Knee deep in the big muddy: the survival of emerging market firms in developed markets.
* This study of Latin American banks located in the United States employs a resource-based framework to explain how subunits of emerging market firms can overcome the challenges of operating in a developed market.
* Our results show that an EMF subunit can draw upon ethnic identity as a valuable and costly-to-imitate resource to achieve competitive parity in the developed market.
* Ethnic resources can be generated not only from ethnically similar customers but also from ethnically similar competitors in the local market.
* In addition, the parent firm' level local and non-local resources can help to achieve competitive parity for the EMF subunits in that country, which in turn positively influences survival.
* However, over expansion can lead to spreading local and non-local resources too thin, thus adversely affecting survival.
Keywords: Resource-based View * Survival * Local Ethnic Density * Local Depth * Geographic Breadth. Emerging Market Firm * Banking
When a firm expands across national borders, the principal hazard faced by its subunits is the liability of foreignness (Hymer 1960/1976, Mezias 2002, Nachum 2003, Zaheer 1995). Liability of foreignness is the sum of the social costs that arise from the unfamiliarity, relational and discriminatory hazards that foreign firms experience over and above those faced by local firms in the host country (Eden/Miller 2004). Because of liability of foreignness, subunits of multinationals entering a host country have a lower likelihood of surviving after entry into the new market.
To overcome the liability of foreignness, some scholars argue that multinational enterprises (MNEs) need to provide overseas subunits with certain firm-specific (home-based) resources, to enable survival and higher foreign subsidiary performance (e.g., Caves 1982, Mitchell/Shaver/Yeung, 1994, Rugman 1981, Shaver/Mitchell/Yeung 1997). These resource-based arguments for overcoming liability of foreignness are well suited for subunits of developed market firms, but their efficacy for subunits of emerging market firms (EMFs) remains unclear. Most EMFs that enter developed markets compete from a position of double disadvantage: The firms incur additional costs of doing business abroad, and in addition they are often resource poor compared to domestic firms in developed markets (Thomas/Eden/Hitt/Miller 2007). Although there are clear exceptions such as the Tata Group and Wipro from India, Embraer from Brazil, and Cemex from Mexico that have stronger resources than many of their developed market competitors (Bartlett/Ghoshal 2000, Khanna/Palepu 2006, Makino/Lau/Yeh 2002), the majority of EMFs have weaker marketing resources (e.g., brand recognition) and technology resources (e.g., fewer patents and copyrights) than most developed market firms. This weakness makes it difficult for EMF subunits to compete effectively by using a resource-exploitation strategy in developed markets (Dawar/Frost 1999, Hoskisson et al. 2000, Khanna/Palepu 1999).
In effect, emerging market firms entering developed markets start off "knee deep in the Big Muddy" or more likely waist or neck deep, to quote the old Pete Seeger song about an army platoon trying to cross the Mississippi River. Do these EMFs have the resources to "cross the river" and survive? Despite the double-disadvantage that exists (presented above), some EMF subunits do manage to survive and even prosper in developed markets. For example, Miami, Florida is host to many agencies of foreign-owned banks, such as Venezuelan-owned Commercebank. U.S. banking services with Latin America are derived not only from trade finance and private wealth, but also from active business investments in the United States. According to Fernando Capablanca, CEO of the Miami agency of Chile's Banco de Credito e Inversiones, "In the last few months at least three Chilean multinationals have settled here, primarily to handle their trade with the US and Europe." Referring to LanChile, the largest single foreign investment in Miami, Mr. Capablanca further noted that, "Eventually, what happens is that you have a lot of banks following these companies, because these are the banks' customers. When you get a few of them, you get a synergy." (Latin CEO: Executive Strategies for the Americas 2002).
Why and how emerging market firms can survive and be successful in developed markets poses an interesting conundrum for resource-based view scholars. We offer two possible answers to this conundrum, both of which revisit the notion that location is critical to organizational survival. First, we argue that an EMF subunit can draw upon ethnic identity as a valuable and costly-to-imitate resource to achieve competitive parity in the developed market. Ethnic resources can be generated not only from ethnically similar customers but also from ethnically similar competitors in the local market. Although ethnicity has been inferred as a driver of location choice in the agglomeration literature (Chang/Park 2005), scholars have been silent on its performance implications. The density dependence literature simply posits a curvilinear relationship between density and performance, assuming intense rivalry at high levels of density (Carroll/Hannan 1989, Hannan/Freeman 1977). However, this literature ignores the coopetition that can arise if a group of ethnically similar firms band together to combat local rivals ("us versus them") rather than compete against each other and it ignores pent-up demand. Our second insight into the conundrum is the role played by an EMF's resources within the host country. We argue that the parent firm's local and non-local resources can help to achieve competitive parity for the EMF subunits in that country, which in turn positively influences survival.
To test our framework, we examine the survival of Latin American banking subunits in the United States. Our dataset provides an appropriate test because it consists of banks headquartered in emerging market countries that share a common Hispanic culture and identity. Moreover, local market entry decisions are especially relevant to service firms because of the need to be located in close proximity to customers (Greve 2000, 2002). It is not uncommon for service firms such as banks to have multiple subunits in one or more local markets of a country. Prior studies have used the organizational learning and search literatures to explain that banks are more likely to establish subunits in local markets for which they have a presence and in neighboring markets (Greve 2000). We build on this earlier literature to examine the role ethnicity plays in the survival of foreign banking affiliates.
The objective of our study is to answer the question: To what extent do home-grown resources and host-country resources influence the survival o fan EMF subunit in a given location within a developed market? Our paper extends previous research in the following ways. First, our resource-based and organizational learning framework suggests that the value of a resource varies from EMF subunit to subunit within a country; as a result, the performance implications of a particular resource are idiosyncratic to each subunit. Second, our study extends Baum and Mezias (1992) and Miller and Eden (2006) by hypothesizing that foreign subunits can share a collective identity resource and therefore, as a group, they have a source of potential cooperative advantage relative to domestic firms that can enhance their ability to survive in developed markets. In our study, we differentiate home-grown resources, host-country local resources, and host-country non-local resources for a particular EMF subunit (because they affect the inter-organizational transfer of knowledge). In addition, we show that local depth (the extent of an MNE's presence in a local market of a host country) and geographic breadth (the extent of an MNE's non-local presence in a host country) each have curvilinear relationships with subunit survival, which builds on prior studies by Greve (2000) and Mitchell, Shaver, and Yeung (1992, 1993, 1994).
Some scholars posit that imitating the best-practicing local firms can reduce the liability of foreignness and improve performance (Kostova/Zaheer 1999). However, imitating best practices of local firms in developed markets is a difficult strategy to follow when the foreign entrant has fewer resources and is less efficient in exploiting its resource portfolio than domestic firms. Our application of ethnic identity to resource-based view thinking shows how EMFs can reduce the likelihood of "drowning" in developed markets (the "big muddy" for them). Ethnicity not only increases brand recognition and the ability to exploit the EMF's own resources, but may also improve access to local resources (e.g., employees, learning spillovers) that can be valuable and difficult to imitate or to substitute in specific market niches (Sirmon/Hitt/Ireland 2007). Lastly, the resource-based view has typically been employed to explain performance within an industry (Barney 1991). For many EMFs operating in developed markets, the goal is not necessarily to achieve above average returns, but rather to achieve competitive parity. To this end, we attempt to show that survival can be an appropriate measure of performance for these contexts.
