 1981, Caves 1982). Thus, the higher the parent R & D intensity, the greater the possibility that the foreign affiliate will be fully-owned (Stopford and Wells 1972, Davidson 1982). Stopford and Wells (1972) found that R & D intensive American firms tended to fully own their foreign affiliates.

The relationship between the ownership structure and the establishment mode (acquisition or new venture) for a foreign affiliate is not well established. Though they are generally regarded as being independent of each other (e.g., Caves and Mehra 1986, Hennart and Park 1991, 1993), Kogut and Singh (1988b) nonetheless argue that the degree of ownership (joint venture) is usually determined in conjunction with the mode of establishment. However, empirical evidence supporting this view is sparse. Stopford and Wells (1972) confirmed the existence of a weak relationship between shared ownership and acquisition, whereas Hennart and Park (1991, 1993) found no evidence in support of any relationship between the degree of ownership and the mode of establishment. Since we believe that the choice of ownership level and the choice of establishment mode are basically two separate but sequential decisions facing firms entering into foreign markets, we expect that the two decisions will be independent of each other.

Host Country-related Determinants

The host attitude towards a particular ownership structure may affect the firm's choice between the two alternative structures. Some host countries prohibit majority foreign ownership of local operations possibly out of concern over loss of national control and the resultant competitive ramifications associated with majority foreign ownership, while others require prior governmental approval for establishing such ownership structures. Still, other host countries take neutral positions regarding the foreign ownership structure in most local industries, except for "strategic" ones. The influence of this prohibition on the choice is self-evident. Regarding the host requirement for prior governmental approval, one can expect that such requirements discourage full foreign ownership of local operations. Gomes-Casseres (1989, 1990) found that restrictive host policies strongly encouraged U.S. firms to establish joint ventures in such countries.

Cultural similarity between the home and host country seems to affect the choice of ownership structure (Kogut and Singh 1988b). The cultural similarity influences the perception of investing firms regarding the costs and uncertainty of alternative ownership structures in foreign markets. Generally, the costs and uncertainty associated with working with equity partners in jointly-owned affiliates would be greater in culturally dissimilar host countries than culturally similar ones and, thus, investing firms would prefer full ownership in culturally dissimilar countries. Alternatively, investing firms may need complementary knowledge and experience to operate successfully in culturally dissimilar countries and frequently joint venture partners can provide such complementary knowledge.(1) In such instances, investing firms may prefer shared ownership to full ownership. Though Kogut and Singh (1988b) found cultural similarity to be a significant factor increasing the probability for foreign firms to choose joint ventures over acquisitions for their operations in the U.S., they did not separate fully-owned acquisitions from partially-owned acquisitions. From a Japanese perspective, entry into culturally dissimilar countries may motivate the Japanese firms to establish fully-owned affiliates to allow easy application of organizational routines developed at home (thus, avoiding the costs and uncertainty involved in assimilating equity partners to their routines) (Abegglen and Stalk 1985, Rapp 1993). We expect that Japanese firms are more likely to choose a full ownership structure when they enter into culturally distant foreign markets than culturally similar ones.

Hypotheses

This study examines the effects of firm-related and host country-related factors on Japanese manufacturing MNCs' choice of foreign ownership structure. We focus on Japanese manufacturing MNC's experience mainly due to their growing importance in global markets, the paucity of studies on their foreign ownership structures, and the availability of relatively consistent and detailed information on their experience in a variety of host countries over a long period of time.

Based on the discussion in the previous section, we propose the following working hypotheses to be tested.

Hypothesis 1a: The larger an investing firm, the more likely it will choose a full ownership structure for its foreign affiliate.

Hypothesis 1b: The larger a foreign affiliate (in relation to the size of the investing firm), the more likely the foreign affiliate will be jointly owned.

Hypothesis 2: An investing firm will be less likely to choose a full ownership structure for its foreign affiliate when diversifying into areas outside of its core business.

Hypothesis 3a: The more experienced in international business an investing firm, the more likely it will choose a full ownership structure for its foreign affiliate.

Hypothesis 3b: The more familiar with a host country an investing firm, the more likely it will choose a full ownership structure for its affiliate in the host country.

Hypothesis 4: The more R & D intensive an investing firm, the more likely it will choose a full ownership structure for its foreign affiliate.

Hypothesis 5: There will be no significant relationship between an investing firm's ownership level and establishment mode of its foreign affiliate.

Hypothesis 6: An investing firm will be less likely to choose a full ownership structure for its foreign operation in a host country that requires prior governmental approval for full ownership of the operation.

Hypothesis 7: An investing firm will be more likely to choose a full ownership structure for its foreign operation in a culturally distant host country than a culturally similar one.

