Parent Control Dynamics and International Joint Venture Performance

This article examines the relationship between foreign parent control and International Joint Venture (IJV) performance over the lifecycle of the IJV. Following Geringer and Hebert (1989); Glaister, Husan, and Buckley (2005), Nguyen and Larimo (2008), the paper conceptualizes foreign parent control across three dimensions including mechanism, focus, and extent. The empirical evidence is based on an analysis of 49 Finnish IJVs established in the 1990s. The result shows that foreign parent firms who adopted a control dynamic approach will see better IJV performance. Broad, tight, and formal control exercised by foreign parent firms over their IJVs leads to better performance of those IJVs in the formation stage. When the performance of an IJV is viewed negatively, foreign parent firms who exercise more control over IJVs will see better IJV performance in the post-formation stage. In contrast, when IJV performance is positive in the formation stage, foreign parent firms are likely to exercise less control over the IJV in the post-formation stage.


Background
In recent years, International Joint Ventures (IJVs) have become a critical part of corporate strategy and firm growth in global markets (Luo & Park, 2004).However, IJVs often experience managerial difficulties in achieving of the intended goals (Park & Ungson, 2001), and thus often have a greater than 50 % failure rate (Parkhe, 1993).Researchers suggest that control problems are one of the primary causes of IJV failures (i.e.Groot & Merchant, 2000).This is said to be due to the fact that in IJVs foreign parent firms are exposed to the risk of opportunism (Zhang & Li, 2001).Furthermore, the operating environment in a foreign market often involves high degrees of uncertainty and risk.Through exercising control, foreign parent firms make sure that their strategies are effectively implemented and their resources are efficiently utilized (Lorange, Morton, & Ghosal, 1986).Insufficient control may result in a leakage of knowledge, proprietary components and capability to outside group (Geringer & Hebert, 1989) or a loss of competitive advantage to other parent firms (Hamel, 1991) or to the competitors (Reich & Mankin, 1986).