Home-grown Resource: Ethnicity as a Collective Identity Resource
According to Nagel (1994, p. 152), "identity and culture are two of the basic building blocks of ethnicity", which he defines as "the product of actions undertaken by ethnic groups as they shape and reshape their self-definition and culture". Identity and culture are important to establishing boundaries and producing meaning. Ethnic boundaries determine who is and is not a member of a group. Ethnic identity therefore represents a specific form of collective identity or group identity (Ashforth/Mael 1989).
While ethnicity and nationality are nearly synonymous in a few countries (e.g., Japan), country boundaries are generally not conterminous with ethnic boundaries. First, many countries share the same ethnic heritage, giving ethnicity a cross-national or interregional dimension. Second, foreign entrants inside a country tend to cluster in communities with other individuals sharing the same identity and culture. Ethnic enclaves within a country result from the immigrant's desire to reunite with family members who had migrated earlier, and to reduce the immigrant's initial unfamiliarity with the host country (Bartel 1989). Local agglomeration, such as Hispanics in Florida or Texas, gives ethnic identity a subnational dimension.
While ethnic identity is normally attached to individuals, organizations can have identities also (Douglas 1986). Identity for organizations is manifested in "their constituent individuals, physical resources, and the shared beliefs and institutions of which individuals function as a 'we'" (Wendt 1994, p. 385). Resource-based view scholars have argued that an organization's culture can be an intangible resource (Barney 1986). Research on organizational learning supports the notion of a relationship between the individual and organizational levels of analysis, suggesting that organizational learning is partly dependent on individual-level learning (Schilling et al. 2003). Based on these arguments, we conclude that ethnic identity can be shared by individuals and organizations, and can be considered a resource to the organization. Firms that share the same ethnic identity therefore share the same group-level ethnicity resource (Wells 1983).
Additionally, the country-of-origin literature suggests that ethnicity can be used by MNEs to overcome the liability of foreignness (Dinnie 2004). In a given host country, due to ethnic ties, subunits of firms from ethnically similar countries perceive each other as "less foreign" than local domestic firms. At the same time, ethnicity is costly to imitate for local domestic firms because ethnicity is embedded in the MNE subunit's identity. Moreover, foreign firms entering a host country tend to cluster with firms of the same nationality (Barry/Gorg/Strobl 2003, Chang/Park 2005). While valuable and difficult-to-imitate resources help firms create competitive advantages, their abilities to do so are affected by the ways in which they are managed (Sirmon et al. 2007) and the relative value of their resources compared with those of competitors (Sirmon/Gove/Hitt 2008).
Ethnically Similar Competitors and Customers
The density dependence literature suggests that competitive and institutional pressures vary in strength depending on density (Hannan/Freeman 1977). Institutional pressures for cognitive legitimacy (Ruef/Scott 1998) tend to be high early in the life of an organizational population, when density is low. However, as the number of organizations grows, interdependencies among organizations cause them to become increasingly competitive (Baum/Singh 1994). Increases in the number of organizations dependent upon the same finite pool of resources ultimately exceed the resources available, resulting in lower individual and collective organizational growth rates. Accordingly, legitimacy pressures are stronger than competitive pressures in low-density environments, but they become relatively weaker as density increases. Research indicates that an inverted U-shaped relationship exists between density and survival rates (Hannan/Freeman 1977, Carroll/Hannan 1989). Miller and Eden (2006) also found that greater local density of domestic competitors reduced the likelihood of survival for foreign entrants.
We contend that two conditions can produce a linear- rather than an inverted-U shaped--relationship between local density and survival: Cooperation among local ethnically similar subunits and a growing underserved ethnic population that leads to unrealized and potential demand for banking services exceeding supply (of banking services). We contend that the first condition can be a result of the second. We discuss each in turn. Coopetition among ethnically similar firms. Among ethnic EMF subunits, the collective identity argument suggests that members of this organizational sub-population compete as a group for local ethnic resources but the competition is not highly intense. Although prior density dependence research posits that local competitive pressures exceed institutional pressures, the collective identity in this case produces a situation whereby competitiveness is lower and cooperation is higher among ethnic EMF subunits. The social identity research provides considerable evidence that members view their group and colleagues in the group more favorably than other groups and their respective members (Kane/Argote/Levine 2005). Group members are perceived by other members in the same group as more trustworthy, honest, loyal, cooperative, and valuable compared with non-group members (e.g., Hewstone/Rubin/Willis 2000). These conclusions about group members' perceptions of each other can be extended from the individual level to the organizational level, supporting Tsai's (2002) views on coopetition and Schilling et al.'s (2003) research on the relationship between individual and organizational learning.
Similar to individuals in a group being more comfortable sharing knowledge with individuals who have the same social identity, firms that have the same ethnic identity are more inclined to share information with each other about the local environment. As the number of firms sharing ethnic identity increases in the local environment, more knowledge transfer is likely to occur across ethnic subunits. Knowledge transfer may include overt knowledge sharing or vicarious learning resulting from information spillovers within the ethnic group--that is, there are more group members from which to learn best practices. As such, increases in the local density of ethnic EMF subunits create "strength in numbers," which facilitates the survival of ethnic group members.
In addition, the EMF subunit benefits from the stock of ethnically similar suppliers of both intermediate goods and factor services. The host country's local ethnic population also provides a pool of skilled ethnic labor resources (human capital) that understands the culture of the foreign firm's subunit (Hitt et al. 2001, Hitt et al. 2006).
An increase in the local density of ethnically similar EMF subunits implies that local customers, especially ethically similar customers, become more familiar with these firms. The increased familiarity reduces liability of foreignness for potential entrants. Because of a stronger sense of identification and attachment to a firm with a particular ethnic heritage, an ethnically similar customer is more likely to establish a closer relationship and have more relational trust with that firm. Research has emphasized the importance of relational resources and trustworthiness as a source of competitive advantage (Barney/ Hansen 1994, Hitt et al. 2006). Relational trust can lead customers to remain in a relationship with its supplier, even though a more economically attractive alternative exists (Saparito et al. 2004). Such customers are also likely to be satisfied with the status quo of the relationship. As such, relations between organizations and customers that share similar ethnic identities entail greater trust, more information sharing and enhanced joint problem solving (Uzzi 1997).
An ethnic MNE subunit is therefore more likely to become embedded in a local community of similar ethnicity than a domestic firm because of its familiarity with the ethnic population's culture, language, and consumer preferences. In this case, ethnic identity makes foreignness a benefit, rather than a liability, in the eyes of ethnically similar local residents. The local ethnic community may be more likely to conduct business and establish closer relationships with ethnically similar MNE subunits than with domestic firms or ethnically dissimilar MNE subunits.
Thus, an increase in the local density of ethnically similar EMF subunits enhances legitimacy and intra-group knowledge sharing that, in turn, increases the value of ethnic identity as a resource in the local environment, thereby enhancing the probability of EMF subunit survival (rather than intensifying competition for local ethnic resources).
Pent-up demand from ethnically similar customers. We also argue that the nature of the local ethnic population has implications for local competitive intensity. Density dependence scholars posit a curvilinear relationship between local density and survival based on the assumption that as the number of firms grows, market size remains unchanged and so firms are forced to compete for increasingly more scarce customers, and thus fewer firms survive. These congestion effects, however, only occur in the upper end of the distribution. If the local ethnic population is (1) currently under-served by local providers and/or (2) growing rapidly in the host location, excess demand for the services provided by the EMF subunits will persist for long periods, even with a number of new entrants. For example, a fast growing Hispanic population in Miami, Florida, coupled with a large percentage of the existing Hispanic population that has forgone banking services, suggests that congestion effects from foreign banking entrants are highly unlikely with this market segment. Moreover, the unique demands and needs of Hispanic consumers have proved costly to serve for the few U.S. banks that have attempted to target this market segment (Medina 2007).