Tests of the Hypotheses

Methodology and Dependent Variable

The database for Japanese foreign direct investments (FDIs) was compiled from Japanese Overseas Investment: A Complete Listing by Firms and Countries, 1992/1993 (Toyo Keizai Shinposha 1992). The survey covers virtually all FDIs by Japanese firms listed on the Japanese stock exchanges (Tokyo, Osaka and Nagoya) as well as major unlisted Japanese firms. The data used in this study include only manufacturing FDIs with at least 10 percent ownership by a Japanese parent listed in Japan Company Handbook. Japan Company Handbook is one of the most comprehensive sources of information on Japanese firms listed on the first section of the Japanese stock exchanges. The database consists of 1,519 uniquely identifiable cases of such FDIs by 402 Japanese firms in 45 countries between 1969 and 1991.

Following previous studies on foreign ownership choice, this study uses the 95 percent cutoff point to capture the alternative ownership structures (Franko 1971, Gatignon and Anderson 1988, Gomes-Casseres 1989, 1990, Hennart 1991).(2) A Japanese parent's ownership share over 95 percent of equity is defined as full ownership, whereas ownership share in the 10 to 95 percent range (inclusive) represents shared ownership. The database lists 546 cases of full ownership affiliates (55 percent) and 447 cases of shared ownership affiliates (45 percent) in the manufacturing sector with at least 10 percent ownership by a single-parent Japanese manufacturing firm. Distribution of the alternative ownership modes across host countries is reported in Appendix 1.

Since some host countries legally prohibited full foreign ownership of local firms, Japanese investments in such countries (e.g., India, Pakistan, Sri Lanka, and China) are excluded (Business International (various years), Price Waterhouse (various years), United Nations Centre on Transnational Corporations 1978-1985, U.S. Department of Commerce (various years) a, b, and c)). This and missing values on some independent variables used in the logit regression reduced the original database to 839 observations, of which 472 (56 percent) are full ownership affiliates, and is fairly representative of the original database. Finally, a note of caution in defining the boundaries of Japanese firms which may belong to industrial groups, or keiretsu. Group membership is found to have significant influence on the strategy, behavior, and performance of a Japanese firm (Genay 1991). While data are not available to enable us to control for the influence of group affiliation, we recognize this possibility and the resulting limitation in our operationalization and measurement of investing firm-related factors to be discussed later.

The ownership structure is captured by a dummy variable which takes a value of one for full ownership and zero for shared ownership. Because of the nature of the dependent variable, this study uses logit analysis to test the hypotheses developed in the previous section. This technique is well-suited for the case in hand, where the phenomenon to be explained can only be measured as [0/1] alternatives.(3) Logit analysis determines the probability (P) of full ownership as a function of a set of independent variables. A positive sign for the coefficient implies that the variable increases the likelihood of full ownership. The model can be expressed as

P([y.sub.i] = 1) = 1/[1 + exp(-a - b[x.sub.i])],

where [y.sub.i] is the dependent variable, [x.sub.i] is the vector of independent variables for the ith observation, a is the intercept parameter, and b is the vector of regression coefficients (Altman et al. 1981).

Independent Variables

The size of the investing firm (ASIZE) is measured by the parent's total global assets at the time (year) of foreign entry. This variable is stated in millions of Japanese yen and introduced in logarithmic form, since the size of assets is expected to have a decreasing impact on the parent's stock of marketable assets or borrowing capacity to finance its foreign affiliate.(4) The source of data is Japan Company Handbook. A positive sign for the coefficient of the variable is predicted.

The size of the foreign operation relative to the parent (RSIZE) is measured by the ratio of the investment size of foreign operation to the parent's total assets at the time (year) of foreign entry. The sources of data are Toyo Keizai Shinposha (1992) and Japan Company Handbook. A negative sign for the coefficient of the variable is predicted.

Relatedness of investment (RLTDNS) is captured by a dummy variable equal to zero if none of the products to be produced by the foreign affiliate was in the same industry group (or a closely related industry group) as the parent's prime product lines (unrelated investment) and one otherwise (related investment), based on product descriptions given in Toyo Keizai Shinposha (1992) and Japan Company Handbook.(5) A positive sign for the coefficient of the variable is predicted.

International business experience (IBEXP) is measured by the length of time (in years) that the parent had been operating abroad at the time (year) of foreign entry (the year of foreign entry minus the year of the oldest foreign investment by the parent as listed in Toyo Keizai Shinposha (1992)). This variable is introduced in logarithmic form since the length of time is expected to have a decreasing impact on the parent's stock of international business experience and knowledge. Consequently, for computational reasons, a value of one year is allowed when the parent's international business experience is less than one year. The source of data is Toyo Keizai Shinposha (1992). A positive sign for the coefficient of the variable is predicted.

Familiarity with a host country (HCEXP) is measured in a similar fashion as IBEXP; the length of time (in years) that the parent had been operating in the host county at the time (year) of entry. The source of data is Toyo Keizai Shinposha (1992). A positive sign for the coefficient of the variable is predicted.

