Management and ownership control associated with foreign operation methods

The term foreign operation method is used in the international business literature to refer to the modes that involve outward transactions (e.g., exports, license, franchising) or FDI (e.g., joint venture) at entry time and/or years after entry time (e.g., Benito, Pedersen, & Petersen, Reference Benito, Pedersen and Petersen1999; Pedersen, Petersen, & Benito, Reference Pedersen, Petersen and Benito2002; Petersen & Welch, Reference Petersen and Welch2002).

The operation methods for an FDI project can be classified into four categories as a result of combining establishment modes (greenfield vs. acquisition) and ownership modes (wholly owned vs. joint venture): joint venture greenfield, partial acquisition, wholly owned greenfield and full acquisition (Brouthers & Hennart, Reference Brouthers and Hennart2007). However, and according to Hennart (Reference Hennart1988), both partial acquisition and joint venture greenfield should be categorized as joint venture, since they are joint hierarchies. Thus, the operation method can be classified into three categories: joint venture, wholly owned acquisition and wholly owned greenfield (Kogut & Singh, Reference Kogut and Singh1988; Nitsch, Beamish, & Makino, Reference Nitsch, Beamish and Makino1996; Anand & Delios, Reference Anand and Delios2002; Meyer et al., Reference Meyer, Estrin, Bhaumik and Peng2009). This study adopts these same criteria and offers a new variable that distinguishes different levels of management and ownership control associated with these three categories.

As Anderson and Gatignon (Reference Anderson and Gatignon1986: 3) state: ‘Control (the ability to influence systems, methods and decisions) has a critical impact on the future of a foreign enterprise. Without this control, a firm finds it more difficult to coordinate actions, carry out strategies, revise strategies, and resolve the disputes that invariably arise when two parties to a contract pursue their own interests.’ Taking this control implies that the MNC must assume both commitment of resources (ownership control) and responsibility for decision-making (management control) because ownership by itself is not necessarily sufficient to guarantee complete managerial control (Nitsch, Beamish, & Makino, Reference Nitsch, Beamish and Makino1996; Jakobsen & Meyer, Reference Jakobsen and Meyer2008). For instance, in some cases the MNC takes over management control and engages directly in the strategic management of the firm, and in other cases it acts more like a financial investor or venture capitalist, advising and possibly indirectly influencing the management, but not taking over direct control. Indeed, Child and Yan (Reference Child and Yan1999) find that financial and nonfinancial resources have different impacts since they provide different bases of control. According to Woodcock, Beamish, and Makino (Reference Woodcock, Beamish and Makino1994) and Nitsch, Beamish, and Makino (Reference Nitsch, Beamish and Makino1996), joint venture, acquisition and greenfield provide different levels of management and ownership control. These operation methods will also imply taking certain forms of risk and at the same time will constitute a way of obtaining different levels of potential returns (Anderson & Gatignon, Reference Anderson and Gatignon1986) (see Table 1).

Researchers often use the resource commitment to distinguish between the choice of a joint venture or a wholly owned subsidiary when working in a foreign market. The level of ownership control is used as an approximate measure of the resource commitment. The higher the level of ownership for a method (i.e., greenfield and acquisition vs. joint venture), the greater will be the resource commitment (Canabal & White, Reference Canabal and White2008) and the return and risk (Anderson & Gatignon, Reference Anderson and Gatignon1986). A firm that does not have all the resources necessary to operate abroad or sufficient knowledge of the host market and that is prepared to risk disseminating its own assets in order to obtain these resources, could choose to work in a particular market through a joint venture (Woodcock, Beamish, & Makino, Reference Woodcock, Beamish and Makino1994). According to Chan and Makino (Reference Chan and Makino2007), another reason to operate abroad with a joint venture is the necessity to conform to institutional pressures obtaining legitimacy (Brown, Reference Brown2012). In this case, the MNC will exchange a percentage of the equity ownership of its foreign subsidiaries for an important resource, legitimacy in the host country (e.g., conforming to a host country regulation related to foreign firms’ entry), thus dealing with the liability of foreignness (Eden & Miller, Reference Eden and Miller2004). This operation method will reduce resource commitment and some forms of risk but often at the expense of returns (Anderson & Gatignon, Reference Anderson and Gatignon1986).