Integrating the coopetition and excess demand rationales, we argue that the EMF subunits continue to benefit from increasing local density of ethnically similar firms. While the agglomeration literature suggests that an inverted-U relationship exists between local density and firm survival, this view ignores the coopetition benefits of ethnic subunits derived from ethnic identity as well as the loyalty, commitment and pent-up excess demand of the local ethnic population as customers and labor resources. As such, a higher density of local ethnic population contributes to the survival of local ethnic MNE subunits. Thus:
Hypothesis 1. There is a positive relationship between local ethnic density and the likelihood of survival of an ethnically similar EMF subunit located in the same local market of a developed country.
Host-country Resources: Local Depth and Geographic Breadth
The second part of our framework focuses on firm-specific resources. EMFs have local and non-local resources that also can contribute to competitive parity in a developed market.
While agglomeration research suggests that location decisions reflect learning from the prior decisions of other firms, especially other foreign firms, the organizational learning literature posits that location choices are also based on knowledge gained from the firm's experience of repeating choices that seemed to be successful. An organization accumulates knowledge from operating a subunit in a particular local geographic space that facilitates information acquisition pertaining to customer preferences, local demand, competitor behavior, and the feasibility of operating a new local subunit to exploit local opportunities. Research suggests that organizations learn from operating in a particular geographic space by replicating successful routines (Levitt/March 1988, Greve 2000).
Within the experiential learning literature, researchers have focused on the notion of search, broadly defined as a process through which organizations analyze and adapt to their external environments (March 1991). Organizations search for alternatives and information about courses of actions such as the appropriate location of subunits (Baum et al. 2000). Search is viewed as a process of allocating resources between exploration of new routines and exploitation of existing routines (Levinthal/March 1993). Experience with particular locations allows a firm's senior managers to acquire information about demand, competitors, and thereby to identify unexploited local opportunities (Greve 1996, 1998).
A number of studies have explored spatial expansion of organizations (Baum/Li/Usher 2000, Greve 1996, 2000). Services firms represent a form of "chain" organization, where subunits conduct roughly the same activities in multiple, different geographic locations. The knowledge learned from the different locations helps to determine the next location choice for a chain organization. The next location selected could be in the same geographic area as one or more of the existing subunits, thus increasing the local depth of the MNE group, or in a different (new) location, thus increasing the geographic breadth of the MNE group.
There is evidence for both types of expansion. Research has shown that banks are more inclined to establish new subunits in local communities where they already have a presence (Greve 2000). Thus, local depth can be a valuable experiential resource for a given EMF subunit in the same local market. On the other hand, geographic breadth provides a MNE with name recognition that can spillover to a given local market and increases the likelihood of attracting new customers and retaining existing ones. Geographic breadth provides slack and experience, and perhaps also the knowledge that allows the firm to better use the ethnic resource in the local market. Shaver et al. (1997) found that foreign firms with multiple subunits in the host country were more likely to survive, suggesting that geographic breadth is positively related to survival. Greve's (2000) study concluded that banks were also likely to establish new subunits in adjacent locations.
Scholars have suggested that experiential resources can help a foreign subunit overcome the liability of foreignness (Zaheer 1995). The experience from operating multiple subunits also involves learning by doing, which reflects "the realization of improvements in productivity through experience production" (Epple/Argote/Devadas 1991, p. 58). The classic learning curve model assumes that learning is cumulative and that it persists over time; however, Argote, Beckman, and Epple (1990) suggested that knowledge acquired via learning by doing may depreciate--that is, organizations can forget over time. As a result, failure to account for depreciation of learning may result in forecasting errors. Moreover, Argote et al. (1990) called into question the persistence of learning after an organization begins production, implying that a new subunit may be more productive than its predecessors, but that these benefits rarely benefit other subunits.
Experience in local ethnic markets can help a firm to develop an effective competitive strategy using knowledge developed from other ethnic markets in the host country. Convenience, name recognition and experience, arising from an EMF's market depth increase the value of ethnic identity and make it more costly-to-imitate for that particular firm, which in turn, enhances the performance of its ethnic EMF subunits.
As the number of local subunits for a given firm increases, the firm encounters local coordination costs and depreciating learning benefits. Even for an EMF that derives benefits from coopetition, increasing its local presence may stretch local resources too thin, and therefore, diminish knowledge transfer at the subunit level. In an ethnic niche market, a firm with high local depth may need to resort to sub-optimal locations for subunits, and thus may be unable to fully exploit local experiential knowledge. Lastly, we expect the coordination costs and resource strain to be especially difficult for EMFs, which in general, have fewer resources, especially managerial capabilities, than foreign firms from developed market countries.
In a similar vein, non-local resources are embodied in a firm's geographic depth. Despite the added ethnic benefits that accrue to an EMF subunit from its parent's geographic breadth in the host country, a substantial number of subunits spread geographically across a country places demands on corporate management. Increased geographic dispersion increases information processing demands that, in turn, lead to higher coordination, distribution, and managerial costs (Hitt/Hoskisson/Ireland 1994). These demands in geographic diversification within a country are similar to those discovered in geographic diversification across countries (Hitt/Hoskisson/Kim 1997, Mitchell et al. 1992).
Moreover, the likelihood of identifying promising opportunities within a country diminishes as the most immediate opportunities have been exploited. At high levels of geographic breadth, resources may be spread too thin across subunits and/or the transactions costs required to coordinate the sharing of knowledge resources across the subunits in the host country are greater than the benefits achieved, and thereby compromising the benefits derived from distance learning. These arguments suggest a nonlinear relationship between geographic depth and subunit survival. Thus:
Hypothesis 2. There is a curvilinear relationship between the geographic breadth of an emerging market firm and the likelihood of survival of its subunit in a local market of a developed country. That is, geographic breadth increases the likelihood of survival at low and moderate levels, but it begins to decrease at high levels of geographic breadth.
Hypothesis 3. There is a curvilinear relationship between the local depth of an emerging market firm and the likelihood of survival of its subunit in a local market of a developed country. That is, local depth increases the likelihood of survival at low and moderate levels, but it begins to decrease at high levels of local depth.
Sample and Statistical Method
Our dataset consists of the U.S. subunits (that is, representative offices, branches, agencies, Edge Act Corporations, and subsidiaries) of banks from Mexico, Central America, and South America. These banks are all headquartered in emerging market countries and share the same Hispanic culture and ethnic heritage. We collected data on the locations of U.S. subunits from the Federal Reserve Board and Federal Deposit Insurance Corporation (FDIC). Our sample initially included all subunits of Latin American banks. We excluded 13 subunits for which relevant data were unavailable.
We examined the survival rate of these EMF subunits during the period 1989 to 2003 in given metropolitan statistical areas (MSA) in the United States. An MSA is generally defined as "an area containing a recognized population nucleus and adjacent communities that have a high degree of integration with that nucleus" (Office of Management and Budget 2000). MSAs represent the most common industry boundaries for studies involving commercial banks (Amel/Rhodes 1988, Barnett/Greve/Park 1994, Berger 1995).
Because we study survival rates, we use event history analysis where the unit of analysis is the event history of each subunit's entry into an MSA. For each subunit, we determine the date of entry into a focal MSA as the beginning of the analysis. We have a dichotomous dependent variable that equals zero if an EMF subunit in a particular MSA survived during a particular year t, and equals one if the subunit was closed. We employ Cox's proportional hazards model, commonly used to model event histories when the statistical function of the hazard rate is unknown or not assumed beforehand (Morita/Lee/Mowday 1993). We tested the proportional hazards assumption and could not reject the null hypothesis that the hazards are proportional; therefore, we stratified the sample by country of origin of each subunit. Stratifying the sample allows the baseline hazard rate to vary across countries of origin (Gimeno et al. 2005).
