Implications for Business
The implications of the theories of horizontal and vertical FDI for business practice are relatively straightforward. First, the location-specific advantages argument associated with John Dunning does help explain the direction of FDI, both with regard to horizontal and vertical FDI. However, the argument does not explain why firms prefer FDI to licensing or to exporting. In this regard, from both an explanatory and a business perspective, perhaps the most useful theory is the market imperfections approach. With regard to horizontal FDI, this approach identifies with some precision how the relative rates of return associated with horizontal FDI, exporting, and licensing vary with circumstances. The theory suggests that exporting is preferable to licensing and horizontal FDI as long as transport costs are minor and tariff barriers are trivial. As transport costs and/or tariff barriers increase, exporting becomes unprofitable, and the choice is between horizontal FDI and licensing. Since horizontal FDI is more costly and more risky than licensing, other things being equal, the theory argues that licensing is preferable to horizontal FDI. Other things are seldom equal, however. Although licensing may work, it is not an attractive option when one or more of the following conditions exist: (a) the firm has valuable know-how that cannot be adequately protected by a licensing contract, (b) the firm needs tight control over a foreign entity to maximize its market share and earnings in that country, and (c) a firm's skills and know-how are not amenable to licensing. Figure 6.6 presents these considerations as a decision tree. Figure 6.6
A Decision Framework
Firms for which licensing is not a good option tend to be clustered in three types of industries:
High-technology industries where protecting firm-specific expertise is of paramount importance and licensing is hazardous.
Global oligopolies, where competitive interdependence requires that multinational firms maintain tight control over foreign operations so that they have the ability to launch coordinated attacks against their global competitors (as Kodak has done with Fuji).
Industries where intense cost pressures require that multinational firms maintain tight control over foreign operations (so they can disperse manufacturing to locations around the globe where factor costs are most favorable to minimize costs).
Although empirical evidence is limited, the majority of the evidence seems to support these conjectures.22
Firms for which licensing is a good option tend to be in industries whose conditions are opposite to those specified above. Licensing tends to be more common (and more profitable) in fragmented, low-technology industries in which globally dispersed manufacturing is not an option. A good example is the fast food industry. McDonald's has expanded globally by using a franchising strategy. Franchising is essentially the service-industry version of licensing--although it normally involves much longer-term commitments than licensing. With franchising, the firm licenses its brand name to a foreign firm in return for a percentage of the franchisee's profits. The franchising contract specifies the conditions that the franchisee must fulfill if it is to use the franchisor's brand name. Thus, McDonald's allows foreign firms to use its brand name as long as they agree to run their restaurants on exactly the same lines as McDonald's restaurants elsewhere in the world. This strategy makes sense for McDonald's because (a) like many services, fast food cannot be exported, (b) franchising economizes the costs and risks associated with opening foreign markets, (c) unlike technological know-how, brand names are relatively easy to protect using a contract, (d) there is no compelling reason for McDonald's to have tight control over franchisees, and (e) McDonald's know-how, in terms of how to run a fast food restaurant, is amenable to being specified in a written contract (e.g., the contract specifies the details of how to run a McDonald's restaurant).
In contrast to the market imperfections approach, the product life-cycle theory and Knickerbocker's theory of horizontal FDI tend to be less useful from a business perspective. These two theories are descriptive rather than analytical. They do a good job of describing the historical pattern of FDI, but they do a relatively poor job of identifying the factors that influence the relative profitability of FDI, licensing, and exporting. The issue of licensing as an alternative to FDI is ignored by both of these theories.
Finally, with regard to vertical FDI, both the market imperfections approach and the strategic behavior approach have some useful implications for business practice. The strategic behavior approach points out that vertical FDI may be a way of building barriers to entry into an industry. The strength of the market imperfections approach is that it points out the conditions under which vertical FDI might be preferable to the alternatives. Most importantly, the market imperfections approach points to the importance of investments in specialized assets and imperfections in the market for know-how as factors that increase the relative attractiveness of vertical FDI.
Chapter Summary
This chapter reviewed theories that attempt to explain the pattern of FDI between countries. This objective takes on added importance in light of the expanding volume of FDI in the world economy. As we saw early in the chapter, the volume of FDI has grown more rapidly than the volume of world trade in recent years. We also noted that any theory seeking to explain FDI must explain why firms go to the trouble of acquiring or establishing operations abroad when the alternatives of exporting and licensing are available.
We reviewed a number of theories that attempt to explain horizontal and vertical FDI. With regard to horizontal FDI, it was argued that the market imperfections and location-specific advantages approaches might have the greatest explanatory power and therefore be most useful for business practice. This is not to belittle the explanations for horizontal FDI put forward by Vernon and Knickerbocker, since these theories also
have value in explaining the pattern of FDI in the world economy. Still, both theories are weakened by their failure to explicitly consider the factors that drive the choice among exporting, licensing, and FDI. Finally, with regard to vertical FDI, it was argued that the strategic behavior and market imperfections approaches both have a certain amount of explanatory power.