R & D intensity (RND) is measured by the ratio of R & D expenditure to sales at the time (year) of foreign entry, supplemented by the 1986 R & D to sales ratio for those pre-1986 cases where data are not available. The source of data is Japan Company Handbook. A positive sign for the coefficient of the variable is predicted.

The establishment mode (EM) is captured by a dummy variable which assumes a value of one for acquisitions and zero for new ventures. An acquisition is defined as the purchase of ownership in a pre-existing local firm in an amount sufficient to confer effective control. A new venture refers to a start-up investment in new facilities. The source of data is Toyo Keizai Shinposha (1992). The coefficient estimate is predicted not to be significant.

Host policy (HP) is captured by a dummy variable. A value of one is given for those host countries requiring prior governmental approval for full foreign ownership of local operations at the time (year) of foreign entry, and zero for others. HP distinguishes between the two types (restrictive vs. non-restrictive) of host ownership policy regimens and captures any changes in a host country's policy regime throughout the study period. However, this dichotomous classification may not realistically capture the influence of a plethora of exceptions and other complex manifestations of host government regulations on ownership structure. Lacking precise data on such regulations, we measure the restrictiveness of host regulations in a conservative manner, risking the possibility of exaggerating the restrictiveness. The sources of data are Business International (various years), Price Waterhouse (various years), United Nations Centre on Transnational Corporations (1975-1985), and U.S. Department of Commerce (various years a, b, and c). A negative sign for the coefficient of the variable is predicted.

Cultural distance (CD) is measured by a composite index showing the overall cultural distance of each host country from Japan using Hofstede's four indices (Hofstede 1980, Hofstede and Bond 1988) and the methodology developed by Kogut and Singh (1988 b).(6) For countries in our sample for which indices are not reported by Hofstede (Luxembourg and Iceland), we use indices of countries that are judged to be culturally similar (Erramilli 1991). For example, we substitute indices of Belgium for Luxembourg. Furthermore, those host countries whose cultural distance score (CD) is less than or equal to the mean (2.9899) of the 50 countries as reported in Hofstede (1980) and Hofstede and Bond (1988) are classified as countries which are culturally similar to Japan in this study. A list of culturally similar and dissimilar countries is reported in Appendix 1. A positive sign for the coefficient of the variable is predicted.

Finally, since this study covers a relatively long time period (23 years), we include a time dummy variable (TIME) in our logit regression to investigate any significant differences that might exist in the choice of ownership structure between earlier and more recent Japanese FDIs. Since the yen-dollar realignment induced by the G-7 Plaza Agreement was, among others, an important facilitating factor for recent Japanese FDIs (e.g., Wilkins 1990), we compare the key determinants of the Japanese MNC's choice before and after the Plaza Agreement. A value of zero is given for pre-1986 Japanese FDIs and one for post-1985 Japanese FDIs. No prediction is made regarding the sign for the coefficient.

Discussion of the Results of the Empirical Analysis

The LOGIST procedure in the SAS program package was used to estimate the parameters of the logit regression model (Harrell 1986). Tables 1 and 2 report results for the full sample and for selected subsamples, respectively. In particular, we analyse the data separately for culturally similar and dissimilar subsamples, for countries with and without overt restrictions on full ownership, and for cases prior to 1986 and after 1985.

Subsample results help to identify any interaction and/or nonlinear effects that may be present. For instance, certain variables may be more important discriminators of the ownership structure dependent variable when Japanese parent firms face overt restrictions on full ownership (HP = 1) than when they face no such restrictions (HP = 0). If firms face no governmental restrictions on ownership structure, they have greater freedom to select the ownership structure of their choice. On the other hand, if governments place overt restrictions on such choices, it is of interest to determine whether other independent variables become significant for this subsample. If none of the independent variables turn out to be significant for the restrictions subgroup, it implies that the governmental restrictions are operationally binding. If other variables are able to statistically discriminate between full and shared ownership, it suggests that selective firms are able to successfully negotiate away the restrictions, or that the overt restrictions are not binding. Similarly, different sets of independent variables may become significant discriminators [TABULAR DATA FOR TABLE 2 OMITTED] for the culturally similar subgroup (CD = 0), the culturally dissimilar subgroup (CD = 1), the pre-1986 subgroup (TIME=0), and the post-1985 subgroup (TIME = 1). For example, Japanese parent firms may have a greater tendency to opt for full ownership in culturally dissimilar countries in order to avoid greater costs and uncertainties associated with joint ownership in such countries, whereas they may be more willing to select a local partner for investments in culturally similar countries. Consequently, different sets of explanatory variables may help discriminate between full and shared ownership for the two subgroups. Finally, the TIME subgroup analysis may provide insights into changes involving Japanese ownership structure preferences over time. Hence, identifying subsample characteristics help provide interesting insights into Japanese parent firms' global ownership structure patterns.
Table 1. Logit Regression Results: Full Ownership versus Shared
Ownership (95 Percent Cutoff, Full Sample)