We estimated the models using the STCOX procedure in STATA/SE 8.0. We used robust standard errors (clustered at the parent level) in order to account for the potential non-independence of spells from each event history (Gimeno et al. 2005). Given that the independent variables vary on an annual basis, we split the sample into spells, as is customary in event history analysis (Gimeno et al. 2005). This resulted in 111 firm-MSA event histories and 1059 spells. Sixty-nine of the 111 subunits failed during the period of study. Following Shaver (1998), we also employed the Heckman (1979) procedure to test for self-selection bias. The procedure indicated minimal selection effects, so the inverse mills ratios were inserted in the models to control for these effects.
We measure local ethnic density with a ratio: The number of subunits of banks from Latin American countries in a particular metropolitan statistical area for a given year divided by the number of people from Hispanic countries in the metropolitan statistical area as reported by the U.S. Census Bureau. Because the census data are reported every 10 years, we interpolate the population for years 1991 to 1999 and extrapolate for years 1989 and 2001-2003.
Local depth. We measure this host-country local resource of an MNE as the number of U.S. banking subunits of a foreign bank in a particular metropolitan statistical area (MSA) in a given year.
Geographic breadth. We measure this host-country non-local resource of an MNE as the number of non-local banking subunits (that is, subunits in other MSAs) of a foreign bank located in the United States for a given year.
Parent-level control variables: Parent size is the natural log of the parent firm's assets in year t. Parent age is the difference between the current year of the study and the establishment date of the parent bank. Data for firm age are obtained from the Federal Reserve, FDIC, Bankersalmanac.com, Moody's, and BankScope. Parent performance is measured using return on assets of the parent. Parent leverage is measured as the debt-to-asset ratio of the parent. Parent level data are obtained from Bank Scope, Mergent Online, and The Banker, while branch and agency-specific data are obtained from the Federal Reserve. On several occasions, the parent bank was contacted to confirm the entry year.
MSA-level control variable." We control for local domestic density with a ratio--the number of domestic banks in the MSA in a particular year, based on FDIC's Call Report data, divided by the non-Hispanic population.
Home-country control variables. Home country real GDP growth (lagged one year) was controlled for in order to capture differences in country-level economic expansion.
Subunit control variables: To control for mode of entry, we included an acquisition dummy variable that equals 1 if the subunit entry was an acquisition and 0 if a greenfield venture. We also include dummy variables for several subunit types: branch office, agency office, Edge Act Corporation, and representative office. We excluded the dummy variable for subsidiary office, treating it as the base category.
Table 1 provides summary statistics and correlations for each variable. We draw your attention to the means for local domestic density (0.26) and local ethnic density (0.02). This preliminary information indicates that there are fewer Hispanic banks per capital than domestic banks per capita in local markets within the United States. Moreover, this preliminary information is consistent with our explanation that local Hispanic populations are relatively underserved customers. Table 2 provides results of hierarchical models testing our hypotheses. For models 1-5, variance inflation factors are well below generally accepted levels suggesting that multicollinearity is not a problem.
In Table 2, the chi-squared statistic for each model is statistically significant. Model 1 of Table 2 presents control variables only. Model 2 includes the main effects for the hypothesized variables. The coefficient for local ethnic density is negative and statistically significant, implying that it reduces the likelihood of failure (increases the likelihood of survival) of an ethnic foreign firm's subunit. These results provide support for Hypothesis 1.
Model 3 introduces the three main effects and geographic breadth squared. The positive, statistically significant coefficient for geographic breadth squared, coupled with the statistically significant negative coefficient for the main effect of geographic breadth suggests an inverted U-shaped relationship--geographic breadth reduces the likelihood of failure (enhances the likelihood of survival) of an EMF subunit to at least moderate levels, but the likelihood of failure (survival) begins to increase (decrease) at high levels of geographic breadth. These results support Hypothesis 2.
Model 4 includes the three main effects and local depth squared. The results indicate that the coefficient for local depth squared is positive and statistically significant while the coefficient for local depth is negative and statistically significant. These results also suggest an inverted U-shaped relationship; local depth reduces the likelihood of failure (enhances survival) at low and moderate levels, but the likelihood of failure (survival) begins to increase (decrease) at high levels of local depth. These results provide support for Hypothesis 3. Controlling for selection biases only had minor effects on the results. The inverse mills ratio was statistically significant only in model 1. Model 5 represents the full model, which includes all squared terms. In a test of robustness, we included the squared term for local ethnic density; however, no evidence of a curvilinear relationship between local ethnic density and survival was found.
Our study decomposed local density to highlight local ethnic density as an important concept. Therefore, we used a ratio--i.e., the number of organizations per capita. In the density dependence literature, the concept of density is the typically measured as the number of organizations in a particular environment and the local human population is used as a control variable. Thus, in another test of robustness, we used the density dependence approach, which now entails local ethnic density, local domestic density and local ethnic population and local domestic population (excluding the ethnic population). After centering the hypothesized variables, the coefficients for the new local ethnic density variable were consistent with the reported results. Our test to determine if multicollinearity is problematic indicated that the average variance inflation factor (VIF) for each model and the maximum individual VIF is well below the accepted level--the highest individual VIF is 4.78 (model 5) and the highest overall VIF is 2.43 (model 5). In each case, we are running two interactions. So, we are well below the cutoff that we listed of 10.0.
We also considered an approach employed by Mitra and Golder (2002) and Van den Poel and Lariviere (2004), in which variables are added sequentially to the model to assess the stability of the parameters and therefore, provided added assurance that multicollinearity does not impact our results because of concerns about the potential affects of multicollinearity in hazard models (Van den Poel/Lariviere 2004). In these robustness tests, we estimated models using a correlation threshold of 0.50. Then, we used a correlation threshold of 0.60. With a 0.50 correlation threshold, we drop several control variables. The coefficients from local ethnic density was negative and significant (p < 0.05). The coefficient for local depth is negative (p < 0.05) and local depth squared is positive (p < 0.001). Similarly, the coefficient for geographic breadth is negative (p < 0.001) while the coefficient for geographic breadth is positive (p < 0.001). These results are consistent with our reported results. For this set of results, the highest individual VIF was 4.45 and the average VIF was 2.08 in the full model.
In another test of robustness, we included a three-way interaction. The coefficient on the three-way interaction between local ethnic density * local depth * geographic breadth was not statistically significant; however, the coefficients on the hypothesized variables remained significant and in the same direction as in model 5. In this model the average VIF was 2.68 and the largest individual VIF was 5.18. Again these VIFs are within acceptable levels. Please note that the hypothesized variables were centered prior to computation of the squared terms (and even the 3-way interaction) to attenuate multicollinearity concerns (also see Zorn 2001, Brambor/Clark/Golder 2006). Our results for the hypothesized variables were robust using a 0.50 or 0.60 correlation threshold with some theoretically driven controls dropped. (1) These results are not reported, but available upon request from the authors.
The present study sought to enrich our understanding of the resource-based view by examining the survival of subunits of Latin American banks in the United States. The resource-based view has focused on a firm's resources relative to those of industry rivals (Barney 1991). International applications of the resource-based view highlight the transfer of resources and capabilities from the parent to foreign subsidiaries and the development of new resources and capabilities in the host country as methods for overcoming liability of foreignness.
