Toward Understanding IJV Control : Parent Inside and Outside Control – An Exploratory Case Study

Previous studies have failed to recognize the difference between parent control over its ‘own people’ and ‘other people’ in IJVs. The paper for the first time in the literature separates parent control into two modes of control –inside and outside control upon the nature of the relationship between parents and respective IJV management teams. We examine inside and outside control through an analytical inductive case study. Our result shows that inside control determined by parent ownership is an endogenous contingent control. It has a broad but selective control focus. It gives a parent adaptability and flexibility in selecting control extent and mechanisms. Outside control is determined by parent overall bargaining power. It is a pre-determined control with a narrow specific focus, and its control extent and mechanisms are predefined. Our study contributes to the literature on IJV control by adding the nature of the relationship between parents and individual management groups into the IJV control equation, unveiling two distinctive forms of control within IJVs.


Introduction
Despite the rapid growth of international joint ventures (IJVs) in terms of both frequency and strategic importance, the estimate 30%-70% failure rate indicates that it is very difficult to run an IJV (Hennart & Zeng, 2002).A critical determinant of IJV performance and success appears to be the control exercised by parents over a venture's activities (Geringer & Hebert, 1989).However, there is little IJV -related research that is focused on control issues.Some scholars have suggested that one-parent dominant control is better than shared control because it reduce unnecessary disputes and conflicts over operational decisions (Ding, 1997;Yan & Gray, 2001) as well as minimizes transaction costs and stabilizes IJVs (Geringer & Hebert, 1989).In contrast, other IJV scholars have argued that shared control creates mutual respect and a sense of fairness that promotes trust and reduces conflict between joint venture parents (Steensma & Lyles, 2000;Yan & Gray, 1994).But, previous studies have produced very controversial results for both dominant and shared control (Yan & Gray, 2001).
There are many possible reasons for the controversial results between control and performance in previous studies: maybe a lack of an integrative approach to studying control in IJVs (Geringer & Hebert, 1989), a lack of multiple-level analyses of all the parties (Luo & Park, 2004), and a lack of correspondence between theory-building and testing (Yan & Gray, 2001).We argue that one of the major factors that have contributed to the empirical inconsistency in the literature is the lack of a clear and meaningful classification of parent controls.IJV management team is responsible for day -to -day IJV operations.And how parent controls are executed by IJV management teams is affected by the nature of the relationship between parents and respective IJV management teams.Previous studies of IJV control have not given any attention to the nature of the relationship between parents and respective management teams and how parent control is affected by such relationships in IJVs.
The current study is addressed at this void.The central question of the study is: How do parents exercise effective control over their IJVs?And follow -on question are: How does the nature of the relationship between parents and respective IJV management teams affect parent controls?How do parents exercise effective control through 'their own people' in the IJV?How do parents exercise control through 'other people' in the IJV? Are they similar or different?How are they different?
Our study contributes to the literature on IJV control.A key insight is the difference between a parent company's control over IJV activities through its 'own people' under inside control and its control through 'other people' under outside control in an IJV.Such difference is a result of the nature of the relationship between parents and respective IJV management groups.A second contribution is the determinants of inside and outside control.Inside parent control is an endogenous and contingent form of control that is determined by parent ownerships.Parent outside control, which is defined in formal agreements between parents, reflects the overall bargaining power of the parent.A third contribution is the difference between inside and outside control on control focus, extent and mechanisms.Inside control provides parents flexibilities to exercise different controls upon strategic importance of control targets.In contrast, outside control is predefined in formal agreements in which control focus, extent, as well as mechanisms used are defined.The most fundamental contribution of this research is that it offers scholars and practitioners a new lens for understanding IJV control and helps decode the contradictory empirical results in previous studies.

Background
Control in the context of IJVs refers to the process by which one entity influences, to varying degrees, the behavior and output of another entity through the use of power, authority and a wide range of bureaucratic, cultural, and informal mechanisms (Geringer & Hebert, 1989).This control makes IJV activities more predictable, thereby decreasing the uncertainty but increasing the likelihood of reaching a desired goal (Ding, 1997;Emsley & Kidon, 2007;Groot & Merchant, 2000).Different types of IJV control based on ownership have been studied such as: dominant control (Ding, 1997;Killing, 1983), share control (Beamish, 1993;Yan & Gray, 1994), and split control (Choi & Beamish, 2004).
"In spite of all the work over the past decades, there is still no consensus on the issue: dominant control, minority control and equal control, as measured by the extent of equity ownership, have all been linked to performance (howsoever defined) but with no consistent results.It was (and still is) therefore obvious, at least to me, that the literature was 'barking up the wrong tree" (Madhok, 2006, p. 6).
Previous studies of IJV control are focused on how control power is divided between parents based upon ownership.Control is not a strict and automatic consequence of ownership (Geringer & Hebert, 1989).Overall parent control is not determined by ownership alone (Geringer & Hebert, 1989;Yan & Gray, 1994); it is significantly shaped by bargaining power (Mjoen & Tallman, 1997;Yan & Gray, 1994).The amount of ownership does give a parent an ability to appoint their personnel to key positions in IJVs (Groot & Merchant, 2000;Mjoen &Tallman, 1997;Wang, Wee and Koh, 1998;Yan & Gray, 1994).
IJV control is carried out through IJV management teams, which are responsible for the day-to-day operations of IJVs.However, previous research on IJVs has not adequately considered IJV management teams (Hambrick, Li, Xin and Tsui, 2001).If we are to understand why some IJVs succeed and others fail, we need to closely examine the role of an IJV's management team (Hambrick et al., 2001;Johnson, Korsgaard and Sapienza, 2002) on the effectiveness of parent controls.Currently, there is no any attention that has been paid to the distinct nature of parents as opposed to respective management groups.Management group composition and behavior are of great consequence to organizational outcomes such as information processing and power and influence patterns (Hambrick et al., 2001).Differences in national backgrounds, language, cultural values and organizational identity further accentuate the inherent complexity of the relationships between IJV parents and between parents and IJV management teams (Inkpen & Beamish, 1997;Johnson et al., 2002).Furthermore, joint venture managers typically represent their respective parents (Newman, 1995).Such coalitional relationships lead to allegiances between the parents and their management group in the IJV (Groot & Merchant, 2000) and thereby affect how parents control IJV's activities through 'their own people' versus 'other people'.

Theoretical control model
Insert Figure 1

about here
In Figure 1, we offer a model that separates parent control into inside and outside control.Inside control in the model is defined as the control exercised by a parent over those activities led by its own representatives or delegates in an IJV.It is an informal control from the IJV perspective, but it may be a formal control from the parent perspective.Its power is determined by parent ownership.In the model, outside control is a control exerted by a parent over those activities led by the other party representatives or delegates.Outside control is formal, and the level of parent outside control power is determined by the parent's overall bargaining power.The scope, extent, and mechanisms of outside control are defined in formal agreements between the parents.In contrast, three dimensions of inside control vary upon the level of strategic importance of the matter as perceived by the parent.
Although the main thesis of the model may apply to all JVs, the nature and extent of the inside and outside control may vary depending on ownership, bargaining power, parent management skill and expertise, and organizational culture within parent firms.It may also vary between JVs in developed countries and those in developing countries.We investigated the model by focusing on IJVs in a developing country and used our theoretical model as a benchmark, comparing our data against the model through an analytical induction method.

Methods
Working from the limited theory about parent control, we conducted an inductive traditional single case study through which we can compare our experiences and gain rich theoretical insights.The traditional single case study enables a deep understanding of a particular social setting through getting as close as possible to the world of managers and interpreting this world and its problems from its inside and offers the benefits of comparative insights within the same organizational context (Dyer & Wilkins, 1991).

Case Selection
The setting of the study was the oil and gas industry.In this study, we conducted a single but rigorous case analysis of an IJV between the Middle East and the People's Republic of China.To ensure confidentiality, the names of the parent companies were coded as CHINOIL (referring to a Chinese state-owned company pursuing outward direct investment for the purpose of oil/gas exploration), and MIDEOIL (referring to the host state-owned company, which was located in the Middle East).The name of the joint venture is the IJV.The goal of the IJV was to explore, appraise, and develop non-associated gas and natural gas liquids in a defined area in the Middle East.

Data Collection
We relied on three primary data sources: first-hand knowledge, archives, and informal discussions.The author was the contracting manager of the IJV during 2004-2008 and was in charge of managing IJV procurement activities and all contracts.Most of the IJV operations were contracted to third parties.In 2006, for example, 88.6% of the IJV operational expenditure was contracted to other parties, including the two parent companies.Not only was contracting extensively involved in the IJV activities, but also its scope covered most of the departments in the IJV, including Human Resources and Geophysical, Exploration, and Drilling departments.The author had a privilege to attend most of the official and unofficial meetings, including board meetings and internal meetings of the Chinese group.The author has developed first-hand in-depth knowledge of the IJV, including parent controls.
The archival data include email, board meeting minutes, internal reports, presentations given at the board meetings, and official agreements.Using first-hand knowledge and extensive archival data, we were able to present a chronological and dynamic view of the IJV.The triangulation of data sources provides more accurate information and improves the robustness of the resulting theory (Anand, Gardner, and Morris, 2007).We triangulated the data in various ways.We not only studied and compared different archival data but also compared those data with the first-hand knowledge.When an inconsistency happens, we contacted the chief operational officer, the exploration manager, and the drilling manager for clarification.This rigorous comparison and analysis not only allowed us to avoid any potential biases but also ensured data validity.We hope to portray the living reality of the IJV and the perceptions of relevant actors.Any data that might reveal the confidentiality of any party were removed, but this did not compromise the data volume and accuracy.

Data Analysis Method
Following Yan and Gray (1994) and Santo and Eisenhardt (2009), we used the analytical inductive method, which is appropriate for this case study.We began with an in-depth case analysis through the lens of a research question: How do parents practice their effective controls through respective IJV management groups?There were no theoretical preferences or a priori hypotheses.We started with a single event that occurred within the IJV to identify the theoretical constructs and relationships influencing the control that the parent exerted over its 'own people' in the IJV.A modification was made along with the second event, and a further comparison was made with more events.Through multiple events analysis and comparison, insights about inside control emerged based on consistent patterns and themes (Santo & Eisenhardt, 2009).As the patterns emerged, other events that had occurred at different times were added to develop propositions and more robust theoretical concepts.

Identification of Outside Control
The IJV is governed by the Board of Directors (BOD) through a series of formal agreements.The BOD is composed of four Chinese directors from CHINOIL and one from MIDEOIL.The sixth member is an independent director from the host country government.The composition of the BOD is consistent with the ownership breakdown: 80% of the members from CHINOIL and 20% from MIDEOIL, not including the independent member.
The members of the management team (MT) are seconded from the parent companies.Five positions out of the seven are held by CHINOIL.These positions include CEO, COO, CHRO, director of the exploration department, and director of the drilling department.The other two positions, CFO and director of the geophysical department, are held by MIDEOIL.The overall management positions in the IJV are proportional with ownership split between CHINOIL and MIDEOIL.The major operations of the IJV happen in the Department of Exploration, Department of Drilling, and Department of Geophysics.
In the area of related -party transactions, parent controls are executed under outside control through formal agreements.The shareholder agreement is the most important agreement in the IJV from which all the other agreements were originated.Parent control of related -party transactions under outside control can be illustrated in the shareholder agreement: All related-party transactions shall be on terms that are no less favorable to the IJV than those that could have been obtained in a comparable arm's-length transaction by the IJV with an unrelated third party.All the related-party transactions with value exceeding $250,000 must be approved by not less than 51% of the Directors nominated by shareholders (non-related party directors) other than the related party.
Clearly, outside control is a predefined control targeting at a specific area.Although not all related-party transactions need an approval of the non-related party directors, those with costs lower than $250,000.00are subject to the annual audit of the BOD.
Proposition 1: Outside control is a prefixed parent control having particular targets.

Scope, Extent, and Mechanism of Outside Control
Under the Shareholder Agreement, there are two tiers of control mechanisms for governing all procurements.The first one is the Contract and Procurement Policy (CPP & P) Procedure, by which all procurement activities including related-party transactions are governed by a competitive bidding process.A definition of related-party transactions is given in the description of the CP P&P procedure: "All contracting and purchasing activities of the Company with or for the benefit of any shareholder or any affiliate of a shareholder".Furthermore, all related-party transactions "shall be conducted on an arm's-length basis".
The second control is the Technical Service Agreement (TSA) that was signed between the IJV and each of the parents.The purpose of the TSA is to offer the IJV an ability to acquire parent services directly for IJV urgent or confidential needs.Such related -party transaction services don't need going through the market competing process.The implementation of the TSA is strictly governed by the shareholder agreement.Transactions through TSA are considered related -party transactions.
All the related-party transactions through either TSA or CPP&P with a value under $250,000.00are subject to the annual audit of the board.Those with contract price equal to or above $250,000.00are solely controlled by non-related party directors no matter which of the procurement procedures are used.Oil and gas industry is a capital intensive field.Contract prices for most of the transactions associated with drilling and exploration are well above $250,000.00.This leads to that most of the transactions between CHINOIL and the IJV drilling department or between CHINOIL and the IJV exploration department need a final approval of the MIDEOIL director.
Proposition 2: Outside control is a content-oriented form of parent control.Its focus is narrow and specific.Proposition 3: Outside control is a centralized parent control.

Insert Table 1 about here
Chinese directors of the IJV drilling and exploration departments are not happy with the fact that they have to seek a final approval of the MIDEOIL director for all the transactions with CHINOIL, complaining that the control power of MIDEOIL over CHINOIL in related -party transactions is unjust because of MIDEOIL's 20% ownership and CHINOIL's 80% ownership.
The outside control power of MIDEOIL over CHINOIL in related-party transactions does not come from its only 20% ownership share.Instead, it is determined by its overall bargaining power.As indicated in Table 1, the overall bargaining power of MIDEOIL is enhanced by its alternative availability, the moderate strategic importance of the IJV to MIDEOIL, its management expertise and local knowledge.Furthermore, its bargaining power in the areas of technology, market access, and material procurement could be high if MIDEOIL were involved in the operational control of the IJV.In contrast, CHINOIL as the majority party (80% ownership) has low alternative availability but high strategic importance of the IJV to CHINOIL, limited overseas business experience (and thus a lack of overseas management expertise) and a lack of access to global markets and material procurement.The overall bargaining power of CHINOIL is lower than its ownership level.
Proposition 4: Outside control power is determined by the overall bargaining power of the parents.

Identification of Inside Control
Inside control of the IJV from CHINOIL are found in the departments of Human Resources, Exploration, Drilling, and Health & Safety Environment (HSE).The common characteristic among the four departments is that they all have Chinese directors seconded from CHINOIL.Such coalitional relationships provide a foundation for CHINOIL to practice its inside control.It is important to note that CHINOIL did try to perform its control over the Geophysical and Procurement and Contracting departments but encountered great difficulty.The manager of the Geophysical Department was seconded from MIDEOIL, and the manager of the Procurement and Contracting Department was independent and not affiliated with either of the parents.Both the managers insisted that all their operations follow the IJV procedures and policies.This finding indicates that inside control is an endogenous and contingent form of control relying on 'its own people' in the IJV.
The ability of a parent to appoint its own people to key positions in the IJV including senior officers and MT members is solely determined by its ownership.Based on the shareholder agreement, the selection of the people for the key positions is consistent with parent percentage interests.So, the parent's inside control power is determined by its equity ownership.
Proposition 5: Inside control is an endogenous and contingent form of control.
Proposition 6: The power of parent inside control is positively proportional to its equity ownership.

Scope, Extent, and Mechanism of Inside Control
Oil and gas exploration is an expensive, high-risk operation.The cost of each exploration well is over $40 million dollars.CHINOIL exercising its control over the operation of exploration and drilling is consistent with its interest.How does CHINOIL exercise its inside control?Is it broad or narrow control?Is it selective or non-selective?Drilling operations include drilling engineering, drilling operations, health and safety management, and budget and payment.CHINOIL did not control the entire drilling operation procedure.Instead it selectively controlled only those components that it perceived as critical, such as drilling programs -one of the components in drilling engineering, special operations that are involved with high operational cost, difficulty and safety, and a portion of procurement such as on-site drilling supervisors.
CHINOIL didn't perform its control over the entire exploration operation either.It selectively controlled exploration strategy, planning, exploratory well location, and data acquisition.It did not intervene with any details of the exploration operations.
Proposition 7: Inside control is broad and selective.
The extent of CHINOIL inside control varies from joint control, dominant control, to centralized control.The variance of the inside control extent is a result of variance among the different functional departments of CHINOIL and level of strategic importance of control target.CHINOIL inside control does not come from one office or one department.Instead, each of the functional departments acts as the parent exercising their control over the corresponding department in the IJV.
Decision-makings of drilling program and special drilling operations have been maintained as a joint control between CHINOIL Engineering Department and the IJV.In contrast, CHINOIL Exploration Department played a dominant role in exploration operations from 2004 to 2006 and a centralized role after 2006.The change of CHINOIL's IJV inside control results from several factors.The first is the change of the key personnel in CHINOIL.The new leader in charge of the IJV management and control favors a close IJV control.The second factor is the establishment of CHINOIL Middle East Division through which CHINOIL further strengths its IJV controls.The third factor may due to the less achievement of the IJV in the past years.Not only are exploration strategy, planning and well location important to the success of the IJV, but also the overall oversea strategy of CHINOIL.CHINOIL's dominant and later centralized control in exploration operations is determined by strategic and economic importance of exploration activities as perceived by the parent.
The major control mechanism used by the CHINOIL Engineering Department is a process-based mechanism that is defined as parent firms exercise control by reporting on relationships or influencing IJV planning and decision-making processes (Geringer & Hebert, 1989).CHINOIL used discussion forums, training programs, and professional experts to influence the IJV decision-making.As for the procurement activities associated with the drilling operation, CHINOIL engineering department used informal social control because they knew that the IJV drilling department would take their words seriously even if they merely said 'You probably need to consider…'.This finding is a result of the difference in social rank between CHINOIL and the IJV and of social exchange rules.
The control mechanism used by CHINOIL Exploration Department is a content-oriented mechanism through which CHINOIL top managers exerts their direct intervention on IJV activities.This is a result of the hierarchical relationship between the parent and the IJV.Although the discussion forums used by CHINOIL looks like a process-oriented control, the essence is not.The IJV follows the documents and procedures of CHINOIL.
Proposition 8: The extent and mechanisms of inside control vary upon the strategic and economic importance of control targets as perceived by the parent.

Discussion
We add to the study of parent control over IJVs.Our core theoretical contribution is a holistic classification of parent control into two modes of control-inside and outside control based on the nature of the relationships between parents and IJV management groups.This classification provides scholars and practitioners with insight into IJV control.The differences between inside and outside parent control help us understand why the recruitment and staffing of IJV's top management positions represents a crucial strategic control mechanism for a JV parent (Geringer & Frayne, 1990;Yan & Gray, 1994).In analyzing different control mechanisms, Wang, Wee and Koh (1998) find that the most important mechanism for all sample firms was "the right to appoint the general manager", followed by "the right to appoint personnel to critical functional areas".By staffing key personnel, a parent can develop more knowledge of the events and circumstances affecting IJV (Tretiak & Holzmann, 1993), and those employees are likely to act in ways that are acceptable to the parent firm (Killing 1983).Based on social exchange theory, everyone expects the appointed employees to have an allegiance to the appointing partner (Groot & Merchant, 2000).Our study indicates that through inside control, the parent gains the adaptability and flexibility to exercise its control over those IJV activities that are strategically important to the parent.
Our second theoretical contribution is the insight into parent control power.Inside control is an endogenous, contingent form of control determined by a parent's equity share.In contrast, outside control is a prefixed form of control determined by a parent's overall bargaining power.Our study suggests that the majority party can effectively exercise its control through its ownership and the minority party can exercise its control through its overall bargaining power.Percent equity share warrants a parent inside control power through appointing its own people to the key positions in the IJV.Previous studies have indicated that IJV board composition and selection of people to key positions are determined by parent ownerships (Groot & Merchant, 2000;Mjoen & Tallman, 1997;Wang et al., 1998;Yan & Gray, 1994).However, ownership alone provides a parent only inside control power and not overall control.The overall control power of a parent includes both inside and outside control.Studying the four joint ventures between the United States and the People's Republic of China (equity share: 51/49, 49/51, 50/50, 50/50), Yan and Gray (1994) find that a parent's overall control power is determined by its overall bargaining power, which is consistent with Mjoen and Tallman (1997)'s study.Our study suggests that by increasing overall bargaining power a parent would increases its overall control power through increasing outside control power, not inside control power.Determining factors of inside and outside control offer both scholars and practitioners insight into how parents seek control power in an IJV based on what they contribute to the IJV.

Insert Table 2 about here
Our third theoretical contribution is how parents exercise their inside and outside control.As listed in Table 2, inside control is broad but selective, whereas outside control is narrow and specific.Inside control is flexible in terms of control extent and mechanisms.Selection of inside control focus, extent, and mechanism depends on the strategic and economic importance of the particular element in questions as perceived by the parent.Inside control power offers a parent adaptability and flexibility in IJV control.In contrast, outside control is predefined, centralized and content-oriented control in our study.It provides a party with an ability to control the other party through formal procedures.A minority partner shall seek outside control as its effective IJV control.Our findings are in agreement with other studies in that the majority party use selective control as an effective form of control over those dimensions that it perceives as critical and the minority party seek narrow, specific content -oriented control (Geringer & Hebert, 1989).
Our last but not at least contribution is a dynamic and systematic view of IJV control through longitudinal data study of the three dimensions of control.Such an approach enables us to gain a close look at IJV control and helps us decode the inconsistent or even contradictory results in the literature on IJV control and performance.IJVs and IJV control are not static.And IJV management teams are a key node in IJV control through which parents influence IJV activities by forming a dynamic interaction chain among parent -IJV management team -IJV activities.Without IJV management team in the equation, parent control falls into a gap.

Conclusion
Given that an IJV has two hierarchical parents and that its management team is made up of members who represent different parent interests, the study of parent control should be tailored to IJV management team composition.The previous research is limited at the parental level and neglect differences at the IJV level.
Our aim was to develop holistic insight into parent control.For the first time in the literature, parent control is separated into inside and outside forms of control based on parent relationships with individual IJV management groups.This classification helps us better understand IJV control.The differences between inside and outside control on operational conditions, determinants, and three dimensions of control (control focus, extent, and mechanisms) provide an insight view of the essence of IJV control and a practical guidance for practitioners in how to seek and balance IJV control.Although our study is focused on a single case, it presents all the standard characteristics of IJVs.These findings may be generalizable to other JVs.The next step will be an empirical test.
In conclusion, our most important contribution is that we bring the nature of the relationship between parents and respective management groups into the IJV control equation, unveiling two distinctive forms of control within IJVs: parent inside and outside control.The results add to the theory on IJV control and offer practical guidance to practitioners.

Figure 1 .
Figure 1.Theoretical model of inside and outside control in IJVs (PA -parent A; PB -parent B; MTmanagement team)