Research gaps and research questions
The issue of IJV control was first raised by West (1959), then by Tomlinson (1970), Franko (1971), Behrman (1977), Rafii (1978).The early research on control elaborated the context of IJVs with respect to the relationship between the international strategy of Multinational Corporations, the strategy of local partners, and IJV control.These works, direct attention towards the need for control of IJVs, rather than towards the actual actions required to control and what to control (Geringer & Hebert, 1989).IJV research was continued by the works of Killing (1983), Schaan (1983) and since then has become an important subject in IJV literature (Calantone & Zhao, 2001;Lee & Beamish 1995;Parkhe, 1993;Yan & Gray 2001a).A substantial number of studies has addressed the importance of control to IJV performance (e.g., Beamish, 1988;Sohn, 1994).However, the results on the impact of parent firm control on IJV performance have been conflicting and inconclusive (Calantone & Zhao, 2001;Chalos & O'Connor, 1998;Geringer & Hebert, 1989;Groot & Merchant, 2000;Lecraw, 1984;Yan & Gray, 2001b).Earlier studies on foreign parent dominant control and its relationship to performance show mixed results: that is, they are either 1) positive (Ding 1997;Killing 1983;Lee & Beamish 1995;Luo, Shenkar & Nyaw, 2001); 2) show no relationship (Calantone & Zhao, 2001;Kogut, 1988;); 3) show a negative relationship (Boateng & Glaister, 2002;Child, Faulkner & Tallman, 2005;Isobe et al., 2000;Osland, 1994).In some other studies the results show a shared control between foreign and local parent firms as being positively related to IJV performance (Beamish, 1988(Beamish, , 1993)).In their recent study, Beamish and Choi and Beamish (2004) find that split control between foreign parent firms and local parent firms has an IJV performance level better than that found in the shared and dominant control models.In addition, it has been noticed that the control and performance relationship may vary in IJVs.
Faced with this research puzzle, researchers suggest that future research should seek to deepen IJV control research along its three dimensions including mechanisms, extent, and focus and its relationship to IJV performance (Chalos & O'Connor, 2004;Geringer & Hebert, 1989;Raswamy, Gomes, & Veliyath, 1998;Woodley, 2006).Furthermore, Shortell and Zajac (1988), Brouthers and Bamossy (2006), Nguyen and Larimo (2008), Nguyen and Larimo (2009) suggest the need to study IJV development processes such as control and performance outcomes.In addition, Kogut (2002) maintains that "no matter what the initial agreement on control and ownership may have been at the start of a venture, environmental change over time may shift the relative bargaining power among partners".This may lead to reconfiguration of the control design of IJVs as these environmental factors are crucial for an IJV's stability (Harrigan, 1985).Moreover, Yan and Zeng (1999) and Zhang and Li (2001) argued that the evolution of the control design is dependent on extraneous variables such as change in government policies, competitive environment.They explained that shifts in a partner's bargaining power may result from inter-partner learning, changes in resource contributions, industry structure, and the host government's regulations on foreign investments.These all lead to the continuously changing structure of organizations (Nielsen, 2007).As a result, IJV control design may not be stable over time and thus it is necessary to investigate how and why a parent firm's control structures implemented in their IJVs change during the lifecycle of the IJV.
The purpose of the present study is, therefore, addressed through the research question: (1) What are the relationships between foreign parent control and IJV performance during the formation stage of IJV lifecycle?