Prior research has not considered the possibility that EMF subunits in a host country can cooperate to exploit a valuable and costly-to-imitate resource, such as ethnic identity. We infer from the results that some EMF subunits can overcome this disadvantage and achieve competitive parity in developed markets. Moreover, our results suggest that prior research may have overstated the effects of the number of local competitors for local EMF subunits that share the same ethnic identity. Therefore, the findings of this study extend resource-based view thinking by examining the potential for a resource to be valuable to a subgroup of firms in a local population, and to be costly-to-imitate by other firms external to this subgroup in the local population. Ethnic identity is a home-grown resource for each EMF subunit that shares the same ethnicity, yet we infer from our study that its value to the subunit is local-market specific. That is, the value of collective identity as a resource to an EMF subunit depends on the local ethnic density and that a Hispanic subunit can improve competitive parity in an environment with a high number of Hispanic subunits per capita.
Although EMFs encounter challenges in developed markets, our results imply that "strength in numbers" for EMF subunits can facilitate competitive parity and thus survival. We infer that collaborative advantage contributes to a linear relationship, rather than a non-linear relationship that has been suggested in prior studies (e.g., Mitchell et al. 1994). Collaborative advantage suggests that businesses operate in a network of interdependent relationships developed through collaboration with the objective of deriving mutual benefits (Lado/Boyd/Hanlon 1997). While firms compete directly with each other, they also cooperate to some degree in the sharing of information, or perhaps, by retrieving or diffusing of information from ethnic group members that form a network (Schilling et al. 2003). Therefore, the EMF learns valuable knowledge from other firms in the local market with which it shares the ethnic identity. Thus, ethnic identity serves as a catalyst for and a facilitator of organizational learning. Collaborative advantage is consistent with the establishment and preservation of the collective identity by the Hispanic subunits in the local environment (Melucci 1989, Swaminathan 2001). By establishing collective identity, the local ethnic density increases thereby enhancing the legitimacy of the business practices of EMF subunits in the local environment, especially in the niche market. In turn, the EMF becomes more embedded in the niche market helping them to better serve local Hispanic customers, and to recruit and retain highly-skilled Hispanic banking employees and managers.
We contend that underlying the ability for ethnically similar firms to exploit collaborative advantage is a growing and underserved Hispanic population. Reports from the Federal Deposit Insurance Corporation forecast growth in U.S. retail banking, in particular, very strong growth from the U.S. Hispanic population. Yet the Hispanic banking sector remains a nascent banking market and a large underserved segment of the U.S. Hispanic population does not use banking services (Seper/Sparshott 2007). However, U.S. banks, for the most part have been unable to achieve the customer loyalty and understand the subtle nuances of the Hispanic population, despite the common language. As such, the forecasted growth in Hispanic population and the upside potential of providing banking services to existing Hispanic customers suggests that local competitive intensity is likely to remain low for the foreseeable future, a view that is also consistent with only a positive linear (rather than curvilinear) relationship between local ethnic density and survival. Our highlights a "slippery slope" for executives of Hispanic banks. On the one hand, there is a golden opportunity for Hispanic banks to enter the U.S. market or further expand within the U.S. market in order to deliver financial services to the U.S. Hispanic bank population. On the other hand, these executives must recognize that overexpansion--even of an underserved ethnic customer base--can produce negative consequences.
By adopting an ethnic niche strategy, EMF subunits are able to reduce their liability of foreignness, and may even benefit from foreignness in some local markets. However, there are limits to ethnic identity as a valuable resource in a local market. Ethnic EMF subunits that modify their strategies to include non-ethnic customers are expected to derive fewer benefits from ethnic identity. Inadequate experience and name recognition with non-ethnic customers in addition to more direct competition with resource-rich domestic rivals may adversely affect survival or at least to their ability to grow beyond serving the ethnic market niche.
Our findings reveal curvilinear relationships for both local depth and geographic breadth with EMF subunit survival. Thus, firms that have a large number of subunits in a country may be unable to give adequate attention or resources to the new subunits established. As such, local depth and geographic breadth provide effective support and enhance new subunits' probability of survival until firms reach a point where learning and knowledge transfer dissipates and they no longer manage and provide needed support to the diverse set of subunits effectively. This suggests that managers should try to identify the optimum number of local markets to serve and the number of subunits they can effectively support, coordinate and govern in a country to ensure the highest level of performance.
Although our study was conducted in an international context, the implications suggest that organizations can cooperate with other organizations that share common traits to create and strengthen a collective identity that can be used to compete with other rivals. However the value of a collective identity as a resource may still depend on the local environment in which an organization operates.
In the present study, we sought to extend our knowledge of the resource-based view by developing a theoretical framework to explain how some EMFs can go "knee deep in the Big Muddy" but still survive and even prosper in developed markets. In general, our results suggest the appropriateness of applying resource-based thinking to emerging market multinationals. The value of resources to an EMF subunit depends on its local environment, as well as, the non-local environment of a developed country. Using a sample of banking subunits from Mexico and Central and South America, our work suggests that a group of EMF subunits can share ethnic identity as a resource that is costly to imitate for domestic firms. Moreover, we hypothesized and found that the value of this resource depends on the fit with the host-country local environment.
Our study adapted learning-based research to the international management literature by showing that local depth and geographic breadth both have curvilinear relationships with survival of a Hispanic subunit, which extends prior work by Greve (2000). Previous studies have focused on international presence--the number of countries in which a multinational operates--as a valuable resource to a MNE. We argue that a MNE can transfer knowledge and best practices within its network of in-host-country subunits thereby helping to reduce costs for the local subunit. Local depth and geographic breadth can enhance a parent firm's name recognition, which can also add value to the subunit's ethnic identity in local environments. Yet, the benefits of these local and non-local resources are tempered by diminished learning and knowledge transfer and increased coordination costs at high levels of local depth and geographic breadth. By over-expanding locally and nationally, an EMF can compromise the resource utilization of its subunit. For an EMF --especially an ethnic MNE--we show that local depth and geographic breadth within a developed market can be valuable resources to a given subunit within the host country. We suggest that this finding, which builds on prior work by Mitchell et al (1992, 1994), is especially relevant in service industries where proximity to local customers is critical.
Use of the resource-based view also extends Miller and Eden (2006) by decomposing local density into components. We show that some components of local density indicate cooperative forces rather than competitive forces and thus are potential resources to a sub-population in the local market. With a niche strategy, local density of host-country firms per capita still can have a negative effect on foreign subsidiary performance; however, local ethnic density enhances the likelihood of survival in a developed market. The traditional view of density dependence suggests that local density, especially in mature industries, reflects competitive pressures. Even in a mature industry such as banking, niche strategies can elevate the importance of legitimacy with the local niche population and of the local market conditions in which collaborative advantage can be sustained.
One limitation of our study is that only U.S. subsidiaries of foreign banks are required to provide financial statement data to U.S. regulators so that data on performance for other subunit forms (e.g., branches) were not available. Although some studies have assessed foreign subunit performance using financial performance (Miller/Eden 2006), analysis of EMF subunit survival is also a well-established practice (e.g., Mitchell et al. 1994). Additionally, we do not use a direct measure of ethnicity, but contend that it is an intangible resource based on the bank's home country. According to Yinger, "ethnicity, nationality, and country of origin are often used as synonyms" (1985, p. 157), thus, nationality is an effective proxy. We also assume that the value of ethnicity to each Hispanic subunit is derived from the local density of ethnically similar competitors and customers.
