INTERNATIONAL JOINT
VENTURES AND THE U.S.
AUTO INDUSTRY
Darwin Wassink
Robert Carbaugh

In 1983 General Motors Inc. and Toyota Inc. formed a joint venture, the New United Motor Manufacturing Inc., to assemble auios in the United
States. For Toyota, the venture was a first attempt to locate production in
America. General Motors viewed the venture as a means of learning how to produce low-cost, high quality, small vehicles. Facing an onslaught of anti-union Japanese firms, the United Auto Workers had to demonstrate that unions would not be an impediment to Japanese production in the
United States. By 1986 the venture was termed a success. This paper considers the welfare effects of international joint ventures among compettng manufacturers, as applied to the U.S. auto industry.

Darwin Wassink is Professor of Economics at the University of WisconsinEau Claire. Previously he served as an economist in Pakistan and Saudi
Arabia.
Robert Carbaugh is Associate Professor of Economics at Central
Washington University. He is author of International Economics and coauthor of The International Monetary System.
ISSN: 088i~390H. THE iNTERNATlONAL

TRADEJOVRNAL,Volu>ne

I No. I. hall 1986

47

48

THE INTERNATIONAL TRADE JOURNAL

The American auto industry is undergoing an evolution in which the
"all American car" is rapidly becoming a thing of the past. Although
American automakers will continue to develop and build their own mid-size and large autos in the United States, they are turning over increasing amounts of small-car production to foreign competitors, mainly the
Japanese. This process intensifies the demands for protectionism among
American auto workers. Reflecting this trend, in 1984 General Motors Inc. and Toyota Inc. of Japan formed a joint venture which assembles subcompact autos for sale in the United States, This paper examines possible welfare effects of international joint ventures among competing auto producers,
INTERNATIONAL JOINT VENTURES
The trend coward the internationalization of the American auto industry is documented by an increasing variety of commerical interrelationships between American and foreign automakers. For example.
General Motors owns 34 percent of Isuzu and 5 percent of Suzuki; Ford owns 25 percent of Mazda and Chrysler owns 15 percent of Mitsubishi.
Internationalization also includes the production of foreign cars in the
United States --such as Accords being produced in the Honda plant in
Marysville, Ohio-- just as It does the production of American-brand cars overseas, such as the Ford Escort being manufactured in Western European plants. It involves the marketing of foreign-produced autos under domestic nameplates, such as the Chrysler "Colt" produced In Japan by Mitsubishi and shipped to the United States. The installation of foreign-produced components in U-S. automobiles also reflects the internationalization trend.
Finally, the General Motors-Toyota and Chrysler-Mitsubishi joint ventures, to assemble subcompact cars in the United States, indicate a fundamental shift in thinking about the way to produce and market automobiles.
If a company is to participate in interniitional business, it must enjoy some advantage over foreign firms. A cost advantage is necessary If the company is to export. Licensing is profitable only if the company owns a

Wassink

and Carbaugh:

I n t e r n a t i o n a l Joint

Ventures

...

4 9

production process or patent that can be transferred overseas for a fee.
Establishing foreign subsidiaries requires the company to have an advantage over local firms in the country where the subsidiary is established.
In a trend that accelerated during the 198O's, companies have linked up with their former rivals in a vast array of joint ventures, A joint venture is a business organization established by two or more companies that combine their skills and assets. It may have a limited objective (e.g. research or production) and be short lived. Joint ventures may be multinational in character. Involving cooperation among domestic and foreign companies.
Joint ventures differ from mergers in that they involve the creation of a new business firm, rather than the union of two existing companies. Table 1
(page 50) provides examples of recent joint ventures between American and foreign companies.
There are several types of international joint ventures. First is a joint venture formed by two firms which conducts a business in a third country.
For example, an American oil firm and a British oil firm may form a joint venture for oil exploration in the Middle East. Next is the formation of a joint venture with local private interests. Honeywell Information Systems
Inc. of Japan was formed by Honeywell Inc. of the United States and
Mitsubishi Office Machinery Co, of Japan to sell information systems, equipment to the Japanese. Last is a joint venture which includes local government participation. Bechtel Co. of the United States, MesserschmittBoeikow-Blom of West Germany, and The National Iranian Oil Co.
(representing the government of Iran) formed Iran Oil Investment Co, for oil extraction in Iran,
The joint ventures between LI,S, and Japanese auto manufacturers have been formed to provide benefits to both the US, and Japanese firms, U.S, auto manufacturers see the joint venture as a method of learning more about the efficiencies of Japanese production techniques. This, they believe, will allow them to reduce production costs and improve efficiency over those levels that would occur if they undertook production on their own. Japanese producers see these joint ventures a*s an additional mt-thod of penetrating