Variables Coefficients (standard error)


Intercept 0.0145 (0.9176)
ASIZE -0.1132 (0.0727)
RSIZE 0.6100 (0.4764)
RLTDNS 0.2223 (0.1958)
IBEXP 0.0976 (0.0972)
HCEXP 0.2125(***) (0.0654)
RND 11.1384(***) (3.3130)
EM -0.2376 (0.1812)
HP -1.0440(***) (0.1627)
CD 0.1651(**) (0.0813)
TIME 0.2219 (0.1970)


Model chi-square 86.83 (p = 0.0001)
N = 839 (F: 472, S: 367)
Correct Ratio (%) 65.1


* p[less than]0.10, ** p[less than]0.05, *** p[less than]0.01
(two tailed)


F: Full ownership, S: Shared ownership


The full sample logit regression (Table 1) has a significant overall explanatory power with a model chi-square of 86.83 (p = 0.001). In addition, approximately 65.1 percent of the sample observations are correctly classified whereas the baseline rate in our sample is 50.8 percent.(7) The subsample logit regressions also have significant explanatory power. From the full sample regression (Table 1), we find the coefficients for host policy (HP), familiarity with a host country (HCEXP), R & D intensity (RND), and cultural distance (CD) variables significant and to be of the correct sign, and the coefficients of other variables to be not statistically significant.(8)

We find the coefficient of HP, a dummy variable capturing the presence of host requirement for prior governmental approval for the full foreign ownership of local firms at the time of entry, significant at the 0.01 level with the negative sign as predicted. In the full sample regression (Table 1), HP has the most significant effect on the probability of a Japanese firm to fully own its foreign affiliate. This suggests that Japanese firms are less likely to fully own their affiliates in overtly restrictive host countries. The findings are consistent with those reported in Gomes-Casseres (1989, 1990). Host restrictions on foreign ownership share influence MNC's choice of ownership structure for its foreign affiliate. This is independent of the host country's cultural distance from Japan and the timing of investment, as evidenced by the results of the cultural and period subsample regressions (Table 2).

The coefficient of HCEXP, the extent of familiarity with a host country, is significant at the 0.01 level with the predicted positive sign in the full sample regression (Table 1). This indicates that Japanese firms are less likely to jointly own their foreign affiliates with local partners as their familiarity with host countries increases. These results are consistent with those reported in Gomes-Casseres (1989, 1990) and Hennart (1991): familiarity with host country increases the probability for an MNC to choose the full ownership structure for its foreign affiliate. However, IBEXP, the extent of international business experience of the parent, is found not to play a similar role (statistically not significant, but with the predicted positive sign).

A further investigation into the importance of HCEXP and IBEXP for subsamples reveals interesting results (Table 2). The coefficient of HCEXP becomes significant with the correct sign for the culturally similar country subsample, the non-restrictive host country subsample, and the post-1985 subsample. The coefficient of IBEXP is significant only with the culturally dissimilar country subsample. One possible explanation for this is that overall international business experience is more important when firms face relatively greater uncertainties associated with investments in culturally dissimilar host countries. Consequently, firms with higher levels of overall international business experience prefer to establish a full ownership structure in such countries. However, given lower levels of uncertainties in culturally similar host countries, in non-restrictive host countries, and during the later period, the ability of IBEXP to discriminate between full and shared ownership structure becomes weaker and the level of familiarity with a host country becomes a more important discriminator. Firms with higher levels of host country experience have a tendency towards full ownership in such countries.

The coefficient of RND, the parent's R & D expenditures to sales, is positive as predicted and significant at the 0.01 level in the full sample regression (Table 1). This suggests that a more R & D intensive Japanese firms prefers full ownership of its foreign affiliate in order to have complete control over its proprietary know-how and established organizational routines.(9) The results are consistent with those reported in Stopford and Wells (1972). This is particularly true for Japanese firms investing in both culturally similar and dissimilar host countries, in restrictive host countries, and over the entire period (Table 2): parent's R & D intensity significantly increases the probability of full ownership of its foreign affiliate. Facing uncertainties associated with restrictive host ownership policies, Japanese firms with higher levels of R & D prefer to establish complete control over their proprietary know-how through full ownership as a compensating factor, and hence the positive relationship. On the other hand, since the perceived threat of loss of proprietary know-how is lower in non-restrictive host countries, RND cannot explain ownership structure choice in such countries. An alternative explanation may be that in non-restrictive countries, as Hennart (1991, p. 494) reasoned in explaining the absence of significance of his R & D intensity variable in his study of ownership strategies of Japanese firms in the United States, the R & D intensive Japanese firms may have alternative motives of acquiring and exploiting technologies.(10) In this case, the high R & D intensity may lead them to both fully own (to exploit their technologies) and to joint venture their subsidiaries (to acquire technologies), which may rationalize the findings of no association between the R & D variable and the dependent variable. On the other hand, in restrictive host countries, given the lack of potential local partners with significant complementary assets to offer, the R & D intensive Japanese firms may not enter - or insist on full ownership - if they have significant proprietary technologies to exploit. Those firms that agree to share ownership will be those that are not technology intensive.(11) Finally, a third explanation may be that the majority of the firms investing in non-restrictive countries have high R & D proportions and hence this variable cannot successfully discriminate between alternate ownership choices.