This chapter made the following points:
Foreign direct investment occurs when a firm invests directly in facilities to produce a product in a foreign country. It also occurs when a firm buys an existing enterprise in a foreign country.
Horizontal FDI is FDI in the same industry abroad as a firm operates at home. Vertical FDI is FDI in an industry abroad that provides inputs into a firm's domestic operations.
Any theory seeking to explain FDI must explain why firms go to the trouble of acquiring or establishing operations abroad when the alternatives of exporting and licensing are available.
Several factors characterized FDI trends over the past 20 years; (1) there has been a rapid increase in the total volume of FDI undertaken; (2) there has been some decline in therelative importance of the United States as a source for FDI, while several other countries, most notably Japan, have increased their share of total FDI outflows; (3) an increasing share of FDI seems to be directed at the developing nations of Asia and Eastern Europe, while the United States has become a major recipient of FDI; and (4) there has been a notable increase in the amount of FDI undertaken by firms based in developing nations.
High transportation costs and/or tariffs imposed on imports help explain why many firms prefer horizontal FDI or licensing over exporting.
Impediments to the sale of know-how explain why firms prefer horizontal FDI to licensing. These impediments arise when: (a) a firm has valuable know-how that cannot be adequately protected by a licensing contract, (b) a firm needs tight control over a foreign entity to maximize its market share and earnings in that country, and (c) a firm's skills and know-how are not amenable to licensing.
Knickerbocker's theory suggests that much FDI is explained by imitative strategic behavior by rival firms in an oligopolistic industry. However, this theory does not address the issue of whether FDI is more efficient than exporting or licensing for expanding abroad.
Vernon's product life-cycle theory suggests that firms undertake FDI at particular stages in the life cycle of products they have pioneered. However, Vernon's theory does not address the issue of whether FDI is more efficient than exporting or licensing for expanding abroad.
Dunning has argued that location-specific advantages are of considerable importance in explaining the nature and direction of FDI. According to Dunning, firms undertake FDI to exploit resource endowments or assets that are location-specific.
Backward vertical FDI may be explained as an attempt to create barriers to entry by gaining control over the source of material inputs into the downstream stage of a production process. Forward vertical FDI may be seen as an attempt to circumvent entry barriers and gain access to a national market.
The market imperfections approach suggests that vertical FDI is a way of reducing a firm's exposure to the risks that arise from investments in specialized assets.
From a business perspective, the most useful theory is probably the market imperfections approach, because it identifies how the relative profit rates associated with horizontal FDI, exporting, and licensing vary with circumstances.
Critical Discussion Questions
In recent years, Japanese FDI in the United States has grown far more rapidly than US FDI in Japan. Why do you think this is the case? What are the implications of this trend?
Compare and contrast these explanations of horizontal FDI: the market imperfections approach, Vernon's product life-cycle theory, and Knickerbocker's theory of FDI. Which theory do you think offers the best explanation of the historical pattern of horizontal FDI? Why?
Compare and contrast these explanations of vertical FDI: the strategic behavior approach and the market imperfections approach. Which theory do you think offers the better explanation of the historical pattern of vertical FDI? Why?
You are the international manager of a US business that has just developed a revolutionary new personal computer that can perform the same functions as IBM and Apple computers and their clones but costs only half as much to manufacture. Your CEO has asked you to formulate a recommendation for how to expand into Western Europe. Your options are (a) to export from the United States, (b) to license a European firm to manufacture and market the computer in Europe, and (c) to set up a wholly owned subsidiary in Europe. Evaluate the pros and cons of each alternative and suggest a course of action to your CEO.
Closing Case Honda in North America
One of the most dramatic trends during the 1980s was the surge in Japanese direct investment in the United States. Leading this trend were the Japanese automobile companies, particularly Honda, Mazda, Nissan, and Toyota. Collectively these companies invested $5.3 billion in North American-based automobile assembly plants between 1982 and 1991. The early leader in this trend was Honda, which by 1991 had invested $1.13 billion in three North American auto assembly plants--two major plants in central Ohio and a smaller one in Ontario, Canada. Honda has invested an additional $500 million in an engine plant in Ohio that supplies its Ohio assembly plants. The company has also established major R&D and engineering facilities at its Ohio plants and has purchased an existing automotive test center--adjacent to the assembly plants--from the state of Ohio for $31 million.