With regard to the management control, the multiple ownership agreements that define a joint venture mean that the firm will face high costs in the initial negotiations to set up a relationship of control between the partners, as well as subsequent costs in managing the relationship (Beamish & Banks, Reference Beamish and Banks1987). Both costs are high because the parties in the relationship can behave opportunistically, putting the success of the operations at risk, since no control mechanism is as effective as ownership. In contrast, wholly owned methods provide firms with greater management control over their foreign operations and their valuable resources (Brouthers, Brouthers, & Werner, Reference Brouthers, Brouthers and Werner2008) because more integrated methods give the firm the authority to direct operations (Anderson & Gatignon, Reference Anderson and Gatignon1986).

Likewise, Woodcock, Beamish, and Makino (Reference Woodcock, Beamish and Makino1994) argue that the concept of resource commitment can be used to differentiate between greenfield and acquisition. Firms that have the knowledge resources available will use the greenfield method, and those lacking these resources will use acquisition, employing financial resources to acquire such knowledge resources previously embedded in another organization (Meyer et al., Reference Meyer, Estrin, Bhaumik and Peng2009). This way, the MNC will control the financial resources used to acquire the ownership but not all the knowledge resources, slightly decreasing its management control. Both greenfield and acquisition avoid the risk of disseminating a firm’s key assets since it reasons that it cannot risk its own resources in a joint venture to obtain the resources it lacks (Delios & Beamish, Reference Delios and Beamish1999). Other reasons to operate abroad with a greenfield or an acquisition are those related to the need to conform to different institutional pressures. The greenfield will give the MNC the opportunity to obtain legitimacy in the eyes of its internal constituents by replicating internal structures established in the past. In the case of the acquisition, the MNC could exchange part of its ownership control, in terms of the tacit knowledge owned by the acquired firm, for credibility among actors in the host country, thus gaining legitimacy.

The greenfield involves a greater level of investment risk as a consequence of the uncertainty associated with the new business project in the host country, with the MNC making the full investment. The acquisition, on the other hand, gives the MNC the security associated with a business project already operating in the host country, thus providing lower risk than the greenfield, but higher risk than a joint venture where the investment is shared with a local partner.

With regard to the levels of management control, Woodcock, Beamish, and Makino (Reference Woodcock, Beamish and Makino1994) maintain that acquisitions involve extra control costs compared with greenfields. The information asymmetry created by the resource deficiency may limit the firm’s ability to understand, and indeed control, the companies it has acquired. When a firm acquires a local company it absorbs a group of employees with their own routines and culture. Integrating these employees will be difficult, particularly if cultural differences exist between the two firms, perhaps because they come from different countries (Hennart & Reddy, Reference Hennart and Reddy1997). Indeed, cultural distance (Yiu & Makino, Reference Yiu and Makino2002), as an important component of the institutional environment, reflects differences between home and host countries (Yiu & Makino, Reference Yiu and Makino2002; Eden & Miller, Reference Eden and Miller2004; Demirbag, Glaister, & Tatogly, Reference Demirbag, Glaister and Tatogly2007). This cultural distance will hinder the parent company’s adjustment to the conditions of the host country (Harzing, Reference Harzing2002). On the other hand, if a firm chooses to work in a foreign market using a joint venture, its level of management control is lower since it must be shared with local partners. These partners tend to be more knowledgeable about normative and regulative institutions such as the relationships with the administration, behavior of the demand, the culture, and so on. Thus, they will be able to behave opportunistically, causing the investor to face high costs to control the operations (Beamish & Banks, Reference Beamish and Banks1987). These high costs of management control, which stem from lower levels of ownership control within a shared investment, will reduce the potential returns for the MNC that chooses a joint venture. For the MNC that chooses a greenfield, the high management and ownership control will increase their potential returns as they rely on their own resources to invest in the host country. When the firm turns to the acquisition as a foreign operation method, it will not have ownership control of all the resources (e.g., knowledge embedded in acquired local firms) or complete management control over the operations. Thus, it will expect a higher level of potential returns than for the joint venture but lower than for the greenfield.

On the basis of the above considerations, we suggest that management and ownership control will vary according to foreign operation method choices. The highest level of control exists when the firm decides to operate abroad by a greenfield. At the opposite extreme, joint venture agreements involve the lowest levels of management and ownership control. Finally, the management and ownership control for the acquisition method is between the other two methods.