We find the coefficient of CD, the cultural distance between Japan and the host country, to be positive as predicted and significant at the 0.05 level in the full sample regression (Table 1). This suggests that Japanese firms are more likely to fully own their affiliates in culturally distant host countries than culturally similar ones. One explanation is that they regard the costs and uncertainties associated with joint ownership of their affiliates with equity partners much greater in culturally dissimilar host countries than culturally similar ones. The subsample results further support this rationale. Cultural dissimilarity plays a particularly significant role in increasing the probability of full ownership of foreign affiliates in non-restrictive host countries (Table 2). Japanese firms' preference for full ownership in culturally dissimilar countries becomes more pronounced in non-restrictive host countries where they have greater freedom to choose between alternative ownership structures.

The coefficient of RSIZE, the relative size of the foreign operation, is statistically not significant with the unexpected sign in the full sample regression (Table 1). This is not consistent with the findings of Kogut and Singh (1988 b) that the size of the foreign affiliate is a significant predictor of joint venture (shared ownership) over acquisition (both fully-owned and partially owned). However, since they did not separate fully-owned acquisitions from partially-owned acquisitions, we cannot directly compare our findings with theirs. A possible interpretation for the positive relation is that investing firms may wish to have more control over their large foreign operations due to the larger share of corporate resources committed to the foreign operations.

RLTDNS, the relatedness of investment, is statistically not significant (but with the expected positive sign) in the full sample regression (Table 1). This positive relationship suggests that when firms diversify into product areas outside of their core businesses, they are more likely to share the ownership of their foreign affiliates to obtain complementary resources, such as product-specific know-how or access to distribution, from their equity partners. These findings seem to be in line with those reported in Stopford and Wells (1972) and Hennart (1991), though the coefficient is not statistically significant in this study.

The subsample analyses, however, provide interesting results (Table 2). The relatedness of investment shows alternating relations for pre-1986 and post-1985 Japanese FDIs: a significant but unexpectedly negative relation for the former and a significant and positive one for the latter. One explanation is that, during the earlier period (pre-1986) when Japanese firms generally are less familiar with the concept of FDI, they tend to seek equity partners more for their complementary local experiences in order to compensate for their limited experiences and consequently are more willing to jointly own their foreign affiliates even if these investments fall into the related category. In contrast, once they become more familiar with the mechanism of FDI during the later period (post-1985), they tend to seek partners only when they need complementary product-specific capability and be more likely to fully own their related investments.

The coefficient of EM, the establishment mode of foreign affiliate, is statistically not significant in the full sample regression (Table 1). Overall, Japanese firms seem to choose the ownership structure of their foreign affiliates independent of how they are established (acquisition or new venture). These results are in line with the findings of Hennart and Park (1991, 1993), but do not support Kogut and Singh's (1988 b) argument that they are related. However, subsample analyses show interesting results (Table 2). The establishment mode is significantly and negatively related to ownership structure for Japanese firms investing in culturally dissimilar host countries, in restrictive host countries, and before 1986. They are more likely to choose shared ownership structure for their foreign affiliates when the affiliates are established via acquisitions. One explanation is that they perceive risks and uncertainties associated with acquisitions to be much greater in culturally dissimilar countries, in restrictive countries, and when they have only limited international business experience, and, thus, prefer to share the risks and uncertainties with local equity partners. An alternate explanation is that restrictive host countries may discourage wholly owned acquisitions to a greater extent than wholly owned greenfield ventures.(12) A final possibility is that the subsample results may be induced by host country regulations that allow full ownership only for new ventures in the culturally dissimilar, restrictive, and pre-1986 cases.(13) Overall, despite the subsample results, the stronger full sample results lead us towards the view that MNC's ownership decision and establishment mode decision are basically separate decision issues.

The coefficient of ASIZE, the absolute size of the parent, is not significant in the full sample regression (Table 1). This is consistent with the findings of Kogut and Singh (1988 b) and Hennart (1991). However, these results could be induced by the intercorrelation between ASIZE and other variables like IBEXP, HCEXP, and RND (Appendix 2).(14) Any significance might have been subsumed by the significance of HCEXP and RND.