Though beyond the scope of the present study, future research should explore the extent to which domestic firms imitate EMF subunits in order to gain legitimacy with the local ethnic community, and the degree to which such ethnically based strategies are successful. Moreover, if domestic banks imitate the strategies of EMF subunits in the local market, how does this practice influence the legitimacy of the domestic firm with the non-ethnic local community?
The notion of local search in local markets and adjacent markets requires some refinement to consider other forms of distance. We have adapted Greve's (2000) and Baum et al.'s (2000) geographic distance lens to access adjacent markets to distance based on ethnicity. Thus, an adjacent market for an ethnic firm is a non-local ethnic market. We also focused exclusively on ethnic identity from the perspective of Hispanic subunits. An opportunity for future research involves the examination of ethnic identity as a resource for subunits from other countries. For instance, future research should examine the degree to which our findings apply in other ethnic settings (e.g., firms from Asian countries investing in developed markets).
It has become more common for firms from emerging markets to expand into developed markets; nevertheless, survival remains a challenge. The present study takes a first step toward understanding how an emerging market firm can draw upon its home-grown resources and host-country resources to avoid drowning in developed markets.
The authors appreciate the comments and suggestions of Steve Boivie, Mason Carpenter, Andrew Delios, Daniel Indro, Kyle Lewis, Dan Li, Yadong Luo, Jim Westphal, Hong Zhu, and anonymous MIR reviewers.
Manuscript received: July 2007 / Revised: February 2008 / Final revision received: June 2008.
Amel, D. F./Rhodes, S., Strategic Groups in Banking, Review of Economics and Statistics, 70, 4, 1988, pp. 685-689.
Argote, L/Beckman, S./Epple, D., The Persistence and Transfer of Learning in Industrial Settings, Management Science, 36, 2, 1990, pp. 140-154.
Ashforth, B./Mael, F., Social Identity Theory and the Organization, Academy of Management Review, 14, 1, 1989, pp. 20-39.
Barney, J. B., Organizational Culture: Can It Be a Source of Sustained Competitive Advantage?, Academy of Management Review, 11, 3, 1986, pp. 656-665.
Barney, J., Firm Resources and Sustained Competitive Advantage, Journal of Management, 17, 1, 1991, pp. 99-120.
Barney, J./Hansen, M., Trustworthiness as a Source of Competitive Advantage, Strategic Management Journal, 15, Special Issue, 1994, pp. 175-190.
Barnett, W. P./Greve, H./Park, D. Y., An Evolutionary Model of Organizational Performance, Strategic Management Journal, 15, Winter Special Issue, 1994, pp. 11-28.
Barry, F./Gorg, H./Strobl, E., Foreign Direct Investment, Agglomerations and Demonstration Effects: An Empirical Investigation, Review of World Economics/Weltwirtschaftliches Archiv, 139, 2003, pp. 583-600.
Bartel, A., Where Do the New US Immigrants Live?, Journal of Labor Economics, 7, 4, 1989, pp. 371-379.
Bartlett, C./Ghoshal, S., Going Global: Lessons From Late Movers, Harvard Business Review, 78, 2, 2000, pp. 132-142.
Baum, J./Mezias, S., Localize Competition and Organizational Failure in the Manhattan Hotel Industry 1898-1990, Administrative Science Quarterly, 37, 4, 1992, pp. 580-604.
Baum, J./Singh, J., Organizational Niches and the Dynamics of Organizational Mortality, American Journal of Sociology, 100, 2, 1994, pp. 346-380.
Baum, J./Li, S./Usher, J., Making the Next Move: How Experiential and Vicarious Learning Shape the Locations of Chains' Acquisitions, Administrative Science Quarterly, 45, 4, 2000, pp. 766-801.
Berger, A., The Profit-Structure Relationship in Banking: Tests of Market-Power and EfficientStructure Hypotheses, Journal of Money, Credit, and Banking, 27, 2, 1995, pp. 404-431.
Brambor, T./Clark, W. R./Golder, M., Understanding Interaction Models: Improving Analyses, Political Analysis, 14, 1, 2006, pp. 63-82.
Carroll, G./Hannan, M. T., Density Dependence in the Evolution of Populations in Newspaper Organizations, American Sociological Review, 54, 4, 1989, pp. 524-441.
Caves, R., Multinational Enterprise and Economic Analysis, New York: Cambridge University Press 1982.
Chang, S.-J./Park, S., Types of Firms Generating Network Externalities and MNEs' Co-Location Decisions, Strategic Management Journal, 26, 7, 2005, pp. 595-615.
Darr, E./Argote, L./Epple, D., The Acquisition, Transfer and Depreciation of Knowledge in Service Organizations: Productivity in Franchises, Management Science, 41, 11, 1995, pp. 1750-1762.
Dawar, N./Frost, T., Competing With Giants: Survival Strategies for Local Companies in Emerging Market, Harvard Business Review, 77, 2, 1999, pp. 119-129.
Dinnie, K., Country-of-Origin 1965-2004: A Literature Review, Journal of Customer Behaviour, 3, 2, 2004, pp. 165-213.
Douglas, M., How Institutions Think, Syracuse: Syracuse University Press 1986.
Eden, L./Miller, S., Distance Matters, pp. Liability of Foreignness, Institutional Distance and Ownership Strategy, in Hitt M./Cheng, J. (eds.), Advances in International Management, Vol. 16, New York: Elsevier 2004, pp. 187-221.
Epple, D./Argote, L./Devadas, R., Organizational Learning Curves: A Method For Investigating Intra-Plant Transfer of Knowledge Acquired Through Learning By Doing, Organization Science, 2, 1, 1991, pp. 58-70.
Gimeno, J. et al., Explaining the Clustering of International Expansion Moves: A Critical Test in the U.S. Telecommunications Industry, Academy of Management Journal, 48, 2, 2005, pp. 297-320.
Greve, H., Patterns of Market Competition: The Diffusion of a Market Position in Radio Broadcasting, Administrative Science Quarterly, 41, 1, 1996, pp. 29-60.
Greve, H., Managerial Cognition and the Mimetic Adoption of Market Positions: What You See is What You Do, Strategic Management Journal, 19, 10, 1998, pp. 967-988.
Greve, H., Market Niche Entry Decisions: Competition, Learning, and Strategy in Tokyo Banking, 1894-1936, Academy of Management Journal, 43, 5, 2000, pp. 816-836.
Greve, H., An Ecological Theory of Spatial Evolution: Local Density Dependence in Tokyo Banking, 1894-1936, Social Forces, 80, 3, 2002, pp. 847-879.
Hannan, M. T./Freeman, J. H., The Population Ecology of Organizations, American Journal of Sociology, 82, 5, 1977, pp. 929-984.
Heckman, J., Sample Selection Bias as a Specification Error, Econometrica, 47, 1, 1979, pp. 153-161.
Hewstone, M./Rubin, M./Willis, H., Intergroup Bias, Annual Review of Psychology, 53, 1, 2002, pp. 575-604.
Hitt, M. et al., Direct and Moderating Effects of Human Capital On Strategy and Performance in Professional Service Firms: A Resource Based Perspective, Academy of Management Journal, 44, 1, 2001, pp. 13-28.
Hitt, M. et al., The Importance of Resources for the Internationalization of Professional Service Firms: The Good, the Bad and the Ugly, Academy of Management Journal, 49, 6, 2006, pp. 1137-1157.
Hitt, M./Hoskisson, R./Ireland, R. D., A Mid-Range Theory of the Interactive Effects of International and Product Diversification on Innovation and Performance, Journal of Management, 20, 2, 1994, pp. 297-326.