Institutional influences on foreign operation method choice

From the perspective of new institutionalism, it can be said that firms will choose a specific level of management and ownership control for which they find institutional support (Meyer & Nguyen, Reference Meyer and Nguyen2005). In the case of the regulative institutions, if they guarantee a high quality host country institutional environment – for example, trained human resources, fiscal incentives, suitable infrastructures, among other factors – the MNC will perceive a favorable setting that facilitates transactions without incurring in a high institutional risk (Demirbag, Glaister, & Tatogly, Reference Demirbag, Glaister and Tatogly2008; Chen et al., Reference Chen, Yang, Hsu and Wang2009). The MNC that finds such institutional support will use integrated operation methods that involve a greater degree of management and ownership control (Brouthers, Brouthers & Werner, Reference Brouthers, Brouthers and Werner1999; Brouthers, Reference Brouthers2002). Given the potential profits expected from the investment, if the firm does not exploit the business itself by using its key resources, it will face high opportunity costs (Brouthers, Reference Brouthers2002). Thus, direct investment through a greenfield offers the firm the opportunity to obtain positive results and to develop competitive advantages at the international level, without facing high management and ownership control costs (Woodcock, Beamish, & Makino, Reference Woodcock, Beamish and Makino1994).

When the institutional environment is relatively favorable, but not quite favorable enough for the MNC to accept the highest level of uncertainty associated with embarking on a solo venture in a host country without local partnership, it could choose to opt for an acquisition. In this case, there is less risk due to the security associated with a business project that is already in operation. However, the probability of developing new competitive advantages based on the MNC’s own competencies will be lower with an acquisition, the firm will face higher management and ownership control costs, and its potential profits could be lower.

In contrast – except in situations regarding specific assets which the MNC intends to control abroad (Anderson & Gatignon, Reference Anderson and Gatignon1986) – when the quality of a host country is not as favorable and the potential outcome of the investment is subject to environmental uncertainty, the firm may be reluctant to use operation methods that increase management and ownership control, and may prefer less integrated methods like the joint venture, as Kao, Kuo, and Chang (Reference Kao, Kuo and Chang2013) found for the context of family firms. The firm will then share both the risk and the potential profits of the investment with a local partner, despite the higher management and ownership control costs this would imply.

Although international experience conditions managers’ attention to different regulative institutions when making decisions regarding international business (Santos-Alvarez & Garcia-Merino, Reference Santos-Alvarez and Garcia-Merino2012), Anderson and Gatignon (Reference Anderson and Gatignon1986) state that the MNC will adopt specific entry mode decisions irrespective of experience. One the one hand, firms with a high level of international experience could use operation methods with a high level of control, in so far as the knowledge resources acquired by experience are available for this investment to be undertaken. On the other hand, firms with a low level of international experience could also use operation methods that allow a high level of control in order to, for example, place acculturated personnel in key positions in the subsidiaries, and thus guarantee management control of the investment. According to Anderson and Gatignon (Reference Anderson and Gatignon1986), the firm will be ready to delegate control only when it has greater international experience, reducing in that case the ownership level in its operations abroad. Thus, it can be established that, irrespective of experience:

Institutional theory regards uncertainty as something to be controlled by strategic actions (Chen et al., Reference Chen, Yang, Hsu and Wang2009). In conditions of institutional heterogeneity between host and home country, MNCs can consider the actions undertaken by other MNCs which have previously entered the country with the aim of decreasing the risk and uncertainty in the decision making related to FDI. Specifically, a firm will pay more attention to other MNCs from the same home country than to those from other countries (Chan & Makino, Reference Chan and Makino2007; Chen et al., Reference Chen, Yang, Hsu and Wang2009), because the former faced the same level of institutional distance as it does. In this way, the firm will gain cognitive legitimacy by imitating decisions previously taken by a large number of organizations (Huang & Sternquist, Reference Huang and Sternquist2007). Lu (Reference Lu2002) finds that this mimetism, both in the firm’s own sector and beyond, is determinant in the operation method choice when the firm establishes a subsidiary abroad.