C/5

-Si

2

<

a.
O

-a o — —

2

c:

6 Ji

^
D. pa

e e

O a-

oi < w

z o w
Pi

O
Q

a:

o

c a, a

a

CL,

OH

c o a.

C

'S o c

o —
—

W
CO

-o

w
H

•?=

1
«

O4

"^

u

ii*

2 a

H

•-'

T.

^

O

V

^

C

1*

III

C

i>

X X O D

I a o

I I
—

O

•
I

I c o

2 ° c >
O
I—.

o oi c/)

C

-a c ? E °
2 0 2

o

3
C

,™

c

c

X o ^ H

Wassink and Carbaugh: Internationa/Joint Ventures .

the American market as well as, and probably more importantly as, a way to reduce their vulnerability to potential protectionist actions by the United
States.
International joint ventures can yield both welfare-increasing effects and welfare-decreasing effects for the domestic economy. Joint ventures lead to welfare gains when: (1) the newly-established joint venture adds to preexisting productive capacity and fosters additional competition; (2) the newly-established joint venture is able to enter new markets that neither parent could have entered individually; (3) the joinr venture yields cost reductions (e.g. economies of scale) chat would have been unavailable if each parent performed the same function separately. However, the formation of a joint venture may give rise to increased market power, suggesting greater ability to influence market output and price. This could especially occur when the joint venture is formed in markets in which the parents conduct business. Under such circumstances, the parents, through their representatives in the joint venture, agree on prices and output in the very market which they themselves operate. Such coordination of activities limits competition, reinforces upward pressure on prices, and lowers the level of domestic welfare.
The example below contrasts two situations: (1) where two competing firms sell autos in the domestic market; (2) where the two competitors form a joint venture which operates as a single seller (i.e. monopoly) in the domestic market. We would expect to see a higher price and smaller quantity in the case where the joint venture behaves as a monopoly. This will always occur as long as the marginal cost curve for the joint venture is identical to the horizontal sum of the marginal cost curves of the Individual competitors. The result of this market power ejfeet is a deadweight welfare loss for the domestic economy--a reduction in consumer surplus that is not offset by a corresponding gain to producers. If, however, the formation of the joint venture entails productivity gains that neither parent firm could realize prior to its formation, domestic welfare would be increased. This is because a smaller amount oi the .domestic economy s resources is now required to produce any given output. Whether domestic welfare rises or

32

THE INTERNATIONAL TRADE JOURNAL

falls because of the joint venture depends on the magnitudes of these two opposing forces.
Figure I (page 53) illustrates the welfare effects of two parent firms which form a joint venture in the market in which they operate. Assume that Sony Auto Co. of Japan and American Auto Co. of the United States are the only two firms producing autos for sale in the U.S. market. Suppose each firm realizes constant long-run costs, suggesting that average total cost equals marginal cost at each level of output. Let the cost schedules of each firm prior to the formation of the joint venture equal MCQ = ATCQ , which equal $10,000. MCQ = ATCQ thus becomes the long run market supply schedule of autos.
Assume that Sony Auto Co. and American Auto Co. initially operate as competitors, charging a price equal to marginal cost. In Figure 1, market equilibrium exists at point A where 100 autos are sold at a price of $ 10,000 per unit. Consumer surplus totals area a'^ b + c. Producer surplus does not exist given the horizontal supply curve of autos. Now suppose the two firms announce the formation of a joint venture known as JV Co., which manufactures autos for sale in the United States. Assume the autos sold by
JV Co. replace the autos sold by the two parents in the United States.
Suppose the formation of JV Co. entails new production efficiencies which result in cost reductions. Let JV Co.'s cost schedule, MC, = ATC, be located at $7,000. As a monopoly, JV Co. maximizes profit by equating marginal revenue with marginal cost. Market equilibrium exists at point B, where 90 autos are sold at a price of $12,000 per unit. The price increase leads to a reduction in consumer surplus equal to area a+ h.Of this amount, area a is transferred to JV Co. as producer surplus. Area b represents the loss of consumer surplus not transferred to JV Co., and becomes a deadweight welfare loss for the U.S. economy (i.e. consumption effect).
Against this deadweight welfare loss lies the efficiency effect of JV Co. which entails unit costs falling from $10,000 to $7,000 per auto. JV Co. can produce its profit-maximizing output, 90 autos, at a cost reduction equal to area d as compared to the costs that would exist If the parent firms