(2) Why and how do these relationships change during the post-formation stage of the IJV life-cycle?International Joint Ventures.Contractor and Lorange (1988) use the term "joint venture" for the creation of a separate corporation by two or more partners with expected proportional share of dividend as compensation.Hennart (1988) defines an equity joint venture as an independent legal entity that is formed when two or more sponsors bring specified assets to that entity and they are paid for some or all of their contributions from the profits earned by the entity.In the present study, an IJV is regarded as a separate entity located in a foreign market formed by one (or more) multinational firm (s) and one (or more) local firm (s) whether through greenfield or partial acquisitions in the target markets IJV lifecycle has not been clearly defined and agreed in IJV literature.When focusing on the co-operative angle or inter-partner relations, researchers have often divided IJV life stages into three stages, as does Buechel (2000) with a formation stage, an adjustment stage, and an evaluation stage; Heide (1994) with a relationship initiation stage, a relationship maintenance stage, a relationship termination stage; Wood and Gray (1991) with a precondition stage, a process stage, and an outcome stage; and Ott (2003) with the three stages being a bargaining stage, a common agency stage, and a repeated games stage.While Kogut (2002) suggested three stages of the IJV lifecycle: creation, institutionalization, and termination, his research focused on only two stages: creation and termination.In contrast, when focusing on IJV success, researchers often distinguish only two IJV lifecycle stages, for example, formation and termination (Reuer, 2000), or pre-incorporation stage and post incorporation stage (Yeheskel, Newburry, & Zeira, 2004), or a formation stage and stable development, or joint venture formation followed by breakdown (Chowdhury & Chowdhury, 2001).In summary, there are different ways to structure IJV lifecycle stages, depending on the focus of the research.In the present study, the focus is on the control of IJVs and its relationship to IJV performance, therefore, the lifecycle stages of IJVs here include a formation stage and a post-formation stage.An IJV formed three to four years previously will often be ready to enter its post-formation stage and that could also be termed an evaluation or reformulation stage (Shortell & Zajac, 1988).
The definition and measurement of organizational performance is always a controversial topic for academic researchers as well as practitioners in many different areas.While there have been many attempts to define and measure the performance of organizations, due to a lack of consensus on this concept, the extant empirical research has not produced a single theory of performance measurement that can be applied across organizations (Tatoglu & Glaister, 1998).Furthermore, the hybrid nature of IJVs, the possibility of incongruence between partners, and the influence of different cultures in IJVs, all result in the valuation of performance becoming very complex and contribute to there being no consensus on the determinants of IJV performance (Child & Yan, 2003: 283-284;Mohr, 2006;Zeng, 1998).Performance is the ultimate test of a firm's strategy (Schendel & Hofer, 1979), a multidimensional construct (Vryza, 1997), and may be evaluated from a different perspective by each IJV partners (Duan & Chuanmin, 2007).Therefore, different perspectives have been adopted in prior research to assess a venture's performance in previous research including a) a single parent firm (such as foreign parent ot local parent) perspective; or b) a dual parent firm perspective; and/or c) an IJV management perspectives.Geringer and Hebert (1991) find a significant correlation between a parent's assessment and that of the IJV manager.Further, Peng and Luo (2000) point out a high degree of correlation between self-reported data and archival data in China.Moreover, Beamish and Delios (1997) concluded from their review that perceptual and objective measures of IJV performance are generally correlated.Thus, in the present study, perceptual measures such as parent firm satisfaction with IJV total performance and financial performance are used to established performance of IJVs.The article is organized as follows: First, it conceptualizes IJV control.Then it analyzes the relationships between foreign parent control and IJV performance during an IJV´s lifecycle.After that, the paper discusses the methodology, the sample, and the results of the study.Finally the study concludes with a discussion of the implications and potential directions of further study.