Years after the firm’s entry to the host country, the perception of uncertainty associated with the institutional gap between the firm’s host and home countries will be reduced. In this context, the MNC can probably attend to other institutional factors (e.g., regulative institutions) in order to identify the most suitable level of management and ownership control for its operation in the host country. However, and according to Pedersen, Petersen, and Benito (Reference Pedersen, Petersen and Benito2002), MNCs entering a foreign country are unlikely to change their operation method with the passing of time. Thus, if an MNC mainly pursues cognitive legitimacy, it is unlikely to change the level of management and ownership control it initially chose since the other MNCs it refers to do not change either. Thus, we can expand Lu’s (Reference Lu2002) finding about the importance of mimetic behavior in explaining the operation method choice both at entry time, as Lu finds, and also years after the time of entry.

The internal institutional environment can also determine the foreign operation method choice. Specifically, the intra-organizational relationships in the MNC include the flows of capital, products and knowledge (Gupta & Govindrajan, Reference Gupta and Govindrajan1991) and human resources (Boyacigiller, Reference Boyacigiller1990) between the subsidiaries. Research has shown that the exchange of functional activities between the parent company and its subsidiaries seeks to achieve a fit and similarity between the organizational structures, systems and practices used by the different subsidiaries in different countries. In this context, the level of strategic autonomy conceded to the subsidiaries reflects the parent company’s intention to reinforce its management practices (DiMaggio & Powell, Reference DiMaggio and Powell1983). Thus, when the parent company wishes to preserve its internally institutionalized practices it increases the pressure to make internal isomorphism dominant (Davis, Desai, & Francis, Reference Davis, Desai and Francis2000; Chan & Makino, Reference Chan and Makino2007; Drogendijk & Andersson, Reference Drogendijk and Andersson2012), hence exerting a greater control and reducing the autonomy of its subsidiaries.

According to Davis, Desai, and Francis (Reference Davis, Desai and Francis2000), a parent company that wants to apply management practices that facilitate intra-firm interdependence will prefer a wholly owned operation method, which allows it to maintain a high level of convergence in the internally institutionalized structures, systems and practices. Organizational replication is a much more natural and efficient process when the firm establishes its own greenfield than when it acquires an already-existing company. This is because created subsidiaries are initially conceived and later developed with the aim that they will inherit the specific advantages of the parent company (Harzing, Reference Harzing2002). But if for various reasons the multinational is not particularly interested in its own practices prevailing, the CEO will not feel internal pressure to choose a method involving a high level of management and ownership control. These arguments lead to the following hypothesis:

If the MNC wishes to preserve its internal institutions, it will transfer the knowledge developed within its network (Gupta & Govindrajan, Reference Gupta and Govindrajan1991). This transfer and integration of knowledge implies the transfer of a set of routines and procedures (Martin & Salomon, Reference Martin and Salomon2003), as well as adapting them to the local context. Thus, the foreign operation methods characterized by different levels of management and ownership control are alternative options the firm can choose for the international transfer of knowledge between the parent company and its subsidiaries (Malhotra, Reference Malhotra2003). Full ownership facilitates efficient transfer of knowledge (Martin & Salomon, Reference Martin and Salomon2003). This operation method, compared with joint venture or other alliances, better satisfies the need to protect the knowledge transferred, and hence reduces the probability of disseminating the firm’s knowledge-based advantages (Hennart & Park, Reference Hennart and Park1993; Belderbos, Reference Belderbos2003; Malhotra, Reference Malhotra2003; Martin & Salomon, Reference Martin and Salomon2003). Likewise, these firms will prefer greenfield to acquisition since acquired firms will bring in routines and new knowledge resources that could make integration of the internally established practices that the multinational intends to transfer more difficult (Hennart & Park, Reference Hennart and Park1993; Belderbos, Reference Belderbos2003). The above leads to the following hypothesis:

Additionally, human resource management and, in particular, the flow of employees (Boyacigiller, Reference Boyacigiller1990; Harzing, Reference Harzing2002; Paik & Sohn, Reference Paik and Sohn2004) are basic tools for replicating organizational systems and practices in multinationals. The literature notes a number of advantages of putting expatriate employees in subsidiaries compared with using only local staff (Paik & Sohn, Reference Paik and Sohn2004). Expatriate employees have a greater understanding of the corporate priorities and strategies than local staff, are more accepting of the rules established by the parent company, and more committed to corporate goals. Consequently, and following Rosenzweigh and Singh (Reference Rosenzweig and Singh1991), expatriates are ideally placed to replicate the structures and procedures established by the parent company. They facilitate the transfer of strategic organizational practices not only between the parent and the subsidiary, but also between subsidiaries. Expatriates from the parent company are subject to dual levels of organizational commitment and identification, which puts them in an ideal position to understand the dynamics of institutional duality (Hillman & Wan, Reference Hillman and Wan2005). They can identify with both the unique objectives of the subsidiary and also with the overall objectives of the MNC, and are therefore in a better position to understand how to align the interests of both parties. The above could explain why authors associate expatriation with internal isomorphism (Harzing, Reference Harzing2002).

However, expatriate parent company nationals can be costly, in terms of monetary costs and in terms of legitimacy losses for the subsidiary in the local environment (Harzing, Reference Harzing2002). Thus, MNCs have to balance the pressure to employ host country nationals for legitimacy in the local environment against the transference of MNC practices through expatriates to seek internal legitimacy.

With regard to the types of operation methods for the foreign investments, those with the highest level of management and ownership control will create the ideal environment for expatriates to replicate the parent company’s procedures. When the MNC looks for internal legitimacy, and therefore makes the decision to transfer practices through expatriates, it chooses to operate with a more integrated operation method. Expatriates may encounter more obstacles to doing the same in acquired companies or joint ventures because they will have to transfer strategic organizational practices to a company that is already operating or to negotiate such practices with their new partners in a joint venture. Indeed, such operation methods are chosen by MNCs when they need to exchange ownership for legitimacy in local markets. Thus, it is more difficult to transfer the routines and organizational practices and also to align subsidiary and parent interests by sending expatriates to these foreign subsidiaries as this can create a sense of hostility (Colakoglu, Tarique, & Caligiuri, Reference Colakoglu, Tarique and Caligiuri2009), therefore assuming a loss of legitimacy for the company in the local environment.

Research context

In order to test the above hypotheses we carried out an empirical analysis of the current level of management and ownership control of the first subsidiary German companies set up in Spain. Specifically, more than 1,000 German firms have invested in Spain over the past century. Germany is consequently one of the five main investor countries – after the United Kingdom, France, Italy and the Netherlands – that provided more than 67.40% of total FDI inflows to Spain between 1993 and September 2011 (Ministry of Economics and Competitiveness, 2011). Although in 2004, the year this research was carried out, Germany was seventh in the ranking of investor countries in Spain, in 2011 it became the fifth biggest investor in this country (with an investment of 0.77 billion euros), after the United Kingdom, France, the United States and Luxembourg.

Primary information from the CEOs of the parent companies in Germany was collected. This is particularly relevant in the current research involving firms originating from the same country that have established operations in the same foreign location, since corporate decisions will depend on managers’ perceptions about the importance of Spanish institutional factors. According to these perceptions, it is possible to adopt different decisions with the same institutions (Demirbag, Glaister, & Tatogly, Reference Demirbag, Glaister and Tatogly2008).

Population

The population in this study consisted of German multinational firms that had direct investments in Spain in 2004, regardless of when they entered this country. Specifically, the unit of analysis is the first subsidiary established by those German MNCs that were still operating in Spain at the moment of the empirical study. Those subsidiaries satisfied the condition of having at least five employees in 2004 operating in any industrial or service sector. This population was identified by collecting information from the Chamber of International Commerce in Germany and the German Chamber of Commerce for Spain. The original population consisted of 926 German MNCs that had made at least one investment in Spain.

Data collection and sample representativeness

The empirical evidence was obtained by using a self-administered, structured questionnaire sent by e-mail and post – and sometimes re-sent by fax – to the CEOs of all the German MNCs in the population. The process of collecting the information was carried out in July 2004. Of the 926 CEOs surveyed, 121 responded to the questionnaire, which represents a real response rate of 13.1%, similar to or higher than that of earlier research involving top managers (see Li, Bingham, & Umphress, Reference Li, Bingham and Umphress2007; García-Cabrera & García-Soto, Reference García-Cabrera and García-Soto2011). In all, 12 responses were rejected in order to maintain internal consistency, leaving a final sample of 109 MNCs with a sampling error of ±6.36%.