Wassink and Carbaugh: International joint Ventures . . .

FIGURE I
WELFARE EFFECTS OF
AN INTERNATIONAL JOINT VENTURE

12.000 -

—

Sony Auto Co.
10,000
American Auto Co.

7.000

- ATC|

JVCo,

Demand

J

I iOO Autos

54

THE INTERNATIONAL TRADE JOURNAL

produced the same output. Area d thus represents additional producer surplus which is a welfare gain for the U.S. economy. The analysis concludes that for the United States, the formation of JV Co. is desirable if area d exceeds area b.
It has been assumed that JV Co. entails cost reductions that are the basis for output and price determination (MCj ^ which are unavailable to either parent as a stand-alone company. Whether the cost reductions benefit the overall U.S. economy depends on many factors. If the cost reductions result from productivity improvements (e.g. new work rules leading to higher output per worker), a welfare gain exists for the economy since fewer resources are required to produce a given number of autos and can be shifted to other industries. However, cost reductions stemming from JV Co. s formation can be monetary in nature. To the extent that JV Co., being a newly-formed company, is able to negotiate wage concessions from domestic workers which could not be achieved by American Auto Co., the cost reductions would represent a transfer of dollars from domestic workers to the economy (consumers in terms of lower prices and JV Co. profits).
This share of the lower costs (area below the dashed line within d) is not a welfare gain for the United States economy. The analysis of the desirability of this joint venture depends on the area above the dashed line in d being greater than the area b.
Another factor is the effect of the joint venture on Sony Auto Co. s profits. Before the establishment of the American joint venture, Sony Auto
Co. earned profits on its auto production injapan. After shifting production to the joint venture, Sony Auto Co. profits are earned in the United States and repatriated to Japan. However, in this case they are subject to the U.S. corporate income tax. To the extent that these tax revenues are an addition to the U.S. economy, that would not have resulted from the separate companies; this would lower the dashed line in area d and increase the welfare benefits to the U.S. economy.
NEW UNITED MOTOR MANUFACTURING INC
A widely publicized international joint venture was announced in 1983 by General Motors and Toyota Motor Corporation, the first and third

Wassink and Carbaugh: International Joint Ventures

largest auto companies in the world. With the approval of the Federal Trade
Commission, the two competitors agreed to form a new separate corporation for a twelve-year period, called New United Motor Manufacturing
Inc. (NUMMl). General Motors and Toyota each own half of NUMMI.
Located at an idle GM plant in Freemont, California, NUMMI produces Chevrolet Novas, a compact car based on the design of the Toyota
Corolla. The Nova does not represent a new car developed for the American market. Toyota designed the manufacturing layout, coordinated acquisition and installation of equipment, and implemented its production system.
NUMMI has a production capacity of 200,000 Novas per year. The Novas are sold to General Motors for distribution through its dealers. General
Motors has the sole responsibility for pricing the vehicle. Toyota and other
Japanese manufacturers supply approximately fifty percent of the value of the vehicle. Major components supplied by thejapanese include engine and transmission, while the Freemont plant conducts stamping and assembly.
American companies supply to NUMMI such items as batteries, glass, headliners, paint and sealants, and air conditioner components.
General Motors' announced goal was to learn the Japanese art of management and small-car manufacturing by getting a first-hand look at how Toyota organizes its operations, motivates its workers, and locates machines and materials. General Motors maintained that if it learned how to build lower-cost cars, it would transfer those cost-saving methods to its other plan,ts. It was estimated that General Motors would save as much as
$ 1,000 per car because it did not have to design a new auto from the ground up. Using Japanese-made components was estimated to save an additional
$700 per car.
NUMMI's production system is based on that of the Toyota Motor
Corporation. Two of the more important techniques used by NUMMI are the "just in time" and "Jidoka" quality principles. The "just in time" system

T h e All-American Small Car is Fading," Business Week. March 12, 1984.