Finally, the coefficient of TIME is not significant in the full sample regression (Table 1), suggesting that the ownership strategy implied by more recent Japanese FDIs (post-1985) is similar to that of earlier ones (pre-1986). However, the period subsamples provide an interesting contrast between pre-1986 and post-1985 FDIs. For instance, the period subsample results show that familiarity with a host country becomes a more important determinant of the ownership choice of the more recent Japanese FDIs than that of the earlier ones (Table 2).

Conclusions

This study offers the first large sample investigation of the factors which affect the choice of foreign ownership structure by non-U.S. MNCs in a global setting, unlike previous studies which have examined the ownership policies of U.S. MNCs in foreign markets or those of non-U.S. MNCs in the United States. Also, this study represents the first attempt to test whether cultural differences influence the choice of foreign ownership structure by MNCs.

The results show that the parent's familiarity with host country, the parent's R & D intensity, and the cultural distance between home and host country have significantly positive influence on the Japanese firm's choice of full ownership of their foreign affiliates. The study also finds that the overt host requirement for prior approval for full foreign ownership operationally discourages Japanese firms from fully owning their affiliates in such countries. To summarize, Japanese firms generally prefer full ownership to shared ownership for their foreign affiliates when (1) the affiliate is located in a more familiar host country, (2) the investing firm is more R & D intensive, (3) they invest in culturally distant countries, and (4) they invest in less restrictive host countries. Furthermore, this study reveals an interesting relation between the MNC's ownership decision and establishment mode decision for its foreign affiliate. In particular, Japanese firms are more likely to choose the shared ownership structure for their foreign affiliates when the affiliates are established via acquisitions in culturally dissimilar host countries, in restrictive host countries, and before 1986.

We have two suggestions for future research. First, would the cultural distance between home and host country play a similarly important role in the foreign affiliate ownership strategy of non-Japanese MNCs? What would be the role of culture in other types of managerial decision-making of MNCs? To the extent that it does play an important role, we should argue that transaction cost-based explanations for such decision-making should be supplemented by cultural contexts of home and host countries. Second, a further investigation into the relationship between MNC's establishment mode decision and ownership decision for its foreign affiliate is warranted. The results of this study do not offer strong conclusive support for the existence of such a relationship. Additional research along these lines will enhance our understanding of the two key aspects of the foreign market entry strategy of MNCs.

[TABULAR DATA FOR APPENDIX 1 OMITTED]

[TABULAR DATA FOR APPENDIX 2 OMITTED]
Appendix 3. Logit Regression Results: Full Ownership versus
Shared Ownership (75 Percent Cutoff, Full sample)


Variables Coefficients (standard error)


Intercept 0.0159 (0.9388)
ASIZE -0.0919 (0.0744)
RSIZE 0.5114 (0.4693)
RLTDNS 0.2243 (0.2005)
IBEXP 0.0533 (0.0980)
HCEXP 0.2565(***) (0.0673)
RND 12.3675(***) (3.4728)
EM -0.2036 (0.1860)
HP -1.0973(***) (0.1640)
CD 0.1472(*) (0.0831)
TIME 0.2437 (0.1988)


Model chi-square 97.54 (p = 0.0001)


N = 839 (F: 512, S: 327)


Correct Ratio (%) 67.5


* p[less than]0.10, ** p[less than]0.05, *** p[less than]0.01
(two tailed)


F: Full ownership, S: Shared ownership


Endnotes

1 We thank an anonymous reviewer for this suggestion.

2 One alternative would be the actual level of ownership. However, this is inadequate since it treats the intervals as constant over the range of ownership level. Implications of the interval between, say, 60 percent foreign ownership and 40 percent would be different from that between 30 percent and 10 percent. In addition, in order to confirm the stability of our findings, we ran a regression using an alternative cutoff ratio (75 percent cutoff point) and obtained results similar to those using the 95 percent cutoff ratio (The results are reported in Appendix 3). The 95 percent cutoff threshold was used in almost all previous studies involving foreign ownership choice.

3 Logit analysis has several advantages as compared to other methods. OLS regression analysis is inadequate when the dependent variable is restricted to a [0/1] space. Frequently, the estimation results of logit and probit analyses for equal data sets are quite comparable (Altman et al. 1981). But the logit approach offers computational advantages due to the iterative technique implied.

4 We acknowledge that the currency changes may potentially affect this variable. However, since detailed information on underlying currency denomination of the parent's global assets is not publicly available, we are not able to accurately capture the influence of currency changes affecting this variable.

5 Ideally, relatedness of investment could be determined by comparing industrial classification codes of the products to be produced by the foreign affiliates with those of the parent's prime product lines. Unfortunately, consistent information on industrial classification of both the foreign affiliate and the parent is not available. As a result, following Harris and Ravenscraft (1991), we rely on product descriptions.

6 Kogut and Singh (1988 b, p. 422) measured cultural distance between the jth country and the U.S. using the following index:

[CD.sub.j] = [summation of] {[([I.sub.ij] - [I.sub.iu]).sup.2]/[V.sub.i]}/4 where i=1 to 4

where,

[I.sub.ij] = index for the ith cultural dimension and jth country; u = the United States; [V.sub.i] = variance of the index of the ith dimension; [CD.sub.j] = cultural distance of the jth country from the U.S.