The Conceptualization of IJV Control
In the organizational literature, management control refers to the process by which an organization influences its members and its units to work in ways that meet the organizational objectives (Glaister & Bluckley, 1998).According to Child et al. (2005:15), control is a central aspect of management, and essential in any system that holds the managers accountable for their actions and decisions.Ouchi (1977: 95) suggested that "control can be conceptualized as an evaluation process which is based on the monitoring and evaluating of behavior or of outputs".Thus, the organizational literature emphasizes how control can be used to manage individuals and subunits.In IJVs, because there are two or more parties involved, their management control is complex (Geringer & Hebert, 1989).In this paper, the control of IJVs is defined as the influence of the foreign parent firm on the IJV operations.Researchers have also acknowledged that control systems are complex and multidimensional (see e.g Berrell, 2007;Das & Teng, 1998;Geringer & Hebert, 1989;Glaister, 1995Glaister, Husan, & Buckley, 2005;Kumar & Seth, 1998;Lu & Hebert, 2005;Raswamy et al., 1998).Unfortunately, the existing research tends to focus on only one or two dimensions of control.In order to be able to capture the dynamic nature of the IJV and conduct IJV control research thoroughly, this study adopts the multidimensional approach to control developed by Geringer and Hebert's (1989).These control dimensions are elaborated upon below.

Control mechanisms
Control mechanisms are structural arrangements deployed to determine and influence what the members of the organization do (Fryxell et al., 2002;Geringer &Hebert, 1989).Control mechanisms consist of a variety of instruments including formal and social controls that are available to firms to exercise effective control over their members (Behrman, 1977;Friedman & Beguin, 1971).Formal control depends on hierarchies, standards, codified rules, procedures, goals, and regulations that specify desirable patterns of behavior (Das & Teng, 1998).These regulations are explicit in their prescription of behavior and in their means of enforceability (Das & Teng, 1998), aimed directly at protecting the assets of the parent firms (Fryxell et al., 2002).The formal control mechanisms help to decrease the potential for opportunism, by controlling the assets through hierarchical means (Mjoen & Tallman, 1997).Formal control includes majority equity holding, determining the composition of the board of directors, rights to appoint key personnel, control of the planning and approval process for capital budgeting and resource allocation and the setting of routines for IJVs (see e.g.Makino, 1995;Mjoen, 1993;Lu & Hebert, 2005).
Social control is designed to promote expectations and mutual commitments through which the JV managers learn to share the common attitudes and knowledge of the organization (Nonaka & Takeuchi, 1995).Social control refers to various mechanisms such as informal communication, information exchange and training, mentoring, development of a common organizational culture, and establishment of personal relationship with IJV managers.These mechanisms foster shared values and norms, without explicitly restricting the behavior of the selected people through the social controls (Chalos & O'Connor, 1998, Das & Teng, 1998;Fryxell et al., 2002;Schaan, 1983).Compared to formal control mechanisms, social control mechanisms have the potential to reduce monitoring and contracting costs, and to support the flexibility and adaptability of the enterprise (Dyer, 1997).

Control focus
Control focus can be further divided into broad control and narrow control (Geringer & Herbert, 1989).In control focus, the parent firms can choose to have a broad control focus and attempt to exercise control over the entire range of the IJV's activities, or they can have a narrow control focus and confine their control activities to the performance dimensions they consider to be critical (Geringer & Hebert, 1989;Groot & Merchant, 2000).Child et al. (2005) maintain that depending on several factors, including the parent firm's competencies and its evaluation of how critical activities are, parent firms may focus their control over activities related to technology in one instance but then focus on market related activities in another instance.The areas of control focus consist of marketing, sales and distribution, procurement, general management and operation, finance and accounting, R & D and development, production and quality, and human resources.Geringer and Frayne (1990) suggest that one of the crucial areas that determine whether the parent firm's intended objectives are achieved is their focus on the control of human resource control.In this paper, the control focus is considered to be broad when parent firms exercise control over two or more areas, and narrow when they exercise control over only one or two areas.