THE INTERNATIONAL TRADE JOURNAL

of parts logistics results in NUMMI carrying only three to four days of parts inventory with little or no warehousing in between. This system of tightly controlled parts inventories and close coordination with suppliers is intended to avoid production bottlenecks and yield operating efficiencies.
The "Jidoka" technique results in machines or the production process itself stopping automatically in abnormal situations, such as a machinery breakdown, permitting speedy correction of the situation. Stopping the production line in case of some abnormality also applies to workers who have a stop button at each job sire and the authority to stop the production process at their discretion.
Another potential area of cost-savings comes from NUMMI's efforts to eliminate the job demarcation problems that plague most American auto plants. NUMMI utilizes simpler and more flexible job classifications, work rules, and procedures which are intended to increase labor productivity. In the manufacturing area, there are two groups of employees—production workers and skilled trade workers. The latter group consists of three classifications-tool and die, power house skilled tradesmen, and general maintenance skilled tradesmen. In contrast, some GM plants have more than one hundred job classifications. On NUMMI assembly lines, employees work in teams of five to ten members, each person performing up to fifteen separate jobs. Every production worker learns the job of everyone else on the team in case of absence. Workers regularly rotate jobs for fairness and to relieve boredom. Each team has complete jurisdiction over its own operation, including quality and any improvements in the production process. NUMMI management thus has greater flexibility in assigning jobs, which permits fewer assemblers and quality inspectors to be hired. In return for these concessions from the UAW union, NUMMI pays its workers prevailing wage and benefit rates for new hires in the industry. NUMMI has pledged that its workers would not lose jobs because of automation.
For Toyota, NUMMI represents a relatively low-cost opportunity to test the transferabitity of its production techniques overseas. It provides
Toyota a quick way to learn how to operate in the United States with a partner who knows the ins and outs of the American auto market. Toyota

Wassink and Carbaugh: International Joint Ventures

57

also views a manufacturing foothold in the United States as insurance against rising protectionism in the United States.
The above advantages were confirmed by Toyota's December 1985 announcement to locate its first US. car assembly plant in Georgetown,
Kentucky. Toyota plans to build as many as 200,000 midsize Camry models a year for the U.S. market by 1989. Toyota indicated that UAW representation at the Georgetown plant will probably come as a matter of course. By choosing Georgetown, Toyota is locating itself amid automotive supply companies, many of which are Japanese. This will help enable Toyota to maintain a tightly controlled parts inventory system with suppliers. It was also indicated that the mid-South provided an atmosphere where Toyota knows the Japanese have succeeded in the past (e.g. Honda's plant in Ohio,
Nissan's plant in Tennessee), What's more, workers in the mid-South have not demanded the same restrictive work rules that are found in most Big
Three plants, which are viewed as a major source of Japanese auto makers' cost advantage.
When the Federal Trade Commission accepted a consent agreement with General Motors and Toyota, the Commission majority cited three major procompetitive benefits for the automobile industry from the joint venture. (I) The joint venture would likely increase the total number of small cars available in America, allowing consumers greater choice at lower prices. (2) The joint-venture car would cost less to produce than if General
Motors were forced to rely on other alternatives. (3) The venture permitted a valuable opportunity for General Motors to complete its learning of the more efficient Japanese manufacturing techniques.
Dissenting Commissioners argued that since the joint venture was the offspring of the first and third largest auto manufacturers in the world, competition would be lessened and the new-car prices would increase. Also,
General Motors and Toyota would be able to exchange anticompetitive

''Toyota is Said to Pick Site in Kentucky for its First Car Assembly Plane in U.S." The
Wall Street journal. Decennber 4, 1985.