Hitt, M./Hoskisson, R./Kim H., International Diversification: Effects on Innovation and Firm Performanee in Product-Diversified Firms, Academy of Management Journal, 40, 4, 1997, pp. 767-798.
Hoskisson, R./Eden, L./Lau, C-M./Wright, M., Strategy in Emerging Economies, Academy of Management Journal, 43, 3, 2000, pp. 249-267.
Hymer, S., The International Operations of National Firms." A Study of Direct Foreign Investment, Cambridge: MIT Press 1960/1976.
Kane, A./Argote, L./Levine, J., Knowledge Transfer between Groups via Personnel Rotation: Effects of Social Identity and Knowledge Quality, Organizational Behavior and Human Decision Processes, 96, 1, 2005, pp. 56-71.
Khanna, T./Palepu, K., The Right Way to Restructure Conglomerates in Emerging Markets, Harvard Business Review, 77, 4, 1999, pp. 125-134.
Khanna, T./Palepu, K., Emerging Giants: Building World-Class Companies in Developing Countries, Harvard Business Review, 84, 10, 2006, pp. 60-69.
Kostova, T./Zaheer, S., Organizational Legitimacy under Conditions of Complexity: The Case of the Multinational Enterprise, Academy of Management Review, 24, 1, 1999, pp. 64-81.
Lado, A./Boyd, N./Hanlon, S., Competition, Cooperation, and the Search for Economic Rents: A Syncretic Model, Academy of Management Review, 22, 1, 1997, pp. 110-141.
Levinthal, D./March, J., The Myopia of Learning, Strategic Management Journal, 14, Special Issue, 1993, pp. 94-112.
Levitt, B./March, J., Organizational Learning, Annual Review of Sociology, 14, 1988, pp. 319-340.
Makino, S./Lau, C-M./Yeh, R-S., Asset-Exploitation versus Asset-Seeking: Implications for Location Choice of Foreign Direct Investment from Newly Industrialized Economies, Journal of International Business Studies, 33, 3, 2002, pp. 403-421.
March, J., Exploration and Exploitation in Organizational Learning, Organization Science, 2, 1, 1991, pp. 71-87.
Medina, H, Compounding Opportunity, Hispanic Bank Marketing, www.hispanicbankmarketing.com., 2007
Melucci, A., Nomads of the Present. Social Movements and Individual Needs in Contemporary Society, Philadelphia: Temple University Press 1989.
Mezias, J., Identifying Liability of Foreignness and Strategies to Minimize their Effects: The Case of Labor Lawsuit Judgments in the United States, Strategic Management Journal, 23, 3, 2002, pp. 229-344.
Miller, S. R./Eden, L., Local Density and Foreign Subsidiary Performance, Academy of Management Journal, 49, 2, 2006, pp. 341-355.
Mitchell, W./Shaver, J. M./Yeung, B., Getting there in a Global Industry: Impacts on Performance of Changing International Presence, Strategic Management Journal, 13, 6, 1992, pp. 419-432.
Mitchell, W./Shaver, J. M./Yeung, B., Performance Following Changes of International Presence in Domestic and Transition Industries, Journal of International Business Studies, 24, 4, 1993 pp. 647-669.
Mitchell, W./Shaver, J. M./Yeung, B., Foreign Entrant Survival and Foreign Market Share: Canadian Companies' Experience in United States Medical Sector Markets, Strategic Management Journal, 15, 7, 1994, pp. 555-567.
Mitra, D./Golder, P., Whose Culture Matters? Near-Market Knowledge and its Impact on Foreign Market Entry Timing, Journal of Marketing Research, 39, 3, 2002, pp. 350-365.
Morita, J./Lee, T./Mowday, R., The Regression-Analog to Survival Analysis: A Selected Application to Turnover Research, Academy of Management Journal, 36, 6, 1993, pp. 1430-1463.
Nachum, L., Liability of Foreignness in Global Competition? Financial Service Affiliates in the City of London, Strategic Management Journal, 24, 12, 2003, pp. 1187-1208.
Nagel, J., Constructing Ethnicity: Creating and Recreating Ethnic Identity and Culture, Social Problems, 41, 1, 1994, pp. 152-176.
Office of Management and Budget, Federal Registry: Standards For Defining Metropolitan and Micropolitan Statistical Areas, National Archives and Records Administration, Washington DC, 2000.
Ruef, M./Scott, W. R., A Multidimensional Model of Organizational Legitimacy: Hospital Survival in Changing Institutional Environments, Administrative Science Quarterly, 43, 4, 1998, pp. 877-904.
Rugman, A., Inside the Multinationals. The Economics of Internal Markets, New York: Columbia University Press 1981.
Saparito, P./Chen, C./Sapienza, H., The Role of Relational Trust in Bank-Small Firm Relationships, Academy of Management Journal, 47, 3, 2004, pp. 400-410.
Schilling, M. et al., Learning by Doing Something Else: Variation, Relatedness, and the Learning Curve, Management Science, 49, 1, 2003, pp. 39-56.
Seper, J./Sparshott, J., Bank on Hispanics, Hispanic Bank Marketing, www.hispanicbankmarketing.com, 2007
Shaver, J. M., Accounting for Endogeneity when Assessing Strategic Performance: Does Entry Mode Choice affect FDI Survival? Management Science, 44, 4, 1998, pp. 571-585.
Shaver, J. M./Mitchell, W./Yeung, B., The Effect of Own-Firm and Other-Firm Experience on Foreign Direct Investment Survival in the United States, 1987-92, Strategic Management Journal, 18, 10, 1997, pp. 811-24.
Sirmon, D./Hitt, M./Ireland, R. D., Managing Firm Resources in Dynamic Environments to Create Value: Looking Inside the Black Box, Academy of Management Review, 32, 1, 2007, pp. 273-292.
Sirmon, D./Gove, S./Hitt, M., Resource Management in Dyadic Competitive Rivalry: The Effects of Resource Bundling and Deployment, Academy of Management Journal, 51, 5, in press, 2008.
Swaminathan, A., Resource Partitioning and the Evolution of Specialist Organizations: The Role of Location and Identity in the U.S. Wine Industry, Academy of Management Journal, 44, 6, 2001, pp. 1169-1185.
Thomas, D., Eden, L., Hitt, M., Miller, S.R., Experience of Emerging Market Firms: The Role of Cognitive Bias in Developed Market Entry and Survival, Management International Review, 47, 6, 2007 pp. 845-867.
Tsai, W., Social Structure of "Coopetition" Within a Multiunit Organization: Coordination, Competition, and Intraorganizational Knowledge Sharing, Organization Science, 13, 2, 2002, pp. 179-190.
Uzzi, B., Social Structure and Competition in Inter-Firm Networks: The Paradox of Embeddedness, Administrative Science Quarterly, 42, 1, 1997, pp. 35-67.
Van Den Poel, D./Lariviere, B., Customer Attrition Analysis for Financial Services using Proportional Hazard Models, European Journal of Operational Research, 157, 1, 2004, pp. 196-217.
Wells, L. T., Third World Multinationals: The Rise of Foreign Investment from Developing Countries, Cambridge: MIT Press 1983.
Wendt, A., Collective Identity Formation and the International State, The American Political Science Review, 88, 2, 1994, pp. 384-396.
Yinger, J. M., Ethnicity, Annual Review of Sociology, 1, 2, 1985, pp. 151-180.
Zaheer, S., Overcoming the Liability of Foreignness, Academy of Management Journal, 38, 2, 1995, pp. 341-63.
Zorn, C. J. W., Generalized Estimating Equation Models for Correlated Data: A Review with Applications, American Journal of Political Science, 45, 3, 2001, pp. 470-490.