Control extent
The control extent refers to the degree or tightness of control which is exercised over the venture (Geringer & Hebert, 1989).Control extent consists of tight control and loose control.With loose control, the parent firms tend to use only one or two control mechanisms and focus their control on only one or two control areas exercised over the IJVs.Furthermore, with loose control, the parent firms are more flexible in their evaluation of the employees' behavior and their performance.IJV managers have to submit reports less frequently and there are fewer meetings between the parent firms and the IJV manager in loose control types.Tight control can be effected through any mechanism that provides the partner with a high degree of certainty that the personnel in the IJV will act as the given partner wishes.Control is tight from a partner's perspective if that partner has the right to make or approve the key decisions (Geringer & Hebert, 1989).Tight control is also manifested if the IJV staff is held strictly accountable for adhering to a complete set of actions such as policies and procedures.Tight control is also related to highly frequent and precise reporting (Child et al., 2005).Control can be tightened through more intensive training of the IJV employees in production and management techniques (Van Sluys & Schuler, 1994).However, tight control may also have adverse side effects.As Child et al. (2005) have argued, if control is exercised in a too domineering manner and/or too frequently, it is likely to lead to significant ill will and to the eventual breakdown of the IJV.

Relational Dynamics between Control and Performance in an IJV's Lifecycle
Performance is an important issue in strategic management (Venkatraman & Ramanujam, 1986).Nielsen (2007) suggests that performance should be discussed in the context of the different stages of the development of an IJV.Previous research on the relationship between control and performance has been dominated by a static approach (Zhang & Li, 2001).However, in order to understand the complex nature of the relationship, it is believed that the dynamic perspective may help to investigate how control dimensions affect performance and how performance, in turn, affects the IJV control decisions of parent firms.In addition, previous researchers propose that "the control system at a relatively young IJV should be viewed as embryonic and should evolve over the life of the venture" (Chalos & O'Connor, 1998: 64).The paper continues by analyzing how foreign parent control structures in IJVs influence performance during the IJV formation stage.Then, the paper studies why and how foreign parent control systems in IJVs change under the influence of IJV performance in the post-formation stage.Luo, Shenkar, and Nyaw (2001) find that broad control exercised by a foreign parent is an important contributor to its satisfaction with its IJV performance.Hoon-Halbauer (1994) maintains that the on-going management process of an IJV works better when foreign staffs possess dominant management control and holding the major decision making power.Brouthers and Bamossy (2006), studying the post-formation process of IJVs established by western and eastern European firms, find that high performance in IJVs is associated with broad control at an early stage of IJV formation.Moreover, Child, Markoczy and Cheung (1994) finds that IJVs in emerging markets like China, local managers are reluctant to make decisions and are afraid of accepting personal responsibility.It, therefore, appears in joint ventures at least local managers may need time to learn to take actions and accept responsibility in managing the IJVs.As a result, it is to be expected that broad control by foreign parent firms leads to better IJV performance during the formation stage.Child et al. (2005) maintain that control is critical for the successful management and performance of any strategic alliance.Mjoen (1993) find that a tight degree of control was associated with better performance.Studying of IJVs in China, Chen (2004) maintains that tight control by local parent firms over IJVs has a negative impact on performance.This is because local parent firms usually do not advantageous expertise of management IJVs.In IJVs, local parent firms primarily contribute land, facilities, labor so they may not have a capability to control the technology and know-how contributed by foreign parent firms.In such situation, the result of tight control by local parent firms is a delay in the decision making process and to hinder implementation of best practices contributed by foreign parent firms, leading to a decrease in the competitive advantage of the IJVs.Therefore, a foreign parent firms should not let a local firm tighten control of an IJV but instead should exercise control itself (Chen, 2004).As a result, tight control by foreign parent firms leads to better IJV performance in formation stage.

Foreign parent control choice in the formation stage of the IJV lifecycle and their effect on IJV performance
Furthermore, Fryxell, Dooley, and Vryza (2002) find that formal control and IJV performance were positively related in younger IJVs but that this relationship became negative in more mature IJVs.As a result, formal control mechanisms are aimed at monitoring the behavior and performance of IJVs, which is critical for stability and efficiency during their early development stage (Larson, 1992).As a result, formal control by foreign parent firms leads to better IJV performance during the formation stage.In summary, we propose that: Hypothesis 1: Broad, tight, and formal control exercised by foreign parent firms over their IJVs leads to better performance in the formation stage.

