Modes of Foreign Entry: a Transaction Cost Analysis and Propositions

A "frontier issue" in international marketing is the appropriate choice of entry mode in foreign markets. The objective of this paper is to offer a transaction cost framework for investigating the entry mode decision. This framework provides 1) a theoretical basis for systematically interrelating the literature into propositions, 2) propositions about interactions which resolve the apparently contradictory arguments advanced to date. Specifically, the paper: * illustrates the feasibility of clustering 17 entry modes into the degree of control the mode provides the entrant; * proposes that the most appropriate (i.e., most efficient) entrymode is a function of the tradeoff between control and the cost of resource commitment * advances testable propositions delimiting the circumstances under which each mode maximizes long-term efficiency. The entry mode literature is reviewed in the context of these propositions, and guidelines are derived for choosing the appropriate mode of entry, given certain characteristics of the firm, the product, and the environment. *Erin Anderson is Assistant Professor of Marketing at the Wharton School, University of Pennsylvania, which she joined in 1981. She received the Ph.D. from the UCLA Graduate School of Management. She has published in the Rand Journal of Economics, Marketing Science, and the Sloan Management Review. **Hubert A. Gatignon has been an Assistant Professor of Marketing at the Wharton School, University of, Pennsylvania since 1981. He obtained an MBA from UCLA in 1975 and a Ph.D. in Management from UCLA in 1981. Professor Gatignon has published in Marketing Science, the Journal of Marketing Research, the Journal of Consumer Research, and the Journal of Marketing. The authors gratefully acknowledge the financial support of the Wharton Center for International Management Studies. The comments of Yoram Wind, Jean-Francois Hennart, and three anonymous reviewers, as well as Charles Goodman and Leonard Lodish, are greatly appreciated, as is the assistance of Shari Powell. Received: August 1985; Revised: November 1985 & March 1986; Accepted: March 1986. Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve, and extend access to Journal of International Business Studies www.jstor.org ® 2 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, FALL 1986 A firm seeking to perform a business function (e.g., production management, distribution) outside its domestic market must choose the best "mode of entry" (institutional arrangement) for the foreign market. The would-be entrant faces a large array of choices, including: a wholly-owned subsidiary, a joint venture (in which the entrant could be majority, equal, or minority partner), or a nonequity arrangement such as licensing or a contractual joint venture. The impact of entry modes on the success of foreign operations is great, leading Wind and Perlmutter (1977) to identify entry modes as a "'frontier issue" in international marketing. Entry modes differ greatly in their mix of advantages and drawbacks. The tradeoffs involved are difficult to evaluate and little understood. Several surveys of how firms actually make the entry mode decision (reviewed in Robinson 1978) indicate that few companies make a conscious, deliberate cost/benefit analysis of the options. What is the best mode of entry for a given function in a given situation? Despite the existence of relevant evidence, the literature does not suggest how the manager should weigh tradeoffs to arrive at a choice that maximizes risk-adjusted return on investment. Instead, much of the literature contains many seemingly unrelated considerations, with no identification of key constructs. Often, a consideration is mentioned as part of a case study, with little indication of how the factor should affect other situations. Further, relevant work is scattered across books and journals in several disciplines, obscured by varying terminology, and separated by differences in problem setup, theory, and method.' The objective of this paper is to develop a theory, expressed in testable propositions, for integrating the literature on entry into a unified framework. The theory, which comes from industrial organization, is explicitly concerned with weighing tradeoffs and with maximizing an economic criterion: long-term efficiency. In particular, the theory includes interactions between determinants of entry modes, interactions that help resolve contradictory arguments in the literature. This review develops testable propositions concerning the following question: Under what circumstances is an entry mode the most efficient choice in the long run? Efficiency in general terms is the ratio of output to input. In the international context we mean the entrant's long-run return on its investment in an entry mode, adjusted for risk. Hence, we address the impact of a mode on both the numerator (returns) and denominator (investment) over the long-time horizon.2 Section one of this paper categorizes modes of entry into varying degrees of control by the entrant. Section two presents a transaction cost theory of entry modes, which generates a set of propositions. Entry mode research is reviewed in the context of these propositions. The paper concludes with suggestions for empirical research. MODES OF FOREIGN ENTRY 3 MODES OF ENTRY AND CONTROL The classical approaches to long-term strategic decisions, such as entry mode choice, emphasize choosing the option offering the highest riskadjusted return on investment in the feasible set. Yet, the literature on the entry mode choice makes little direct mention of risk or return. Instead,' the issue is structured in terms of the degree of control each mode affords the entrant (Daniels, Ogram, and Radebaugh 1982, Robinson 1978, Robock, Simmonds, and Zwick 1977, Vernon and Wells 1976). But why such emphasis on control? The Preeminent Rok of Control Control (the ability to influence systems, methods, and decisions) has a critical impact on the future of a foreign enterprise. Without 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 (Davidson 1982). Further, the entrant can use its control to obtain a larger share of the foreign enterprise's profits. In short, control is a way to obtain a higher return. Yet control, while obviously desirable, carries a high price (Vernon 1983). To take control, the entrant must assume responsibility for decisionmaking, responsibility a firm may be unwilling or unable to carry out in an uncertain foreign environment. Control also entails commitment of resources, including high overhead. This in turn creates switching costs, reducing the firm's ability to change its institutional arrangement should its choice turn out to be suboptimal. Resource commitment also increases the firm's exposure, i.e., the possibility of losses due to currency changes (Davidson 1982). Thus, to assume control is also to assume some forms of risk. Control, then, is the focus of the entry mode literature because it is the single most important determinant of both risk and return. High-control modes can increase return and risk. Low-control modes (e.g., licenses and other contractual agreements) minimize resource commitment (hence risk) but often at the expense of returns. Firms trade various levels of control for reduction of resource commitment in the hope of reducing some forms of risk while increasing their returns. Hence, focusing on control is consistent with the classical risk-adjusted return perspective. The viewpoint adopted in this paper is that international entry mode choices are most usefully and tractably viewed asa tradeoff between control and the cost of resource commitments, often under conditions of considerable risk and uncertainty. Preserving flexibility should be a major consideration of most firms in making the tradeoff. Flexibility, the ability to change systems and methods quickly and at a low cost, is always an important consideration, particularly in lesser-known foreign markets (where the entrant is likely to change systems and methods as it learns the 4 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, FALL 1986 new environment). This view is consistent with Holton (1971), who argues that control, risk, and flexibility are principal considerations (Mascarenhas 1982). Classifying Modes of Entry The objective of this review is to suggest major factors that determine what degree of control maximizes long-run efficiency. The theory of the efficiency of modes of entry treated in section two relies on the existence of a mapping to a control dimension but not on any particular mapping. For purposes of illustration, we suggest, in this section, a mapping from entry modes to the degree of control they afford the entrant. As a caveat, there are many ways to gain control and many variations within any one form of entry mode (Kindleberger 1984, Hayashi 1978). For example, a minority partner might exercise influence out of proportion to ownership, due to such factors as a special contractual arrangement, expertise, or status as a government body. Hence our discussion is very general, and exceptions to our mapping can be found. Consequently, this discussion is intended to demonstrate the feasibility of a mapping, empirical tests of which are independent from the theory proposed in section two. Indeed, a valuable research contribution would be the development of a detailed theory of the relationship between control and governance structure. A Suggested Clustering of Entry Modes Although there is no tested, accepted theory as to how much control each entry mode affords, both the "'management" (Root 1983) and the "economic" (Calvet 1981, Caves 1982) streams of research offer information as to the clustering of entry modes. Figure I illustrates how 17 entry modes can be grouped in terms of the amount of control (high, medium, low) an entrant gains over the activities of a foreign business entity. As shown in Figure 1, dominant equity interests (wholly-owned subsidiary or majority shareholder) are expected to offer the highest degree of control to the entrant (Root 1983, Davidson, 1982, Bivens and Lovell 1966, Friedman and Beguin 1971, Killing 1982). Balanced interests (plurality shareholder, equal partnership and balanced contracts) are shown as medium-control modes based on the notion of a "credible commitment" (Williamson 1983) or "hostage." Firms forming a venture with a high likelihood of trouble (such as equal partnerships) will have difficulty locating a suitable partner. To attract a partner, the entrant mayneed to put up something to lose, a sort of good-faith collateral, known as credible commitment. For example, in a slightly unbalanced venture, the over 50%-partner may concede favorable contract clauses (such as veto power). These clauses can be so favorable that a firm may have more control with a 49% share than a 51% share (Friedman and Beguin 1971). Or the commitment may be the most critical positions in the MODES OF FOREIGN ENTRY 5 Figure 1 Entry Mode Classified by the Entrant's Level of Control High-Control Modes: Dominant Equity Interests Wholly-owned subsidiary Dominant shareholder (many partners) Dominant shareholder (few partners) Dominant shareholder (one partner) Medium-Control Modes: Balanced Interests Plurality shareholder (many partners) Plurality shareholder (few partners) Equal partner (50/50) Contractual joint venture Contract management Restrictive exclusive contract (e.g., distribution agreement, license) Franchise Nonexclusive restrictive contract Exclusive nonrestrictive contract Low-Control Modes: Diffused Interests Nonexclusive, nonrestrictive contracts (e.g., intensive distribution, some licenses) Small shareholder (many partners) Small shareholder (few partners) Small shareholder (one partner) foreign entity: the exposed partner can demand to fill them with its own personnel, a method preferred by Japanese multinationals (Hayashi 1978). In a 50-50 relationship, the hostage is a peculiar one the venture itself. Friedman and Beguin (1971) point out that equality in equity capital can "lend a special feeling of partnership to the two partners" (p. 372), adding "the risk of deadlock itself acts as a powerful incentive to the partners, encouraging them to find solutions to disagreements by discussion and compromise" (p. 377). In certain nonequity modes, moderate control comes from daily involvement in the operation and from expertise. These modes include: -Contract management (an ongoing relationship) in which the entrant performs specified functions and in which the entrant has representation on the management committee that oversees the venture's activities, Contractualjoint ventures, Restrictive exclusive contracts, Franchising (a form of licensing in which the use of a business system is granted).3 Franchising offers medium control because the typical agreement includes incentives to adhere to the system's rules and allows a high degree of monitoring of the franchisee's activities. 6 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, FALL 1986 Contracts that are exclusive but nonrestrictive or nonexclusive but restrictive. Either restrictiveness or exclusivity give the entrant moderate control, though by different means. Restrictive contracts circumscribe the other party's freedom of action, while exclusive contracts (simultaneously a reward and a protection against competition) motivate the other party to cooperate (Stern and El-Ansary 1982). Low-control modes are ones in which the entrant has diffused interests. These include nonexclusive, nonrestrictive contracts (multiple unrestricted licenses and intensive distribution) and minority equity positions. We reiterate here that there are many ways to gain control. Our list is not exhaustive: In particular, we note that entrants may build stable relationships or networks with other parties in which the long-term interests of both parties allow the development of norms. Although this area has received relatively little research attention, developments have been made by Hakansson (1984) and Williamson (1985). We now turn to propositions concerning the degree of control that is most efficient for a variety of conditions. Given a ranking of entry models on control, it is then possible to recommend an entry mode for a given entry situation. A TRANSACTION COST ANALYSIS OF FOREIGN MODES OF ENTRY What is the best entry mode for a given setting? Obviously, a large number of factors bear on the answer. The intent of this review is to propose constructs and mechanisms derived from a unified theoretical framework. This framework is similar to the general approach of several new theories of foreign investment (Kindleberger 1984, Caves 1971, 1982, Hennart 1982, Rugman 1982), which concern why multinational firms exist. In this paper, we develop these theories for one specific issue: choice of mode of entry. Our analysis builds on the existing literature by proposing detailed relationships between constructs corresponding to the ideas presented in the more general theories. As we develop these empirically testable propositions, we contrast them with predictions from other, frequently more well-known frameworks to suggest how they differ and to spur empirical research designed to sort out competing predictions. We also review findings that bear on each proposition. Our mapping from governance structure to control (Figure 1) looks something like a progression from less integrated to more integrated. Williamson (1979) suggests that degree of integration proceeds from complete non-integration (classical marketing contracting between two parties) to complete integration (one entity "contracts" internally to perform a function), passing through intermediate points. Underlying this progression is the transference of authority from paper (a contract) to entities (arbitrators, parties to a transaction), culminating in the consolidation of authority by one party. This progression of authority is a growing degree of control. MODES OF FOREIGN ENTRY 7 Control and integration are closely related, since integration gives a firm legitimate authority to direct operations. Hence, we employ a theory of vertical integration to generate propositions about the desirability of various modes of entry offering various degrees of control. The theory, transaction cost analysis, combines elements of industrial organization, organization theory, and contract law to weigh the tradeoffs to be made in vertical integration (and by extension, degree of control) decisions. We begin with the assumption that the market being entered has at least enough potential that the firm can recoup the overhead of a high-control entry mode. If this is not the case, high-control modes are not worth considering (Williamson 1979). However, for markets large enough to break even on the fixed cost of a high-control mode, the entrant has a choice to make. In these circumstances, the efficiency of an entry mode depends on four constructs that determine the optimal degree of control, following a transaction cost analysis. These constructs are: 1. transaction-specific assets: investments (physical and human) that are specialized to one or a few users or uses; 2. external uncertainty: the unpredictability of the entrant's external environment; 3. internal uncertainty: the entrant's inability to determine its agents' performance by observing output measures; 4. free-riding potential: agents' ability to receive benefits without bearing the associated costs. Figure 2 A Transaction Cost Framework For Analyzing the Efficiency of Entry Modes (RNACTION-SPECIFIC ASSETS Pl P2 P3 P4i < \ C~~~~~~EGREE OF CONTROL3 t NENAL UNCERTAINTY ) FREE-RIDING POTENTIAL P9 8 JOURNAL OF INTERNATIONAL BUSINESS STUDIES, FALL 1986 Figure 2 is an overview of the framework, which shows that these four factors should be positively associated with the entrant's degree of control. The four factors, their rationale, and their corresponding propositions about entry modes are discussed one by one in this section. Figure 3 restates the propositions in a more accessible form and lists the conditions under which the modes in the high-control cluster are most appropriate. TRANSACTION-SPECIFIC ASSETS Transaction cost analysis approaches the entry mode question with the following promise: a low level of ownership is preferable until proven otherwise. We label this premise the "default" hypothesis. The default hypothesis accords with an assumption fundamental to economics, that is, that market outcomes tend to be efficient when competition is strong. Competitive pressures drive parties to perform effectively at low cost and to deal with each other in fairness, honesty, and good faith lest they be replaced. Hayashi (1978) gives the example of foreign sales agents competing to carry out distribution for a Japanese entrant. Hayashi finds that where competition among agents is active, the resulting business relationship is highly cooperative. In general, where suppliers of a good or service are readily available, a firm may take advantage of their expertise and economies of scope and scale in performing their specialized function by writing a contract with one supplier, confident that a new supplier may be found if the relationship is unsatisfactory (Williamson 1981b). Accordingly, firms are advised to avoid integration whenever the supplier market is competitive. In this way, the firm can have both a high return and low risk. By not integrating (or investing directly), a firm avoids the drawbacks of a company division. Overhead is minimized, as is company politics, communication distortion, and the possibility that an inside division will become obsolete or inefficient because it is shielded from the pressures of daily competition for contracts (Williamson 1975). Integration (or direct investment) is, however, justified when the market mechanism no longer encourages performance, i.e., when competitive pressure is low. Williamson (1979) argues that most transactions begin when competition is intense but some degenerate into lock in ("small numbers bargaining") when the contract partner becomes irreplaceable. Then the partner may extract new contract terms, become inflexible, and otherwise violate the letter and spirit of the agreement ("opportunistic" behavior, i.e., self-interest seeking with guile) with relative impunity. Degeneration into lock in occurs when "transaction-specific assets" of considerable value accumulate. These are investments (physical and human) that are valuable only in a narrow range of transactions, that is, specialized to one or a few users or uses (Williamson 198 lb). An example of a physical transaction-specific asset is a stamping machine to make MODES OF FOREIGN ENTRY a Ln Ln Ln Ln WD rm, u