With regard to the representativeness of the sample, the mean subsidiary size (in terms of number of employees) in the population of German MNCs with FDI in Spain is 309.1 employees. The mean subsidiary size in the sample of 109 participating MNCs is slightly lower (236.3 employees). The proportion of subsidiaries from industrial sectors in the population is 52.4%, while this proportion rises to 67.9% in the sample and includes industries such as machine manufacturing (14.7%), electronic equipment manufacturing (11%) and vehicle manufacturing (10.1%). The reverse happens for the service sectors. Specifically, the proportion of subsidiaries from service sectors in the population is 47.6%, while this proportion drops to 32.1% in the sample and includes services such as ICT (10.1%), and banking, insurance and consulting (7.3%).

Measures

With regard to the level of management and ownership control, we asked the CEOs about the operation method (joint venture, acquisition, or greenfield) their firms had used to enter the Spanish market and the operation method they were using when the study was carried out. Thus, it could be determined if the operation method used by the MNC had changed at all. In this study the dependent variable used – management and ownership control – allows us to jointly analyze the three foreign operation methods through a single simultaneous equations model.

For the regulative institutions, we measure the institutional quality of the host country in terms of the political-legal and economic incentives to foreign investment. For the former we measured how attractive the investment support, interest rates, and fiscal pressure were for the firm’s operations (Demirbag, Glaister, & Tatogly, Reference Demirbag, Glaister and Tatogly2007). For the economic institutions, we measured the availability and quality of both the human resources and the physical infrastructures (Chan, Isobe, & Makino, Reference Chan, Isobe and Makino2008). Thus, the CEOs assessed the educational level, professional and multilingual training, productivity of human resources, cost of trained human resources and the public administration’s commitment to training. As for the physical infrastructures: terrestrial and airport infrastructures, the health service, security and climate were all measured. All these items were adapted from Galán, González-Benito, and Zúñiga-Vicente’s (Reference Galán, González-Benito and Zúñiga-Vicente2007) study of location factors. We measured how favorable all these aspects were for the firm’s operations at the moment of carrying out the study by using a 5-point Likert scale from 1=‘very favorable’ to 5=‘very unfavorable’ (reverse score).

For the isomorphic pressures of the internal institutional environment, three questions regarding the following were included in the questionnaire: (1) search for fit in the internal structures; (2) firm’s desire to transfer know-how to the subsidiary; and (3) firm’s intention to seek internal legitimacy through managers’ expatriation. In turn, external isomorphic pressures were measured by using six variables specially designed for this work. With that aim, data provided by the firms about their operation method at entry time and the date of their entry in Spain (Lu, Reference Lu2002) were used. These operation methods at entry time had not been changed throughout all the years the firms had been established in Spain. Thus, these firms’ behavior, according to Pedersen, Petersen, and Benito (Reference Pedersen, Petersen and Benito2002) findings, offers conclusive proof regarding the low probability of an MNC changing its initial foreign operation method. More specifically, we measured general joint venture, acquisition, and greenfield mimetism by the proportion of sampled German firms that had entered Spain with a joint venture, acquisition or greenfield, respectively, before the respondent firm itself entered Spain. In turn, we measured sector joint venture, acquisition and greenfield mimetism using the percentage of German firms from the same sector as the firm that had operated in Spain with a joint venture, acquisition or greenfield, respectively, before the firm itself entered Spain. All these variables define the external cognitive institutional environment.

Finally, as a control variable we measure the MNC size in terms of number of employees.

Description of sample

The size of the subsidiaries in Spain is very diverse, ranging from a minimum of five employees to a maximum of 5,400, as is the size of the MNC (ranging from a minimum of ten employees to a maximum of 417,000). 65.5% of the sample subsidiaries had a turnover of more than 10 million euros in 2004. The German MNCs in the sample had been operating for an average of 70.90 years and had made their first direct investment in Spain on average 19.23 years ago. Specifically, 25% of the MNCs (27 firms) entered the Spanish market through FDI before 1977. The entry process speeded up in the following 9 years (from 1977 to 1986), with the entrance of another 25% of MNCs being registered. In the following 12 years (from 1986 to 1998), 75% of the MNCs taking part in the study had entered Spain. Finally, the remaining 25% of the MNCs entered after 1998.