IJV performance during the formation stage and the changes in foreign parent control in the post-formation stage
Lorange (1997) maintains that because IJVs evolves over time, the balance of a parent firm's control should be shifted accordingly.In the same vein, Brouthers and Bamossy (2006) argue that changes in IJV structure and control relationships may result in lower transaction costs and improved IJV performance.In managing IJVs, Lee, Chen, and Kao (1998) suggest that foreign parent firms need to adopt a flexible mindset.Groot and Merchant (2000) suggested that IJV performance is directly related to parents firms' attitudes toward control.Several studies have focused on how IJV performance affects IJV reconfiguration.Vaidya (2000) argues that control related failures are likely to occur if control practices are not re-evaluated and modified in response to changing circumstances.Simonin (1997) argues that a parent firm's competence at successfully designing subsequent control structures is ultimately reflected in future IJV performance.Killing (1983) observes that partners might loose or strengthen control over the IJV in response to the IJV's on going performance.According to Brouthers and Bamossy (2006: 8), control changes may lead to improved cooperation and performance.
In post-formation stage, when performance results carrying over from the formation stage are positive the use of formal control mechanisms to mitigate opportunism in the later stage may be short-lived (Fryxell et al., 2002).This is due to the fact that formal control mechanisms are based on formal contracts between parent firms, and therefore, valid for a certain period of time (Dyer, 1997).When a parent firm's needs and strategies for IJVs change, they often call for repeated contracts renegotiation, leading to very high transaction costs.In addition, in the post-formation stage of IJVs, when the relationship between foreign parent and local parent firms matures, formal control mechanisms can cause conflicts and promote distrust between parent firms (Ring & Van de Ven, 1994).Consequently, in the post-formation stage of the IJV lifecycle, social control mechanisms (based on trust and commitment) are more economical, that is once the initial cost of establishing social controls averages out to become less than the cost of periodic contracting (Fryxell et al. 2002).Furthermore, Fryxell et al. (2002) find that social control and IJV performance were positively related only in the presence of effect-based trust between the parents, which is built during the development of IJVs.On the other hand, if parent firms rely totally on social control in the early in IJV lifecycle stage they may expose the potential for opportunism in the formation stage.Thus, trust often appears between parent firms in the post-formation stage rather than during the formation stage of the IJV lifecycle.In the post-formation stage, there is a need to find a balance between control and maintaining harmony between foreign parent and local parent firms.Kauser (2007) suggests that where there is a pre-existing trust, control should be monitored by social mechanisms.In the same vein Chen, (2004) maintains that in the later stage social control helps to increase a parent firm's confidence in its IJVs.This confidence can lead to high expectations of mutual learning and high degree of commitment by parent firms to the IJV and so promotes knowledge sharing and creates a more competitive bundle of resources (Dyer & Nobeoka, 2000).In short, when IJV performance in the formation stage of the IJV lifecycle is positive, foreign parent firms tend to exercise social control over their IJVs in the post-formation stage.
Furthermore, in post-formation stage, local managers also learn how to manage and direct IJVs towards their targeted goals.Robins et al. (2002) suggest that "mature" ventures often use local personnel supplied by local partners at many levels, thus giving local parent firms broader control over IJVs.Further, previous research has suggested that IJVs require greater autonomy in order to adapt to the social, political, financial, legal and cultural expectations of the IJV host environment (Yeheskel, Newburry, Zeira, 2004).Moreover, Yeheskel et al. (2004) proposed that foreign parent firms that allow IJV managers to formulate and implement policies in functional areas that are sensitive to the host country would lead to better IJV performance.Superior performance of an IJV in post-formation stage is also achieved if foreign firms focus their control on their resource contribution not on the whole range of IJV activities (Yan & Child, 2004).Moreover, during the post-formation stage of IJV as the level of trust increase between foreign and local firms, the need for control diminishes (Nielsen, 2007).Therefore, parent firms seek to focus their control over particular activities, rather than control all activities (Kauser, 2007).Merchant and Groot (2000) suggest that when recent IJV performance rates are relatively high, parents firms will utilize more looser control over the IJV and narrower control foci.As a result, when IJV performance in the formation stage of the IJV lifecycle is positive, foreign parent firms tend to exercise narrow control over their IJVs in the post-formation stage.
In addition, when an IJV has superior performance, the partners tend to loosen control since the IJV's management team has proved its expertise and that helps to maintain harmonious relations with the local parent firms (Child et al. 2005).Similarly, Chen (2004) proposed that over time, tight control activities exercised by foreign parent firms might decrease due to the influence of the IJV management getting stronger at the post-formation stage than it was in the formation stage.This is because it permits the expectations of labor unions, clients, suppliers, and other local stakeholders to be properly considered.This would positively affect worker commitment and productivity and increase the fit between the IJV and its host environment.As a result, this will increase the effectiveness of IJVs and leads to better IJV performance.Gaining more autonomy in the post-formation stage is also important for an IJV's survival.In foreign markets, IJVs often need to be able to react quickly to fast-changing markets by reducing long decision-making processes and bureaucracy.Supporting the argument for more autonomy for IJV management, Bleeke and Ernst (1991) contend that operating decisions are best made by those who are focused purely on the JV.Yeheskel et al. (2004) point out that foreign parent firms are willing to permit more autonomy when their IJVs operate under more certain and stable circumstances.In addition, Nielsen (2007) argues that the venture mature over time and partners are more familiar with each others' idiosyncrasies, thus the uncertainty within transactions decreases.Consequently, intensive control is not necessary.In addition, Dianne (1997) suggests that as the IJV has progressed well, foreign parent firms should have clarified the criteria and time frames for the gradual transfer of control and power to their local partners.As a result, when IJV performance during the formation stage of the IJV lifecycle is positive, foreign parent firms tend to exercise loose control over their IJVs in post-formation stage.
On the other hand, undesirable IJV performance prompts structural instability in control system (Yan, 1999).Poor IJV performance is often caused by at least one of the partner firms failing to achieve its objectives, thus creating stimuli for changing the ineffective existing control system.Yan and Gray (2001b) suggested that a factor that could produce change in an IJV is a shift in IJV performance.When investigating the evolution of IJVs, Child (2002) concluded that over time, foreign parent firms are likely to increase their equity share for the following reasons: a) the need to buy out unreliable partners; b) the unwillingness or inability of local partners to finance IJV expansion.In their study, Brouthers and Bamossy (2006) also found that tighter controls were normally instituted after poor results and/or misdeeds.As a result, when IJV performance in the formation stage of IJV lifecycle is negative, foreign parent firms tend to exercise more control by exercising broad, formal, and tight control over their IJVs in post-formation stage.In summary, we propose that: Hypothesis 2: When IJV performance has been negative in the formation stage, foreign parent firms tend to exercise formal, broad, and tight control over their IJVs in the post-formation stage of the IJV lifecycle.In contrast, when IJV performance has been positive in the formation stage, foreign parent firms tend to exercise social, narrow, and loose control over their IJVs in the post-formation stage of the IJV lifecycle.