38

THE INTERNATIONAL TRADE JOURNAL

information, thus blunting competition. Any efficiencies that General
Motors might learn from Toyota would be limited mainly to assembly and stamping processes, since Toyota designed the new car and produced the sophisticated components in Japan. Furthermore, it was argued that
General Motors had already implemented many Japanese techniques without Toyota's help, and that process would continue unabated. It was maintained that General Motors could learn about efficiences from its two other Japanese partners and did not need Toyota.
Critics of NUMMI also maintained that it would result in job losses for
Americans. At the NUMMI plant, up to 5,000 new jobs would be generated.
However, only fifty percent of the Nova is sourced in the United States, the remainder representing Japanese production. Because the Nova was to replace GM"s Chevette, with almost one hundred percent American content, the result would be an overall job loss for American workers.
Moreover, most sophisticated systems and components for the Nova would be produced injapan, providing high-skilled jobs for the Japanese. Back in
America, workers put the final pieces together, which resulted in lowskilled jobs that would become increasingly automated in the years ahead.
In terms of the model presented in Figure 1, it appears that the cost reductions that accrue to NUMMI, compared to the costs achieved by
General Motors when it produces compact autos independently, can be attributed to two sources. First, there will be significant cost reductions from the imported components produced by Toyota. This estimated cost reduction of $700, as reported in Business Week, is not a saving to the U.S. economy. It reflects a reduction based on the lower production costs in
Japan. But this benefit would equally accrue to the U.S. consumer if Japanese autos were instead imported. The same would be true of the cost savings

3,,

"The GM Toyota Linkup Could Change the lndu.stry," Businers Week. December 24,
1984. ^
"General Motors Corp. and Toyota Motor Corp.: Proposed Consent Agreement With
Analysis to Aid Public Comment," Federal Registeryol 48, No. 250, December 28, 1983, pp.
67246-67257.

Wassink and Carbaugh: International Joint Ventures . . .

from design and research done in Japan rather than in the United States.
The important cost savings for the U.S. economy are the productive efficiencies that result from the adoption of Toyota labor practices and assembly line organization of the NUMMI plant. The amount of these savings has not been made public by NUMMI and is difficult to evaluate from the outside. However, critics of the FTC decision to consent to the joint venture estimated NUMMI's efficiency gains to be only $200 per vehicle, resulting in $40 million in total efficiency gains (NUMMI's production capacity equals 200,000 Novas per year). It is likely that General
Motors would contend that these figures underestimate NUMMTs efficiency gains. But it appears that the immediate benefits of NUMMI will be modest. The strongest support for the joint venture might be based on the potential learning-by-doing that could be transferred later to the organization of production in other auto manufacturing facilities of Cieneral
Motors. This possibility goes beyond our model and will continue to be a hypothetical benefit for years to come. It will not only be difficult to measure these gains, but nearly impossible to construct a counterfactural of how organization change would have occurred without this joint venture experience. As a result, it seems inappropriate to attribute large efficiency benefits to NUMMI.
Critics of the joint venture have expressed concern that the cooperation of Toyota and General Motors wilt lead toa joint exercise of market power. In 1984 Toyota and General Motors sold 1,559,980 vehicles rated as subcompacts or compacts out of an industry total of 5,300,000 small cars sold. As the dominant firms in the industry, Toyota and General Motors might use their market power so as to raise auto prices. If other companies followed the price increases of General Motors and Toyota, the losses to the
American consumer could easily exceed the efficiency gains on the small

Future of the Automobile Industry, Hearing Before the Subcommittee on Commerce,
Transportation, and Tourism, U.S. Hou.st- of Representatives, Ffbruyry 8, 1984, p. 286.