As a result of these investments, Honda now employs 10,000 workers in its central Ohio plants and pumps a payroll of $7.3 million per week into the local economy. Of the 854,879 cars that Honda sold in the United States during 1990, nearly two-thirds were built at its three North American assembly plants--the vast majority of them in Ohio. Honda says the domestic content of its American-built cars is 75 percent, meaning that three-fourths of the final cost of a car is accounted for by North American labor, components, and other costs. The remaining 25 percent of the cost is accounted for by imported parts.
Honda had considered establishing auto assembly operations in North America as early as 1974 but ruled out investment then because of the high cost of North American labor. In 1977, Honda announced it had selected a site in the small town of Marysville, Ohio, for a motorcycle assembly plant. Motorcycle production would test the ground for the possible manufacture of automobiles. This experiment was deemed necessary because Honda's internal feasibility studies still predicted that high labor costs and poor productivity would make North American-based automobile production unprofitable. However, Honda quickly realized that its assumptions about US workers' poor productivity were unfounded, and in 1979 it announced plans to construct an automobile assembly plant adjacent to its Marysville motorcycle plant. Two years later, in
November 1982, the first US-built Honda was assembled, and by 1984 the plant was producing 150,000 automobiles per year.
Throughout the 1980s, Honda's direct investment in North America produced complementary investments by many of its Japanese suppliers of component parts. By 1989, at least 29 major Japanese supplier companies had established transplant manufacturing facilities in Ohio to supply Honda with component parts. In addition, 33 other Japanese firms had invested in the United States to supply Honda and several other Japanese and US automobile manufacturers. Honda required many of these companies to build their plants close to its Ohio complex so they could introduce a just-in-time production system, in which parts are delivered to the assembly plants just as they are needed. This technique virtually eliminates the need to hold in-process inventories and is regarded as a major cost savings. In addition, Honda wanted major suppliers close by so they could conveniently collaborate on the design of major components and on techniques for reducing costs and boosting quality.
A number of concerns seem to underlie Honda's decision to invest in North America. First, it is widely assumed that many Japanese firms, including Honda, did this largely to circumvent the threat of protectionist trade legislation, which seemed very real following the rapid increase in Japanese automobile exports to North America during the 1970s and early 80s. The threat of protectionism--especially the 1981 Voluntary Restraint Agreement under which Japanese companies agreed not to further increase their imports into the United States--may have accelerated Honda's late-1980s investments in Ohio. A second concern was probably the sharp rise in the value of the Japanese yen against the US dollar during 1987. This dramatically increased the cost of exporting both finished automobiles and component parts from Japan to North America. This also may have accelerated Honda's investments in the late 1980s.
However, it is also necessary to consider Honda's investment in North America in the context of its long-term corporate strategy. As a latecomer to automobile production in Japan, Honda had always struggled to be profitable in the intensely competitive Japanese auto industry. Against this background, Honda's North American assembly plants can be seen as part of a strategy designed to circumvent Toyota and Nissan and to make major inroads in the United States market ahead of its Japanese rivals. Underlying this strategy was Honda's strong belief that products need to be customized to the requirements of local markets. To paraphrase Hideo Sugiura, the former chairman of Honda, there are subtle differences, from country to country and from region to region, in the ways a product is used and what customers expect of it. If a corporation believes that simply because a product has succeeded in a certain market it will sell well throughout the world, it is likely destined for large and expensive errors or even failure. To produce products that account for local differences in customer tastes and preferences, Sugiura claimed that a company needed to establish top-to-bottom engineering, design, and production facilities in each major market in which it competed. Thus, in the late 1970s, Honda decided to invest in North America. Its success can be judged by the fact that although it was only the fourth largest automobile manufacturer in Japan in 1990 (with 9.3 percent of the market, compared to Toyota's 32.5 percent), it was the second largest Japanese automobile manufacturer in the United States (with 6.14 per
cent of the market, compared to first-place Toyota's 7.6 percent).
http://www.honda.com
Source: A. Mair, R. Florida, and M. Kenney, "The New Geography of Automobile Production: Japanese Transplants in North America," Economic Geography 64 (1988), pp. 352 - 73; H. Sugiura, "How Honda Localizes Its Global Strategy," Sloan Management Review, Fall 1990, pp. 77 - 82; S. Toy, N. Gross, and J. B. Treece, "The Americanization of Honda," Business Week, April 25, 1988, pp. 90 - 96; and P. Magnusson, J. B. Treece, and W. C. Symonds, "Honda: Is It an American Car?" Business Week, November 18, 1991, pp. 105 - 9.
Case Discussion Questions
Drawing on the market imperfections approach to FDI, explain why Honda chose to invest in production facilities in the United States, as opposed to contracting with an established US auto company to produce its cars under licensing in the United States?
Which of the theories of FDI reviewed in this chapter best explain Honda's FDI into the United States?
Are there aspects of Honda's investment in the United States that are not explained by the theories of FDI reviewed in this chapter? What are these aspects and how would you explain them?