(1) We thank an anonymous reviewer for this suggestion.
Associate Professor Stewart R. Miller ([mail])
School of Management, University of Texas, San Antonio, USA.
Associate Professor Douglas E. Thomas
Anderson School of Management, University of New Mexico, Albuquerque, USA.
Professor Lorraine Eden--Professor Michael Hitt
Mays Business School, Texas A & M University, College Station, USA.
Table 1. Summary Statistics Variable Mean S.D. Min Max 1 Parent Age 83.42 44.97 7 195 2 Parent Size ($mill) 15.70 24.80 0.14 436 3 Parent Performance 1.19 2.91 -48.00 21.59 4 Parent Leverage 9.41 6.41 -87.5 67.49 5 Edge Act 0.05 0.22 0 1 6 Rep. Office 0.07 0.26 0 1 7 Branch 0.27 0.44 0 1 8 Agency 0.49 0.50 0 1 9 Acquisition 0.08 0.28 0 1 10 Home Real GDP 2.34 3.94 -13.38 12.82 Growth Lagged 11 Local Domestic 0.26 0.11 0.11 0.47 Density (per capita, 1000 population) 12 Local Ethnic 0.02 0.01 0 0.034 Density (per capita, 1000 population) 13 Local Depth 1.25 0.55 1 4 14 Geographic 1.14 1.48 0 5 Breadth Variable 1 2 3 4 5 1 Parent Age 2 Parent Size ($mill) 0.57 3 Parent Performance -0.04 -0.12 4 Parent Leverage -0.02 -0.19 0.39 5 Edge Act -0.08 -0.14 0.01 0.06 6 Rep. Office 0.14 0.21 -0.06 -0.05 -0.06 7 Branch 0.06 0.06 -0.11 -0.01 -0.14 8 Agency -0.05 -0.09 0.07 -0.04 -0.23 9 Acquisition -0.10 -0.08 -0.06 0.05 -0.01 10 Home Real GDP -0.07 -0.03 0.08 -0.04 0.00 Growth Lagged 11 Local Domestic 0.13 0.08 0.01 0.10 -0.21 Density (per capita, 1000 population) 12 Local Ethnic -0.18 -0.36 0.08 0.18 -0.04 Density (per capita, 1000 population) 13 Local Depth 0.05 -0.03 0.16 0.04 -0.11 14 Geographic 0.53 0.59 -0.04 -0.15 0.00 Breadth Variable 6 7 8 9 10 1 Parent Age 2 Parent Size ($mill) 3 Parent Performance 4 Parent Leverage 5 Edge Act 6 Rep. Office 7 Branch -0.17 8 Agency -0.27 -0.59 9 Acquisition 0.01 0.04 -0.02 10 Home Real GDP 0.01 -0.05 0.00 -0.06 Growth Lagged 11 Local Domestic -0.21 0.56 -0.24 0.05 -0.02 Density (per capita, 1000 population) 12 Local Ethnic -0.36 0.37 -0.08 -0.03 0.02 Density (per capita, 1000 population) 13 Local Depth -0.01 -0.18 -0.03 -0.1 0.07 14 Geographic 0.28 -0.21 0.08 -0.21 -0.02 Breadth Variable 11 12 13 1 Parent Age 2 Parent Size ($mill) 3 Parent Performance 4 Parent Leverage 5 Edge Act 6 Rep. Office 7 Branch 8 Agency 9 Acquisition 10 Home Real GDP Growth Lagged 11 Local Domestic Density (per capita, 1000 population) 12 Local Ethnic 0.58 Density (per capita, 1000 population) 13 Local Depth -0.02 -0.11 14 Geographic -0.24 -0.44 -0.01 Breadth The cutoff for significance at the 5% level is the absolute value of 0.06. Table 2. Survival of EMF Subunits Model 1 Model 2 Model 3 Parent Age 0.02 0.04 0.02 (0.29) (0.35) (0.41) Parent Size 0.04 -0.22 -0.30 (0.21) (0.26) (0.26) Parent ROAA 0.01 -0.03 -0.06 (0.04) (0.03) (0.04) Parent Leverage -0.03 -0.01 -0.01 (0.02) (0.02) (0.03) Edge Act -0.07 -0.84 -0.86 (0.59) (0.75) (0.79) Rep. Office 0.81 -0.60 -0.99 (0.50) (0.70) (0.76) Branch -1.03 -3.78 ** -4.44 ** (0.65) (1.00) (1.12) Agency -0.791 -2.39 ** -2.85 ** (0.40) (0.62) (0.57) Acquisition 1.55 ** 0.98* 1.12* (0.38) (0.43) (0.50) Home Real GDP 0.04 -0.01 -0.01 Growth (Lagged) (0.07) (0.06) (0.06) Inverse Mills Ratio -2.41 ** 0.30 0.13 (0.76) (1.01) (1.08) Local Domestic Density -2.721 -1.65 -4.92 (per capita, 1000 pop) (1.509) (3.51) (3.49) Local Ethnic Density -156.89 ** -145.47 ** (per capita, 1000 pop) (52.01) (50.94) Local Depth -1.55 ** -1.46 ** (0.56) (0.55) Geographic Breadth -1.29 ** -1.71 ** (0.26) (0.24) Local Depth Squared Geographic Breadth Squared 0.38 ** (0.08) Observations 1066 1066 1066 Partial Log-Likelihood -113.97 -84.43 -75.86 Degrees of Freedom 12 15 16 Model Chi-squared 66.09 *** 93.00 *** 97.32 *** Model 4 Model 5 Parent Age 0.08 0.06 (0.35) (0.41) Parent Size -0.17 -0.26 (0.27) (0.26) Parent ROAA -0.04 -0.06 (0.03) (0.04) Parent Leverage -0.01 -0.01 (0.02) (0.03) Edge Act -0.73 -0.80 (0.74) (0.76) Rep. Office -0.31 -0.75 (0.67) (0.72) Branch -3.48 ** -4.19 ** (0.95) (1.06) Agency -2.14 ** -2.64 ** (0.58) (0.55) Acquisition 0.91* 1.04* (0.42) (0.50) Home Real GDP -0.01 0.00 Growth (Lagged) (0.06) (0.05) Inverse Mills Ratio -0.02 -0.16 (0.96) (1.04) Local Domestic Density -1.36 -4.56 (per capita, 1000 pop) (3.59) (3.55) Local Ethnic Density -161.67 ** -149.73 ** (per capita, 1000 pop) (53.66) (51.86) Local Depth -2.60 ** -2.37 ** (0.93) (0.91) Geographic Breadth -1.31 ** -1.70 ** (0.26) (0.23) Local Depth Squared 0.80 * 0.67 * (0.35) (0.34) Geographic Breadth Squared 0.37 ** (0.07) Observations 1066 1066 Partial Log-Likelihood -83.06 -74.92 Degrees of Freedom 16 17 Model Chi-squared 96.57 *** 95.32 *** Model stratified by home country. Robust standard errors (shown in parentheses) were adjusted for clustering by subunit. ([dagger]) indicates significant at 10%; * 5%; ** 1%; *** 0.1% levels.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||RESEARCH ARTICLE|
|Author:||Miller, Stewart R.; Thomas, Douglas E.; Eden, Lorraine; Hitt, Michael|
|Publication:||Management International Review|
|Date:||Dec 1, 2008|
|Previous Article:||Domsch, M. E./Lidokhover, T.: Human Resource Management in Russia, Ashgate: Aldershot 2007.|
|Next Article:||Market size, legal institutions, and international diversification strategies: implications for the performance of multinational firms.|