Methodology
To test the hypotheses, the study uses a survey research method to collect data for the empirical evidence.In the survey, the questions about joint venture control and performance were collected directly from those involved in IJV operations.The measurement of variables is based on a 5 point-scale.The methods used in this study to analyze the data are description statistics and the Chi-square test.The purpose of the methods is to determine how well an observed set of data fits an expected set of hypotheses.The method is particularly useful to find out whether an IJV control structure which is made up of different elements of the IJV control dimension (formal, social, broad, narrow, tight, and loose) has a normal distribution or whether the structure has been formed under the influence of performance factors.

Data collection and measurement
The study here is a part of an on-going research project focusing on IJV behavior, strategies, partner selection, control structure, and performance of Finnish firms.The target firms and investments were identified as follows: 1) The FDI data base collected by the project leader starting from the late 1980s based on press releases regarding IJVs published in several Finnish business magazines and newspapers, 2) Annual reports and websites of the 250 largest Finnish firms as reported by the leading business magazines, 3) based on the earlier surveys focusing on IJVs and wholly owned subsidiaries of Finnish firms conducted by the project leader.From the resources, we identified 340 qualifying IJVs formed by Finnish firms since 1985 and in operations until at least 2002.The 340 qualifying IJVs involved 250 Finnish parent firms.Of those 250 firms, several firms were very difficult to contact either because they had been restructured or because they had gone out of business.The aim of contacting the firms was primarily to identify the best informants.In some firms there was no longer anyone employed who had sufficient knowledge to further study.This left a total of 200 Finnish parent firms.Due to time and cost constraints a postal questionnaire and online web survey were used to gather the data.
The participants were those managers who directly involved in the establishment and operation of the IJVs.To enhance the quality of the data, the respondents were contacted by phone in December 2006 to explain the key points of the study and the questionnaires.In exchange for their cooperation with the study and to provide motivation and accurate responses, the respondents were assured of anonymity and were promised a summary report of the findings, and were also entered into a draw for three gifts.Following one reminder at the end of the January 2007, 54 questionnaires were collected from 49 participating firms at the end of February 2007, providing a response rate of 24,5 % which is relatively similar to that of earlier respective studies in Finland (see Larimo & Rumpunen, 2006).The sample was carefully examined for any systematic response bias using t-tests.Respondents and non-respondents were compared by their age, size, international experience, and IJV experience.No statistically significant difference was found.Thus, there was no response bias to be found in the final sample.After taking out the uncompleted questionnaires, the final sample related to 49 IJVs.Among these IJVs, 45% were established 1988-1995, and 55% 1996-2006; 53 % through acquisitions, and 47% through greenfield investment; 76 % involved 2 partners and 24 % involved 3 partners; 61% were established with an indefinite term, 22% with a duration of less than 5 years, and 17 % stated to have an intended duration of more than 5 years; 41 % had a degree of Finnish ownership of between 10%-49%, 10% had equal ownership, and 49 % had Finnish majority ownership upon establishment; 71% were located in emerging economies, and 29% in developed economies; 63% dealt with industrial products, 27 % with consumer products, and 10 % offered both consumer and industrial products.
All three control dimensions were measured with a 5 point-scale.Based on the list of different mechanisms and focus areas provided, the respondents were asked to assess 1) their method of monitoring and control, and 2) their focus of monitoring and control at the formation and post-formation stages (Cronbach's Alpha=0.89, and 0.92, respectively).The of IJV performance was also based on a 5 point-scale, the respondents were asked if they were satisfied with IJV performance on both financial and total performance scales with 1= "very dissatisfied" to 5= "very satisfied" at both formation stage and post-formation stage where Cronbach's Alpha values were = 0.91 and 0.94, respectively.

4.
3.1 Parent firms' control choice and IJV performance in the formation stage of the IJV life-cycle Performance was measured using seven different subjective measures.Respondents were asked on a 5-point Likert scale, first to weight a suggested set of measures, and then to express their degree of satisfaction with each of the seven measures.The two most important measures of performance were total performance and financial performance.Among 49 respondents, 32 Finnish parent firms (65%) exercised formal, broad, and tight control over their IJVs during the formation stage.The respective mean of the IJVs total performance was 4.6 (with 1= very dissatisfied to 5= very satisfied) and financial performance was 4.3.The remaining 17 Finnish parent firms (35%) exercised other control structure such as social, narrow, and loose control.Those parent firms demonstrate a respective total performance mean of 2.5, and one for financial performance of 2.4.Therefore, formal, broad, and tight control exercised by foreign parent firms in the formation stage of the IJV lifecycle leads to better IJV performance.Also, based on the chi-square test, 2= 16.5 the result was significant at P<0.005 (df=5) (see table 1).Thus, the results supported hypothesis 1 for both description statistic and chi-square test.4.3.2IJV performance in the formation stage and parent control in the post-formation stage of the IJV life-cycle According to the respondents, 75% of foreign parent firms (37 firms) were satisfied with the IJV's total performance (mean 4.3) and financial performance (4.1) resulting from IJV operations in the formation stage of the IJV lifecycle.Of these 37 foreign parent firms, 95 % foreign parent firms exercised social, narrow, and loose control over their IJVs in the post-formation stage of IJV lifecycle.On the other hand, the 25% of foreign parent firms (12 firms), who were not satisfied with their IJV performance in the formation stage of the IJV lifecycle, all exercised formal, broad, and tight control over their IJVs in the post-formation stage of the IJV lifecycle.In addition, according to a chi-square test 2= 15.27, the result was significant at P<0.01 (df=5) (see table 1.).Thus, the results supported hypothesis 2 for both description statistics and chi-square test.So that if IJV performance in the formation stage of the IJV lifecycle has been positive, foreign parent firms tended to exercise social, narrow, and loose control over their IJVs in the post-formation stage of IJV lifecycle.In contrast, if IJV performance has been negative in the formation stage of an IJV's lifecycle, foreign parent firms tend to exercise formal, broad, tight control over their IJVs in the post-formation stage of IJV lifecycle.
All in all, the relationship between the design of the control structure and performance is not just one-way, but rather is reciprocal and dynamic.With an initial control system designed by a foreign parent firms, there will be two situations affecting IJV performance.In the first case, when the performance is negative, foreign parent firms will respond to this situation by exercising more control over IJVs to restore order and make sure that the IJVs achieve planned targets.However, control also carries costs, especially with broad control and tight control.In addition, foreign parent firms may also signal their distrust of local parent firms, by insisting on a lot of procedures and excessive paperwork that may flow from formal control and tight control.Therefore, when IJV performance is positive, foreign parent firms may prefer to exercise less control over IJVs to create trust with local parent firms and to reduce associated costs (see Figure 1.).