Scale validity

We carried out a confirmatory factor analysis using structural equations by maximum likelihood. The results show an acceptable goodness of fit, since all the absolute (CMIN=71.31, p=0.19; RMSEA=0.03), incremental (NFI=0.79; TLI=0.95), and parsimony (CMIN/DF=1.15) fit measures are within recommended limits. We then calculated the Cronbach’s α and the composite reliability coefficient to evaluate the reliability of the scales used. The results show a high reliability for the constructs physical infrastructures (α=0.74; composite reliability=0.73) and human infrastructures (α=0.82; composite reliability=0.80), and a moderate reliability for the construct political-legal institutions (α=0.64; composite reliability=0.67). These results confirm the validity of the scales for the factors measuring the institutional quality of the host country.

Determinants of level of management and ownership control

The descriptive analyses carried out show that 68.8% of the sample firms had established operations in Spain with a high level of management and ownership control (greenfield), 26% opted for a low level of management and ownership control (joint venture), and finally 4.6% opted for an intermediate level of control (acquisition). This small number of firms made it impossible to measure the general and industrial acquisition mimetism, although it was not a problem to keep the dependent variable with three levels.

Table 2 shows the means and standard deviations of the independent, dependent, and control variables in the study, and the correlations between them. The joint influences proposed in the theoretical model were determined using a simultaneous equations analysis by maximum likelihood, which allowed us to test the proposed hypotheses (see Figure 1).

Figure 1 Institutional antecedents of level of management and ownership control

Although our sample is not very large, due to the difficulty in convincing CEOs to participate in this kind of study (see Li, Bingham, & Umphress, Reference Li, Bingham and Umphress2007), the relations shown in the global model (Figure 1) are stable and we expect them to remain so when the number of observed cases rises. This stability is supported by the estimation of partial models we made, which provide a better sample size to variables ratio (N/v) than global models, as well as a simplified SEM including only those constructs that had previously shown a significant effect on management and ownership control. This simplified model shows a high goodness of fit (CMIN=101.67, p=0.07; RMSEA=0.04; NFI=0.91; TLI=0.96; CMIN/df=1.24) and confirms the effects identified. The literature argues that five cases per variable are sufficient for normal and elliptical distributions, and 10 cases per variable for other distributions (e.g., Bentler & Chou, Reference Bentler and Chou1987). Although the N/v ratio is at the bottom edge of this interval for the global model, it is well within the recommended limits for the simplified model.

The estimated global model allows us to confirm the proposed hypotheses in the majority of cases. First, the results show that the quality of political-legal institutions does not have a significant influence on the choice of a particular level of management and ownership control. In contrast, the more positive the perceived quality of human and physical infrastructures in the host institutional environment, the more likely the firm is to opt for a higher level of management and ownership control for its subsidiaries there. Thus, we can accept Hypothesis 1 partially.

Second, the results show that the more frequently German firms previously operating in Spain had chosen a level of management and ownership control (greenfield or joint venture), the more likely the respondent firm would be to adopt the same level of control with the aim of obtaining cognitive legitimacy. These results confirm Hypothesis 2a. With regard to the relation between the respondent firm’s choice of management and ownership control and the previous choices of the firms from the same sector, the results show that only when firms from the same sector have previously chosen high levels of management and ownership control, does the respondent firm choose the same level of control for its subsidiaries. The non-significance of sector mimetism for the joint venture could be due to the low number of firms from the same sector opting to operate in Spain using a joint venture, since the sectorial variability was substantial. All this allows us to accept Hypothesis 2b partially.

With regard to internal isomorphism in terms of seeking cognitive legitimacy, the results show that the greater the sample firms’ desire to achieve a fit in the organizational structures, systems, and practices, the greater their propensity to raise the level of management and ownership control in their subsidiaries. Thus, we can accept Hypothesis 3a. In contrast, the firm’s desire to make internal isomorphism prevail through transfer of know-how to its subsidiary does not necessarily imply the adoption of a high level of control over them, so we must reject Hypothesis 3b. Finally, and in contrast to what Hypothesis 3c postulates, the results indicate that the stronger the parent company’s intention to seek legitimacy through expatriates, the lower its propensity to use a foreign operation method that increases management and ownership control.

To summarize, the joint influence of the regulative and cognitive dimensions of the institutional environment is able to explain 99% of the management and ownership control the German firms choose in order to continue their operations in Spain.