60

THE INTERNATIONAL TRADE JOURNAL

number of cars produced by NUMMI.
If the projected ef ficienq' gains of $200 per vehicle for annual output of
200,000 vehicles are achieved by NUMMI, this $40 million gain would be offset by a price increase of only $7.50 per vehicle for the 5,300.000 small cars sold in 1984. If only General Motors and Toyota raise prices on their models to reduce competition for the Nova, and other small car producers do not follow the price increase, a price increase of $25.65 on their models would offset the efficiency gains of NUMMI. If price increases in small cars spill over into price increases for large cars, as would be expected, it would require smaller price increases to offset the projected efficiency gains of
NUMMI.
Pricing behavior of General Motors and Toyota in the 1986 model year is difficult to evaluate. Toyota prices are likely to be influenced by changing exchange rates, as well as changing market conditions in the United States.
In the initial price announcements of General Motors, the increases in prices for models carried over from the 1985 model year showed approximately a five percent increase for small cars comparing 1985 model initial prices and ] 986 model initial prices. For large and luxury model cars, the price increases were substantially lower, while for midsize models the price increases by General Motors were larger. However, if the price increases on small cars were only one percent higher in 1986 than with full competition, this would amount to a $50 to $70 increase which would more than eliminate NUMMI's efficiency benefits to American consumers.
CONCLUSIONS
The overall impact of international joint ventures in the U.S. automobile industry will depend on two factors that are difficult to quantify.
Most analysts indicate that the immediate efficiency gains from jointventure auto production are relatively small. But if these gains can be transferred to all auto production by U.S. firms in a more rapid and efficient manner as a result of this process, the gains could be significant. The potential for losses to the U.S. economy depends on the effect of the joint venture on the competitiveness of the automobile market. If joint-venture

Wassink a n d Carbaugh: international foint Ventures . . .

6 1

production simply replaces former domestic production, as the Nova replaces the Chevette, and the cooperating firms develop implicit arrangements to segment the market and avoid direct competition, the resulting market power could lead to higher profits at the expense of the
American consumer. Any business arrangement that allows firms in an oligopolistic market to cooperate tends to raise economists' suspicions.
Whether the FTC decision to consent to international joint ventures was good or bad will remain difficult to evaluate.

62

THE INTERNATIONAL TRADE JOURNAL

REFERENCES

1.

A. Altschuler, The Euture of the Automobile
Massachusetts: MIT Press, 1984).

(Cambridge,

2. F.R. Artisien and P.J. Buckley, "Joint Ventures in Yugoslavia,"/ottr«*j/ of Intemational Business Studies, 16, No. 2 (1985), pp. 111-136.
3.

RJ. Carbaugh and D. Wassink, "joint Ventures, Voluntary Export
Quotas, and Domestic Content Requirements," Quarterly Joumal of
Business and Economics, 24, No. 2 (1985), pp. 21-36.

4.

W.R. Cline, International Trade in Automobiles (Cambridge,
Massachusetts: MIT Press, 1984).

3. E.J. Contractor, "A Generalized Theorem for Joint Venture and
Licensing Negotiations," Joumal of International Business Studies, 16,
No. 2 (1985), pp. 23-50.
6.

R. Crandall, "Import Quotas and the Costs of Protectionism," The
Brookings Review (Summer, 1984), pp. 8-16.

7. J. Grunwald and K. Flamm, The Global Factory: Eoreign Assembly in
International Trade (Washington, D.C: The Brookings Institution,
1985).
8.

R.D. Hall, The International Joint Venture (New York: Praeger
Publishers, 1984).

9.

K.P. Harrigan, Strategies for Joint Ventures
Massachusetts: Lexington Books, 1985).

(Lexington,

10. KJ. Hladik, International Joint Ventures (Lexington, Massachusetts:
Lexington Books, 1985).
11. J.P. Killing, Strategies for Joint Venture Success (New York: Praeger
Publishers, 1983).

Wassink

and Carbaugh:

I n t e m a t i o n a l Joint

Ventures . . .

6 3

1 2 . C. Oman, New Forms of International Investment in Developing
Countries (Paris: Organization for Economic Cooperation and
Development, 1984).
13. Organization for Economic Cooperation and Development, LongTerm Outlook for the World Automobile Industry (Paris: 1984).
14. R. Reich, "Collusion Course: GM's Pact with Toyota is a Cover for
Surrender," New Republic (February 27, 1984).
15.

, "Japan Inc., USA; If You Can't Beat Them, Joint Venture
With Them," New Republic (November 26, 1984).