7 The baseline rate, that is the classification rate that would have been obtained by chance, is equal to [a.sup.2] + [(1 - a).sup.2], where a is the proportion of full ownership in the sample (56.3%) (Morrison 1974).

8 The lack of significance of variables like ASIZE and TIME could be due to multicollinearity with other variables (Appendix 2).

9 Abegglen and Stalk (1985) and Rapp (1993) argue that Japanese firms prefer to establish, ceteris paribus, existing organizational and other related routines abroad in order to better preserve/expand global market share and ensure firm survival.

10 We thank an anonymous referee for this suggestion.

11 We thank an anonymous referee for this suggestion.

12 We thank an anonymous referee for this explanation.

13 We thank an anonymous reviewer for this suggestion. However, lack of precise data on sufficiently large number of host countries does not allow us to investigate this possibility.

14 Following the suggestions of an anonymous referee, we ran the overall regression with ASIZE, but excluding IBEXP, and with IBEXP, but excluding ASIZE. Given the significant correlation between ASIZE and IBEXP, this procedure potentially helps identify which variable is the dominant variable to discriminate between full and shared ownership. The results are:

ASIZE, excluding IBEXP: -0.0905 (coefficient), 0.0690 (std. error)

IBEXP, excluding ASIZE: 0.0506 (coefficient), 0.0923 (std. error)

Both variables appear to be insignificant in the absence of the other.

References

Abegglen, J. and G. Stalk, Kaisha: The Japanese Corporation, New York: Basic Books 1985.

Altman, E. I., R. B. Avery, R. A., Eisenbeis, and J. F. Sinkey, Jr., Application of Classification Techniques in Business, Banking and Finance, Contemporary Studies in Economics and Financial Analysis, vol. 3, Greenwich, CT: JAI Press 1981.

Anderson, E. and H. Gatignon, Modes of Foreign Entry: A Transaction Cost Analysis and Propositions, Journal of International Business Studies, 17, 1986, 3, pp. 1-26.

Business International (various years), Investing, Licensing and Trading Conditions Abroad.

Buckley, P. and M. Casson, The Future of Multinational Enterprise, London: McMillan 1976.

Caves, R., Multinational Enterprise and Economic Analysis, New York: Cambridge University Press 1982.

Caves, R. and S. Mehra, Entry of Foreign Multinationals into U.S. Manufacturing Industries, in M. Porter (ed.), Competition in Global Industries, Boston, MA: Harvard Business School Press 1986.

Cho, K. R. and P. Padmanabhan, Acquisition versus New Venture: The Choice of Foreign Establishment Mode, UCD-CBA Working Paper 1992-9, Denver, CO: University of Colorado at Denver 1992.

Davidson, W., Global Strategic Management, New York: John Wiley 1982.

Erramilli, M. K., The Experience Factor in Foreign Market Entry Behavior of Service Firms, Journal of International Business Studies, 3, 1991, 479-501.

Farge, N. and L. Wells, Jr., Bargaining Power of Multinationals and Host Governments, Journal of International Business Studies, Fall 1982, pp. 9-23.

Franko, L., Joint Venture Survival in Multinational Corporations, New York: Praeger 1971.

Gatignon, H. and E. Anderson, The Multinational Corporation's Degree of Control Over Subsidiaries: An Empirical Test of a Transaction Cost Explanation, Journal of Law, Economics, and Organization, 4, 1988, 2, pp 305-336.

Genay, H., Japan's Corporate Groups, Economic Perspectives (Federal Reserve Bank of Chicago), Jan.-Feb. 1991, pp 20-30.

Gomes-Casseres, B., Ownership Structures of Foreign Subsidiaries, Journal of Economic Behavior and Organization, 11, 1989, pp 1-25.

Gomes-Casseres, B., Firm Ownership Preferences and Host Government Restrictions: An Integrated Approach, Journal of International Business Studies, 21, 1990, 1, pp 1-22.

Grieco, J., India's Experience with the International Computer Industry, International Organization, Summery, 1982, pp. 609-632.

Harrell, F., Jr., The LOGIST Procedure, in R. P. Hastings (ed.), SUGI Supplemental Library User's Guide (5th ed.), Cary, NC: SAS Institute 1986.

Harris, R. and D. Ravenscraft, The Role of Acquisitions in Foreign Direct Investment: Evidence From the U.S. Stock Market, Journal of Finance, July 1991, pp 825-844.

Hennart, J. F., A Transaction Costs Theory of Equity Joint Ventures, Strategic Management Journal, 9, 1988, pp 361-374.

Hennart, J. F., The Transaction Costs Theory of Joint Ventures: An Empirical Study of Japanese Subsidiaries in the United States, Management Science, 37, 1991, 4, pp 483-497.