Contribution and Conclusions
With regard to control in IJVs, research has traditionally been modeled solely by the relative degree of ownership (Mjoen & Tallman, 1997).However, ownership may not be the optimal means of control in every situation and may be but a minor issue in governance.Guide (2001) states that a firm does not need majority equity ownership to exercise managerial control.Sohn (1994: 299) points out that social knowledge may help MNCs reduce dependency on ownership as means of control over their subsidiaries.In addition, when control of IJVs was discussed separately from the ownership structure, most previous research focused on only one dimension of control, mainly control mechanisms (e.g.Mjoen, 1993;Vryza 1997;Xiansheng, 1998;Yan & Gray 2001a).However, to be able to recognize the control issues within IJV clearly, Geringer and Hebert (1989) and Lu and Hebert (2005) and Raswamy, Gomes, and Veliyath (1998) proposed that future research should examine the multidimensionality of IJV control including control focus, control mechanisms, and control extent, and also the relationship of each one to IJV performance.
The first contribution of the present study to IJV theory is to fill this gap by providing a better understanding of the multidimensionality of parent firm control and each element's influence on IJV performance.In addition, the present study is the first to examine the change of IJV control during the IJV lifecycle since Zhang and Li (2001: 342) concluded "how control design of IJVs evolves over time remains unclear".This is important because it examines organizational process.The research of process makes an important contribution because it explains how managers can influence firm performance through organizational control (Chakravarthy & Doz, 1992).In the relationship between IJV control and performance the present study points out that the conflicting results from previous studies are, perhaps, because they have assumed a monotonic relationship between control and performance (Fryxell et al., 2002).As it takes one step further away from this approach, the present study proposes that the relationship between IJV control and performance depends both on which lifecycle stage IJVs are at, and on the results of IJV performance during the previous stage.In particular, the present study proposes that when foreign parent firms choose to exercise broad, formal, and tight control at the formation stage of the IJV lifecycle, IJV performance is likely to be positive.In contrast, if foreign parent firms choose to exercise narrow, social, and loose control over IJVs at the formation stage, IJV performance is likely to be negative.
In addition, foreign parent firms will redesign their control exercised over IJVs over time according to IJV performance.In the post-formation stage, when IJV performance is negative, foreign parent firms need to exercise more control over IJVs in an attempt to improve performance.When IJV performance is positive foreign parent firms may reduce their degree of control over IJVs in order to reduce the costs associated with control, and to promote a higher degree of trust, the potential for which had probably already been observed during the initial stage, and evidenced through strong performance.This finding is consistent with Zhang and Li (2001) who report that during IJV development process, IJV performance becomes a motivating force stimulating the evolution of IJV control design.The present study extends this by specifying which particular control structure is needed during the period of such changes.In short, it can be concluded that there is no optimal choice of IJV controls that are stable over time.Control of IJVs needs to be re-adjusted over time in accordance with IJV performance.As a result, this study may help to explain previous conflicting results, such as why in some cases dominant control forms yield better performance while in others they do not but instead minor control or shared prove effective control forms.This finding contradicts Glaister and Buckley's (1998) which maintains that the nature of management control does not vary with the development of IJVs.
This study does have several limitations.First, the sample size of the study is rather small and the analysis based only on data from foreign parent firms.For further studies, researchers could use the framework of this study with a bigger sample size and incorporate the viewpoints of both local and foreign parent firms to study the dynamics of parent firm control.In addition, the study defines an IJV lifecycle as consisting of two stages; further study could test the framework of this study using three or more stages.Finally, although performance is the main motive driving IJV control change, other factors should be taken into account.Future studies may wish to investigate control dynamics during the IJV lifecycle in accordance with the changes in operating environments of IJVs, the changes to parent strategies in IJVs, and the relationship development between local firms and foreign firms.

Figure
Figure 1.IJV Control model