Why Joint Ventures Die So Quickly

With the increasing competition in the global market, as well as the implications of 1992 in Europe, managers are realizing that just keeping abreast of technological and marketing breakthroughs will be an accomplishment in itself Already, the need to be on top of new technologies, to shift out of declining industries into growing ones, and to learn how to buy and sell in unknown markets is beyond the capabilities of many large international firms. It is because of this strain that managers are focusing on the formation of joint ventures.

Joint ventures have an attractive quality in a period of turbulence for they often serve to reduce or exploit the variance of the market by stabilizing competition, especially industries which have to cope with excess capacity. They also can be designed as a means of exploring new capabilities and opportunities. Ideally, when a venture shows promise, one of the partners may exercise an option to buy the venture; the other partner walks away with the capital gains.

My comments are directed, therefore, towards developing an understanding of how joint ventures serve to lessen competitive threats or provide options to respond to market and technological fluctuations. My analysis takes the perspective that a joint venture is cooperation within a competitive context.


Narrowly defined, a joint venture occurs when two or more firms pool a portion of their resources within a common legal organization. This is done by selecting among alternative modes the ways by which two or more firms can transact. Thus, a theory of joint ventures must explain why a particular mode of transacting is chosen over other alternatives such as acquisition, supply contract, licensing, or spot market purchases.

Reasons given by CEOs for choosing joint ventures range from the benefits of sharing risk or exploiting economies of scale and size, to the exchange of technologies and differential abilities. In many cases, these are promoted by governments who stipulate shared ownership as the only channel by which to invest in a country. The most common, however, are fear, profit and learning.


Usually labelled as “transaction cost theory,” the fear motive can be boiled down to the following: whenever two firms transact on a long-term basis, problems arise from the difficulty of setting future prices, guaranteeing quality and delivery, and safeguarding technological and strategic decisions.

No matter how well contracts are designed, they may fail to provide effective guarantees. The necessity of stipulating contractual conditions increases, of course, with the complexity of the transaction and the difficulty of monitoring behavior. Contractual clauses regulating the development process are particularly troublesome to write and to enforce. A supplier which initially gives a low price, for example, may claim unexpected costs in developing a new process. The fears in relying on an outside supplier are heightened when the buyer must design around specific components and is, thus, precariously dependent upon the goodwill in the relationship.

A joint venture is frequently seen as the best alternative. By requiring mutual commitment of investment, it provides incentives for both parties to perform according to their obligations. An analogy, which is well-understood in the context of broader East-West relations, is the principle of nuclear determent, whereby stability is maintained by holding both sides “hostage.” Similarly, a joint venture holds each other’s investment vulnerable to loss in the case of breach of contract or poor performance.


Perhaps one of the strongest reasons for doing a joint venture is the pursuit of profit. Increased profitability can be gained by one of two ways: First, between firms in an oligopoly, joint ventures can stabilize competition and improve industry returns. There is considerable empirical research to confirm this. If you consider the recent joint ventures of Asea Brown Boverie and Westinghouse in the power generation business, and General Motors and Toyota in automobiles, aspects of oligopolistic cooperation are clearly present.

Second, enhanced profitability can be derived from the reduction of costs or the creation of new products and technologies which can influence the competitive positioning of the partners in their industries. For example, the tie-up between General Motors and its Korean partner (in which it has a minority investment) facilitates the export of low-cost vehicles from Korea through a well-enhanced distribution network in the U.S. But the cooperation between the two companies also serves to slow the penetration of untamed Japanese competitors, who are seeking to upgrade their auto lines into higher-priced levels on the basis of profits earned on their commodity vehicle sales.

Of course, fear and profit are not mutually exclusive. Many of the concerns noted earlier are also relevant when firms cooperate by contract, by merging or by frequent lunches between the top management, in order to improve their strategic positioning. Yet, if the cooperation entails the revealing of secrets, the transfer of technologies, or the sharing of brand labels, the fear of the misuse of these assets will drive the partners to seek ways to enforce compliance with the agreement.



Amid the dour discussion on how fear motivates cooperation, it is important to balance this perspective by considering the role of joint ventures in creating and transferring knowledge among firms. This explanation views joint ventures as a means bywhich firms learn new, or seek to retain old, capabilities. Sharing knowledge is especially important in ventures between firms from different industries who seek to pool their distinct competences.

An example is the venture between Honeywell and Ericsson in the development of a telecommunications office switch for the American market. Honeywell had considerable in-house expertise in software features desired by the end user, as well as the ability to run a development facility in the U.S. Ericsson, in turn, had the switch technology, and several years of experience in the international development and sales of the product. The development efforts resulted in a product which is fully adapted to the American market.

Of course, knowledge can be transferred by other means than a joint venture, such as through a license or outright sale. Again, we are confronted with understanding why a joint venture is a better way to transfer some kinds of knowledge. However, the choice in this case may not be driven by fear, but by the difficulty of transferring knowledge which is organizational in character. One reason why joint ventures are commonly used among firms in international markets is, in fact, to exchange the distinctive managerial skills of countries.


There has been considerable emphasis paid to joint ventures in the media as heralding a new era of cooperation and strategic alliances.

Yet, as the above motives suggest, cooperation clearly occurs within a competitive context. Not only does a joint venture face the usual competition of the market, but it also is caught in the conflict between the partners. As a result, ventures are unusually fragile.

Some insight into this fragility can be gained from Figure I. This table gives hazard rates for a sample of 92 manufacturing joint ventures located in the U.S. A hazard rate is the ratio of those ventures which terminated during a year over those ventures at risk. The number at risk is the initial number of joint ventures (i.e., 92), minus those which terminated in years one and two, and those which are still alive but have not yet reached three years of longevity.


Using this data, I was able to investigate the factors which led to earlier terminations, making a distinction between dissolution and acquisition. Dissolution clearly represents a business failure on the one hand, but can also reflect a fundamental conflict among the partners. Acquisitions suggest that one of the parties, or an outside party, places a higher value on the venture. It is less a statement about business failure or conflict than differences in valuation.

To test some hypotheses on the causes of termination, I applied a hazard model which has been most commonly used in medical studies. These studies ask, for example, how smoking increases or decreases the likelihood of cancer. A similar approach can be made in the study of joint ventures, where the division of equity shares, for example, influences the speed at which the ventures terminate.

Based on this approach, some interesting statements can be made. Acquisitions of joint ventures are more likely to occur under two kinds of industry conditions: few competitors and unexpected growth. In concentrated industries (i.e., industries where a few firms dominate), joint ventures often serve two functions: one is to restrict output and thus avoid price wars; the second function occurs when one firm can no longer go it alone in the industry, but is not willing to divest entirely.

The Asea Brown Boveri venture with Westinghouse is a good example. The industry is concentrated and mature. Therefore, in the case of termination of the venture, it benefits the industry that one of the parties acquires the operation, rather than both parties investing separately in further plant capacity. More interestingly, the venture contract gives ABB a call option to buy the venture and Westinghouse a put option to sell. The joint venture is, then a likely divestiture on the part of Westinghouse, which has agreed to the arrangement either because it cannot commit to divest, or it provides a hand-holding service until ABB can run the U.S. operations alone.

The analysis of dissolutions presents a radically different profile. A primary factor in dissolution is whether or not the parties have other business agreements. Like the mutual hostage positions described earlier, the evidence shows that ventures are more stable if the threat of dissolution is deterred by its possible impact on other relationships. Stability, in other words, is stronger between partners who have a history of partnership.

The study of joint venture termination provides two fundamental findings: first, ventures are often options to divest or to expand depending on the market; second, their stability is strongly affected by the familiarity and commitment of the partners.


An understanding of the reasons why joint ventures end provides a few sound rules on how they should be designed. There is no fail-safe rule for designing any business plan and its implementation. But a joint venture is especially difficult for one simple reason: it is under the ownership of more than one firm.

Joint venture contracts do not make good reading. It is unlikely that any advice can make these contracts slimmer, but the following considerations might eliminate the need to reread the terms as often:

Design the venture to guarantee your sleep. In every venture, there are worries about the loss of control over technologies and brand labels.

One way to eliminate the problem of technology leakage is to begin with the assumption that a leak will occur. What is it worth to you? If there is a price at which the technology can be sold, then sell it as part of the capital contribution of your firm to the venture. If it is, in a sense, priceless, then do not share it.

The questionable quality of a joint venture’s products can cause serious damage to brand reputations. In their venture, GM-Toyota simply agreed not to share or create a common brand label. Nor did Honeywell and Ericsson share a brand label. Market reputations are hard and expensive to establish. Sharing brand labels is often necessary, but some loss of sleep can be expected.

Do not burden the joint venture. Cooperative arrangements are frequently a spectacle of each member trying to do the least amount of work; it is logical that partners will try to make sure they are getting something out of the venture. Therefore, the temptation is to burden the venture with excessive channels of remuneration, such as through the setting of transfer prices on goods sold to or bought from the venture, or by imposing licensing fees and royalties.

There is a fundamental problem with such a policy: the financial evaluation of the venture will look very different to each partner. Unless the venture is healthy, not only will one partner be more upset than the other, the management of the venture itself will be jeopardized.

Design the venture so that both partners are equally interested in the profitability of the concern. The only way to structure such an outcome is to let dividends be the primary channel by which profits are divided. Backdoor channels should only be encouraged when governments impose rules on how equity is shared at different rates, for example, on dividends and licensing fees.

Choose the right benchmark for evaluation. Joint ventures are often not popular decisions. Engineers are upset that technology is “given away” or bought from the outside; business managers worry about the effect on competitive positioning. Thus, negotiators face a second negotiation at home. The common temptation is to oversell the venture, a temptation especially appealing when someone else will be assigned to manage the cooperation. No wonder that one joint venture president said the first thing one should do before starting the job is to tear up the business plan and start again.

Unreasonable expectations will also plague the long-term evaluation of the venture. It is essential that the joint venture managers remember why they did it in the first place. Consider all the heat and discussion over whether American firms have allegedly been robbed blind by Japanese partners. The recent agreements signed by Honeywell in computers, Firestone in tires, Dresser in trucks, and Westinghouse in power generation suggest another interpretation: were these firms no longer willing to go it alone? In this case, the benchmark should be how the return on the venture compares to an outright divestment.

Build for the future. No rule will assure the success of a joint venture. As in a business, the working out of conflict and challenges requires the participation of effective managers. It is important that the venture be supported by a wider relationship among the partners.

A harder task is to work out how the relationships of each partner-with other players (and potential, or existing, competitors) in the industry-affect the cooperation. But as many firms have learned to their dismay, the growing number of joint ventures leads to unusual patterns of coalitions in an industry.

In a time of industry redefinition, it is not surprising that the spate of joint ventures should lead to competitors who, if not directly cooperating, are only once-removed in their kinship. No joint venture should be accomplished without an analysis of the wider industry cooperative relationships as they stand today, and as they may evolve tomorrow.

Bruce Kogut, who received his PhD from MIT in 1983, and who has taught at the Stockholm School of Economics in Sweden, is currently an Assistant Professor of Management at the Wharton School. The author would like to thank AT&T for its support given under the auspices of the Reginald H. Jones Center for Management Policy, Strategy and Organization.

Chief Executive magazine (published since 1977) is the definitive source that CEOs turn to for insight and ideas that help increase their effectiveness and grow their business. Chief Executive Group also produces e-newsletters and online content at chiefexecutive.net and manages Chief Executive Network and other executive peer groups, as well as conferences and roundtables that enable top corporate officers to discuss key subjects and share their experiences within a community of peers. Chief Executive facilitates the annual “CEO of the Year,” a prestigious honor bestowed upon an outstanding corporate leader, nominated and selected by a group of peers, and is known throughout the U.S. and elsewhere for its annual ranking of Best & Worst States for Business. Visit www.chiefexecutive.net for more information.