Hennart, J. F. and Y. R. Park, Greenfield vs. Acquisition: The Strategy of Japanese Investors in the United States, in H. Vestergaard (ed.), An Enlarged Europe in the Global Economy, Copenhagen, Denmark: Copenhagen Business School 1991.

Hennart, J. F. and Y. R. Park, Greenfield vs. Acquisition: The Strategy of Japanese Investors in the United States, Management Science, 39, 1993, 9, pp. 1054-1070.

Hladik, K., International Joint Ventures: An Economic Analysis of U.S. Foreign Business Partnerships, Lexington, MA: Lexington Books 1985.

Hofstede, G., Culture's Consequences, Beverly Hills, CA: Sage Publications 1980.

Hofstede, G. and M. Bond, The Confucius Connection: From Cultural Roots to Economic Growth, Organizational Dynamics, Spring 1988, pp. 4-21.

Japan Company Handbook (various years).

Kim, W. C. and P. Hwang, Global Strategy and Multinationals' Entry Mode Choice, Journal of International Business Studies, 23, 1992, 1, pp. 29-53.

Kogut, B. and H. Singh, Entering the United States by Joint Ventures: Competitive Rivalry and Industry Structure, in F. Contractor and P. Lorange (eds.), Cooperative Strategies in International Business, Lexington, MA: Lexington Books 1988a.

Kogut, B. and H. Singh, The Effect of National Culture on the Choice of Entry Mode, Journal of International Business Studies, 19, 1988b, 3, pp. 411-432.

Lecraw, D., Bargaining Ower, Ownership, and Profitability of Transnational Corporations in Developing Countries. Journal of International Business Studies, Spring-Summer 1984, pp. 27-43.

Morrison, D., Discriminant Analysis, in R. Ferber (ed.), Handbook of Marketing Research New York: Wiley 1974, pp. 2442-2457.

Price Waterhouse (various years), Information Guide (Doing business series).

Rapp, W., Japanese Foreign Direct Investment in the Product Cycle, in Bird and Beechler (eds.), Proceedings of the Sixth Annual Meeting of the Association of Japanese Business Studies, New York: Association of Japanese Business Studies 1993.

Root, F., Entry Strategies for International Markets, Lexington, MA: D. C. Heath 1987.

Rugman, A., Inside the Multinationals, New York: Columbia University Press 1981.

Stopford, J. and L. Wells, Jr., Managing the Multinational Enterprise, New York: Basic Books 1972.

Toyo Keizai Shinposha, Japanese Overseas Investment: A Complete Listing by Firms and Countries, 1992/1993, Tokyo, Japan: Toyo Keizai Shinposha 1992.

United Nations Centre on Transnational Corporations 1978-1985. National Legislation and Regulations Relating to Transnational Corporations New York: United Nations Publications.

U.S. Department of Commerce (various years) a. Economic Trends and Their Implications for the United States, Washington, DC: International Trade Administration.

U.S. Department of Commerce (various years) b. Investment Climate in Foreign Countries (Annual and periodic reports), Washington, DC: International Trade Administration.

U.S. Department of Commerce (various years) c. Overseas Business Reports (International marketing information series), Washington, DC: International Trade Administration.

Wilkins, M., Japanese Multinationals in the United States: Continuity and Changes 1879-1990. Business History Review, Winter 1990, pp. 585-629.

Prasad Padmanabhan, Associate Professor of Finance, Department of Finance, San Diego State University, San Diego, CA, U.S.A.

Kang Rae Cho, Associate Professor of International Business College of Business Administration, University of Colorado at Denver, Denver, CO, U.S.A.
COPYRIGHT 1996 Gabler Verlag
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1996 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Padmanabhan, Prasad; Cho, Kang Rae
Publication:Management International Review
Date:Jan 1, 1996
Words:7896
Previous Article:Entry mode and performance of Japanese FDI in Western Europe.
Next Article:Performance factors of subsidiaries abroad: lessons in an analysis of German subsidiaries in France.
Topics:


Related Articles
Factors affecting price earnings ratios and market values of Japanes firms.
Strategic mindsets and decision-making autonomy in U.S. and Japanese MNCs.
An empirical analysis of factors affecting cross-border acquisitions: U.S.-Japan.
Comparing the profitability of firms in Germany, Japan, and the United States.
The Trojan horses: Japanese and U.S. presence in Europe.
Comparative analysis of export-oriented and foreign production-oriented firms' foreign market entry decisions: a reply.
Distinctive feature of domestic and international joint ventures.
Ownership strategy of multinationals from ASEAN: the case of their investment in Sino-foreign joint ventures (1).
The dynamics of Japanese firm growth in U.S. industries: the Penrose effect.
Determinants of ownership-based entry mode choice of MNEs: evidence from Mongolia.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters