Incentives for the emergence of vertical restraints.

Item Type text; Dissertation-Reproduction (electronic)

Authors Durham, Yvonne.

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 05/10/2021 14:22:35

Link to Item http://hdl.handle.net/10150/186842 INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedtbrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely. event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

U-M·I University Microfilms International A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor. M148106·1346 USA 313/761·4700 800/521·0600

Order Number 9506975

Incentives for the emergence of vertical restraints

Durham, Yvonne, Ph.D. The University of Arizona, 1994

V·M·I 300 N. Zeeb Rd. Ann Arbor, MI 48106

INCENTIVES FOR THE EMERGENCE OF VERTICAL RESTRAINTS

by

Yvonne Durham

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF ECONOMICS

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

1 994 2

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Final Examination Committee, we certify that we have read the dissertation prepared by------Yvonne Durham entitled Incentives for the Emergence of Vertical Restraints ------~------

and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of __~J)~Q~C~,iv~_~~~~--~~~b~f~/~V~s~O~f~~,2F~------

_j /L2 17V Date I ,; A .... _ 7~-= 7/""2--"Llqy V\../v~t·· ~ ! Date

Date

Date

Date

Final approval and acceptance of this dissertation is contingent upon the candidate's submission of the final copy of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

Dissertation Director Date 3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules ofthe Library.

Brief quotations from this dissertation are allowable without special permission. provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department of the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances. however. permission must be obtained from the author. SIGNED'¥~ ~ 4 ACKNOWLEDGEMENTS

I would like to thank the members of my dissertation committee: Mark Isaac for his continued support as dissertation director. Stan Reynolds for his insightful comments and encouragement. and especially Vernon Smith. for providing me with emotional and financial support to continue my studies in the face of seemingly overwhelming obstacles.

I am also very grateful for the financial and technical support provided by the

Economic Science Laboratory. Shawn Lamaster and Zhu Li provided invaluable programming help; Pat Kiser always knew which forms needed to be turned in where and to whom; and Steve Rassenti always had an open door. For these things and for their constant friendship. I thank them.

Finally. I would like to thank my family for their constant support and my friend

Don Wells for teaching me that economics can be fun and rewarding. A special thanks goes to all the wonderful friends and educators who were patient with my questions and generous with their time. 5

TABLE OF CONTENTS

LIST OF FIGURES ...... 7

LIST OF TABLES ...... 8

ABSTRACT ...... 9

1. AN EXPERIMENTAL EXAMINATION OF DOUBLE MARGINALIZATION AND VERTICAL RELA TIONSHIPS ...... 11

1.1 Introduction and Literature Review ...... 11 1.2 Experimental Design ...... 15 1.3 Experimental Procedures ...... 20 1.4 Experimental Results ...... 22 1.5 Conclusions ...... 31 1.6 Endnotes ...... 33

2. A THEORETICAL MODEL OF SYSTEMS EXCLUSIVE DEALING ...... 35

2.1 Introduction and Literature Review ...... 35 2.2 Model Specification ...... 39 2.3 Game Description ...... 42 2.4 Equilbrium Without Exclusive Dealing ...... 43 2.5 Equilbrium With Exclusiv~ Dealing ...... 45 2.6 Results ...... 51

3. DID NINTENDO USE EXCLUSIVE DEALING CONTRACTS TO MAINTAIN ITS MONOPOLY POSITION IN THE U.S. HOME VIDEO GAME INDUSTRy? ...... 62

3.1 Introduction ...... 62 3.2 History of Nintendo ...... 64 3.3 The U.S. Home Video Game Industry Before Nintendo ...... 66 3.4 The Market Test ...... 68 3.5 The Security Chip ...... 73 3.6 Licensing Restrictions ...... 74 3.7 Marketing Practices ...... 77 3.7.1 Nintendo Power ...... 77 3.7.2 Nintendo Hotline ...... 78 3.7.3 Keeping Parents Happy ...... 79 3.7.4 Game Management ...... 80 6

TABLE OF CONTENTS - Continued

3.8 Competitors and Advances in Technology ...... 82 3.8.1 The Super Nintendo Entertainment System ...... 85 3.8.2 The Continuing Market for NES ...... 86 3.8.3 Licensee Benefits ...... 87 3.9 Lawsuits ...... 89 3.9.1 Atari Garnes Corporation ...... 89 3.9.2 Atari Corporation ...... 91 3.9.4 Federal Trade Commission ...... 93 3.10 Legal Treatment of Exclusive Dealing ...... 96 3.11 Theoretical Evaluation ...... 106 3.12 Conclusion ...... 112 3.13 Endnotes ...... 114

APPENDIX A: EXPERIMENTAL INSTRUCTIONS ...... 120

APPENDIX B: EXPERIMENTAL RESULTS ...... 127

APPENDIX C: REGRESSION RESULTS ...... 138

REFERENCES ...... 141 7

LIST OF FIGURES

FIGURE 2.1, Exclusive Dealing Ranges ...... 52 FIGURE 2.2, Percentage of Customers System 2 Must Deliver ...... 57 FIGURE 2.3, Percentage of Customers System 1 Must Deliver ...... 58 FIGURE 2.4, Exclusive Dealing and Retail Prices ...... 60 FIGURE 3.1, Home Video Game Industry Sales ...... 72 FIGURE 3.2. Exclusive Dealing Ranges ...... l07 8 LIST OF TABLES

TABLE 1.1, Price, Quantity, and Profit Predictions ...... 19 9

ABSTRACT

The incentive that an upstream firm has to integrate or to impose vertical

restraints arises because the actions taken in the downstream market affect upstream

profits. In this dissertation, the existence of an incentive to impose vertical restraints

is studied experimentally and the use of exclusive dealing as a vehicle for

monopolization is considered both theoretically and empirically.

The first chapter investigates the presence of a vertical externality, and therefore

an incentive to impose vertical restraints, in eleven experimental markets. The

upstream markets are characterized by a single seller. When the downstream market

consists of three firms, no evidence of a vertical externality is found, and prices and

profits are consistent with the vertically integrated outcome. When the downstream

market is characterized by a singl~ firm, there i~~ evidence of a vertical externality. An

~nalysis of individual market behavior reveals th~t downstream firms approximate their

best response functions, with some indication that they improve over time.

The second chapter is a theoretical examination of the use of exclusive dealing to obtain monopoly power when the products are systems of goods with a vertically differentiated component. Two upstream firms sell the differentiated component directly to the consumers and the other components through a set of downstream firms. Conditions under which each firm will be able to monopolize the market using exclusive dealing are determined. The effect on prices is also examined. 10

The third chapter is a case study of the U.S. home video game industry.

Nintendo of America, Inc. used exclusive dealing contracts with the software companies that designed games for its system. These companies were prevented from selling games made for Nintendo in any other format. The claim that Nintendo achieved and maintained its virtual monopoly through the use of these contracts even though its console was of a lower quality than those of its competitors, is investigated using industry data an the model from Chapter 2. According to the model, conditions in the industry were such that Nintendo would have been able to maintain monopoly control. causing the higher quality firm to exit. 11

1. AN EXPERIMENTAL EXAMINATION OF DOUBLE MARGINALIZATION AND VERTICAL RElATIONSHIPS

1.1 Introduction and Literature Review

Vertical relationships among firms have been of continuing interest to industrial

organization economists and antitrust authorities. Although the relationship between

upstream firms and downstream firms is in some ways similar to the relationship

between a firm and its customers, a vertical relationship between firms can be more complex. In a vertical structure, some decisions, such as the final price, the amount of services, or the promotional effort are made after the intermediate good is sold by the upstream firm and hence no longer in its control. Since these additional decisions affect the upstream firm's profit, an upstream firm may have an incentive to attempt to control them. The maximum profit that can be earned by the vertical structure is the integrated profit -- or the profit that an integrated firm, controlling all the decisions made by the structure, could earn. Much of the literature in this area has focused on the issue of vertical control and the set of vertical restraints sufficient for the vertical structure to obtain the integrated profit.

Incentives to integrate or to impose vertical restraints can come from a variety of sources. The monopoly incentive, or the incentive that arises when the downstream market is not competitive, comes from this inability to control downstream decisions. 1

A monopoly producer of an intermediate good wi1l charge its downstream customers 12 a price that is greater than its marginal cost. If the downstream market is not competitive, specifically if it is also monopolized, then the price of the final good will also be marked up above cost. Each time the downstream firm takes any action to increase the quantity it sells ( i.e., lowers price, increases promotional effort), the upstream firm's profit is increased. However, the downstream firm does not take this into account when it makes its decision and therefore tends to restrict quantity relative to the vertically integrated solution. This is referred to as the vertical externality and provides an incentive for the use of vertical restraints.

A well-known illustration of this basic externality is due to Spengler (1950). He uses a fairly simple model to examine the efficiency of vertical integration when the downstream market is monopolized. In his model, the downstream firm's only decision is the retail price. When both the upstream and downstream markets are monopolized, the presence of the vertical externality, which has corne to be known as

"double marginalization," causes the final price to be higher and the aggregate profits to be lower than in the integrated outcome. The upstream firm charges a price higher than its marginal cost, causing the cost facing the downstream firm to be higher than the vertical structure's cost. The downstream firm then behaves as a monopolist with this marginal cost and prices above it. Therefore, two successive markups occur which cause a restriction of quantity. The distortion caused by the downstream monopolist lowers upstream profits, which provides the upstream firm with an incentive to control downstream actions. The vertical structure earns lower aggregate profits and consumers surplus is decreased because of the quantity decrease and price increase. 13 In Spengler's model, if the downstream market is competitive, the vertical

externality is eliminated. In this case, the final price is forced down to the downstream

marginal cost, and the structure earns the vertically integrated profit. The competition

in the downstream market effectively controls price for the upstream monopolist.

Welfare is enhanced and aggregate profits are increased. The incentive to vertically

integrate or impose restraints is removed in this simple case. Spengler's theory is an

argument for the efficiency of vertical integration and the use of competition to

duplicate its results when the downstream market is not initially competitive.

Vertical relationships among firms are vastly underexplored in the experimental economics literature. Because this theory is basic to much of the work done in the vertical restraints and control area, it seems to be a good choice for the initial experimental test in this area. Because of the control possible in the laboratory, the experimental setting would give the theory its best shot at succeeding. The purpose of this paper is to report an initial test of Spengler's model (as described in Tirole

[1988]). Two types of experiments are discussed here. In both types, the upstream market consists of a single seller. The types differ by the downstream market structure. In one, the downstream market is characterized by a monopolist, while competition is introduced in the other.

Previous experimental work has studied problems of a somewhat similar nature.

Plott and Uhl (1981) investigated the role of middlemen in double auction markets.

Some subjects, known as traders, first participated as the buyers in one market with other subjects who were sellers. The traders then became the sellers of those units in 14 another market with a set of subjects who were designated as buyers. It was found that the competitive solution was supported with the middlemen present and that contrary to popular belief, the middlemen did not profit at the expense of the buyers and sellers. In all cases, three or four traders were used. This interdependence of markets was also studied by Goodfellow and Plott (1990) in a similar setting. There was an input market which involved six sellers and an output market consisting of six buyers. Four subjects were assigned to be producers. The producers were able to purchase units in the input market, transform them according to a particular transformation function into units of output, and then sell those units in the output market. The double auction institution was used in each market, and support was found for the competitive theory.

Fouraker and Siegel (1963) studied bilateral bargaining in single-shot and repeated settings. In these experiments, a buyer and seller were paired. The seller chose price and the buyer chose quantity. In examining a vertical structure in which both the upstream and downstream markets are monopolized, the choice space is essentially the same as these bilateral bargaining experiments. The upstream firm chooses price, and the downstream firm, probably taking final demand into consideration, indicates the quantity it wishes to purchase for that price. In Fouraker and Siegel, the pairs were randomly chosen, and each subject's partner was anonymous to the subject. They examined the tendency for these pairs to be closer to the Bowley solution (analogous to the double marginalization solution here) or the Pareto optimal solution (analogous to the integrated solution). The Bowley solution was supported in 15 their treatment with incomplete information and repeated play and with complete information when the payoffs were split equally at the Bowley solution. The results are somewhat mixed when there is complete information, repeated play, and the payoffs are split equally at the Pareto optimal solution. Several of the pairs found the Pareto solution under these conditions.

1.2 Experimental Design

The basic assumptions of Spengler's theory are as follows. The upstream firm is assumed to choose a contract which the downstream(s) firm either takes or leaves.

This contract consists only of an upstream price. The downstream firm(s), whose opportunity cost is normalized to 0, then chooses a price. After this, the consumers respond by making their purchases. All firms are assumed to know demand. Each level of production is characterized by constant marginJI costs, Cu and Co' where the subscripts denote upstream and downstream. The upstream firm knows both Cu and

Co' while the downstream firm(s) knows only its own cost, Co. Assuming a linear demand, D(P) = a - bP, where a, b > 0, a downstream monopolist will choose Po to maximize 16 giving a reaction function of

Po = 1/2b ( a + bPu + bCo ).

The upstream monopolist will choose Pu to maximize

nu = ( Pu - Cu ) ( a - bPo )

subject to the downstream firm's reaction function. The subgame perfect Nash equilibrium prices, quantity, and aggregate profit when both markets are monopolized are:

PL; = 1/2b ( a + bCu - bCD ) ,

PD = 1/4b ( 3a + bCD + bCu ) ,

Qu = Qo = 1/4 ( a -bCD -bCu ), and

IIlI + nD = 3116b ( a - bCu - bCD f·

An integrated firm would choose P to maximize

II = ( P - Cu - CD ) ( a -bP ).

The integrated price, quantity, and profit are: 17

P = l/2b ( a + bCo + bCu ) ,

Q = 112 ( a - bCo - bCu ) , and

II = 1/4b ( a - bCo - bCu )2.

If the downstream market is characterized by competition, then the downstream reaction function will be Po = Pu + CD' The equilibrium then consists of the integrated final price. quantity. and total profit and an upstream price of

Pu = 1I2b ( a + bCu - bCo ).

As long as a > b( Cli + CD ), the final price in the integrated structure (or competitive downstream structure) is less than the final price in the monopoly/monopoly structure, and profits are higher. Using this linear structure for demand, the upstream monopolist will charge the same upstream price regardless of the structure downstream, so the restriction of quantity associated with a downstream monopolist will lower his profits.

In order to follow the assumptions of the theory as closely as possible, the institution used in these experimental markets is the posted-offer institution. 2 Smith

(1981) found that the posted-offer institution is the pricing mechanism that is most supportive of the monopoly price. It is also appropriate in this case because the downstream firm is assumed to be presented with a take-it-or-leave-it price. 18 Demand for each treatment was a discrete form of the linear demand

described above with a = 121 and b = .2. All prices and costs during the experiments were denoted in experimental pesos. Downstream marginal cost was constant at 100

pesos and upstream marginal cost was constant at 310 pesos. The derived demand

that the upstream firm faces when the downstream market is monopolized is the

marginal revenue cUl\'e associated with the final demand minus the downstream

marginal cost. A linear demand curve was used because it provided the cleanest marginal re\'enue curve to work with. In all of the experiments, both the upstream and downstream markets had a capacity of 54 units. In the competitive case, three idcntical firms shared this capacity. The theoretical predictions for prices, quantity, and profits for these designs can be found in TABLE 1.1.

Becausc the model assumes constant marginal costs, competitive firms earn zero economic profits in equilibrium. In order to assure that the subjects in the experiments did not earn zero accounting profits, a commission of four pesos was paid to each subject whenever he/she sold a ur.it in both types of experiments. Price offers were restricted to being made in five peso increments to avoid a distortion in predictions stemming from this commission. The only effect the commission had was to sharpen the predictions, since without them, the downstream is indifferent between selling and not selling the last unit. Because the theory predicts that the upstream firm will squeeze all of the profit out of the downstream on the last unit, the commission provides an incentive to trade the last unit. Since prices could only be made in five 19 peso increments, the upstream firm had no way to extract the four-peso commission from the downstream firm on the last unit.

TABLE 1.1

Price, Quantity, and Profit Predictions

Q

\1onopoly'\1onopoly 410 555 10 520 490

Monopoly'Competition 405 505 20 580 80

!\nte: Pores and profits are in pesos, and profits include commissions but are net of the lump sum payments for the upstream subJC·cts. Although demand IS linear. the discreteness causes upstream prices to vary between the two types. The actual predicted upstlram poce with a continuous demand would be 407.5 pesos.

An exchange rate of 2.5 pesos for each cent was used for the monopolists in each experiment. However, an exchange rate offour cents for every peso was used for the competitive firms since the competitive firms were earning only commissions in equilibrium. The difference between downstream profits at the Nash prediction and the profits at one unit below the prediction is the four-peso commission. It was decided to design the experiments so that a subject's indifference between choosing 20 the predicted quantity and one unit less would not involve a significant portion of the

market. This, along with the linear demand curve, meant that the upstream

monopolists in these markets would have earned a great deal of money at the

predicted prices. Using a large exchange rate would have flattened out the payoff

function more than was felt suitable, so a lump sum payment was taken from the

upstream firm each period. Theoretically, this should not affect a subject's maximizing

choices. The lump sum payment was made small enough to allow for a wide variety

of choices and mistakes. so final payments to the monopolists still tended to be quite

large.)

1.3 Experimental Procedures

Thirty-four undergraduates at the University of Arizona participated in these

experiments. Four monopoly/competition and seven monopoly/monopoly experiments were run. Experiments generally lasted about one and a half hours for both types, and

subject payments averaged about S16.50 for the monopoly/competitive experiments and

S28 for the monopoly/monopoly experiments. Each experiment consisted of two practice periods and 15 Cictual periods, and the subjects were informed of this at the beginning of each experiment. The upstream firm was known as the producer, and the

downstream firms were known as traders. 21 In the monopoly/competition experiments, subjects were brought into the room

and told that one of them would have the opportunity to be a single seller and that we would allow them to earn the right to be that seller. Subjects were given a short trivia

quiz. and the high scorer was awarded the role of producer.4 Each subject was given

a set of instructions and record sheets. a buyer demand schedule. and several offer forms. A copy of the instructions for this experiment can be found in Appendix A. In this environment. the traders produced to order to avoid any problems associated with carrying inventories or exposing the traders to risk. Each period. the producer was asked to set a price. which the traders were shown immediately. The traders were then invited to submit a contingent price to the buyers. The buyers. who were programmed following the buyer demand schedule. ordered from the traders. given the set of offered prices. The buyers were programmed to approach the low-priced trader first, and in the case of a tie. to randomly choose a trader to deal with. These orders were submitted to the traders. after which the traders made purchase requests from the producer. If the producer did not offer to supply enough units to satisfy the traders. the traders' purchase requests were processed in a random order of which the traders were informed at the beginning of each period. In each experiment. one su bject was paid a fixed payment of S15 to run papers back and forth between the subjects and the experimenter to speed processing.

In the monopoly/monopoly experiments. subjects were brought in one pair at a time and randomly assigned the roles of producer and trader. Since subjects were face to face in the monopoly/competition experiments. it was decided to allow these 22 experiments to be conducted face to face also. The producer had enough information to calculate the maximum profit that the trader could earn given each price the producer offered. However. the trader had enough information to be able to compute only his own profit. TI-::s provided an asymmetry of information. Each period. the producer was asked to post a price that was shown to the trader. The trader then offered a price to the buyers. and knowing the quantity demanded at that price. made its purchasing decision.

1.4 Results

Results from all experiments are shown in Appendix B. The competitive downstream price for the monopoly/competition experiments and the downstream best response for the monopoly/monopoly experiments are determined each period from the actual upstream price. There are several issues which should be considered when examining this data. First. do the subjects coordinate their actions in such a way as to achieve the predicted final price and quantity outcomes in the market? Second. do the upstream and downstream subjects behave as the theory predicts? In other words. are final predicted outcomes achieved. and are they achieved through both firms behaving as theory predicts or in some other way? Third. does the change in the downstream market structure affect the efficiency of the markets? 23

It seems fairly evident from an examination of the monopolylcompetition graphs

that the theory is working well. The downstream firms appear to behave competitively

as the theory predicts, and the upstream firm generally takes advantage of this. With the exception of mel 04, the upstream monopolists lock onto the predicted price of 405

pesos by the tenth period. By the eighth period, almost all of the downstream firms are charging a price of Pu + 100 (or the downstream marginal cost). During the last

8 periods. occasional attempts by certain traders are made to raise price, but these traders immediately find it unprofitable and return to the competitive price the next period. In the fourth monopoly/competition experiment, although the monopolist had some difficulty in finding the profit-maximizing price, the downstream firms were behaving competitively. No price distortion is present in the downstream market.

5 Using t-tests , the hypothesis that the true mean of the downstream prices across experiments is 505 pesos cannot be rejected in any of the last 5 periods.6 The hypothesis that the true mean ofthe upstream prices across experiments is 405 cannot be rejected in any of the last 5 periods either. Spengler's monopolylcompetitive theory is supported, and the vertical structure, as well as each of its components, behave in general as the theory would predict.

These experiments tended to be uninteresting for the subjects, which is why only four were run. Once price settled down to the competitive price, the traders' profits were in effect determined randomly by the buyer program. Downstream subjects informed the experimenter that they felt as if they had little control over the outcome. Three firms provided enough competition in the downstream market to 24 compel the firms to price a marginal cost. Because the downstream firms found that they could do no better than act as price takers, the upstream firm had no incentive to place any restraints on them. He could have done no better by imposing any restraints. As in the Plott and Uhl middlemen experiments, the middlemen (or traders in this case) took no profit at the expense of the buyers or upstream sellers. Spengler's assertion that competition in the downstream produces the vertically integrated results and eliminates the incentive for vertical restraints because ofthe removal ofthe vertical externality is supported.

In examining the monopoly/monopoly experiments, one can see that there is much more variation in the data than there was in the monopoly/competition data.

Subjects seemed to spend time exploring the profit possibilities. It appeared to be especially difficult for the upstream firm to determine its best strategy as evidenced by the comments of many of the upstream subjects, who expressed concern that they could not figure out how the downstream firm was behaving and therefore had difficulty deciding which actions to take.

The conjecture that these experiments result in the Nash prediction consists of testing the two hypotheses that PD = 555 pesos and Pu = 410 pesos. Using the downstream prices across each of the seven experiments, we cannot reject the hypothesis that the true mean of the final price is 555 pesos in each of the last five periods.7 However, we can reject in periods 13 and 14 that the true mean of the upstream prices is 410. Therefore, overalI we would conclude that there is evidence in these experiments favoring the Nash prediction. However, the rejection of the 25 hypothesis that the true mean of upstream prices is 410 in two of the last five periods may indicate that this is a case where the structure as a whole behaves as theory would predict, but may not be the result of the individual firms behaving as predicted.

In four of the last five periods, we can reject the hypothesis that the mean of the downstream prices is equal to the integrated or monopoly/competition price of 505 pesos in favor of the hypothesis that it is greater than 505. This coincides with the results in Fouraker and Siegel, who found support for the Bowley (Nash) rather than the

Paretian (integrated) equilibrium in their bilateral bargaining experiments under incomplete information conditions. Subjects here have more information than what was given the Fouraker and Siegel subjects, though it is presented in a different way.

Rather than a table of profits, demand is given and subjects can then either compute their own profits or explore the profit options using the market. The upstream firm has enough information to calculate the downstream profits if he wishes.

An interesting issue to consider in these markets is whether the downstream firms are indeed creating a price distortinn. This can be tested by examining whether the true mean of the actual downstream prices in each period is different from the mean of what the competitive responses would have been to the upstream prices in each of the last five periods. The only period in which we cannot reject the hypothesis that these two means are the same is in period thirteen. This demonstrates that the vertical externality is present in these markets.

Spengler's assertion is that the vertically integrated solution can be duplicated by introducing competition in the downstream market and that both aggregate profit 26 and consumer surplus will increase because of this. Using the usual measure of consumer plus producer surplus to measure the efficiency of the marketS, in all but one of the last five periods, we can reject the hypothesis that the monopoly/monopoly markets are on average as efficient as the monopoly/competition markets in favor of the hypothesis that the monopoly/competitive markets are more efficient.

To summarize the monopoly/monopoly results presented so far, there appears to be strong evidence supporting the Spengler conclusion of a vertical externality and the argument that the elimination of that externality increases welfare. There are also several questions involving individual behavior and coordination in these experiments that deserve more detailed attention. For example, how close is the downstream firm to its best response function, and does it get better at finding this best response over time? Does the downstream firm's proximity to its best response function affect how close the upstream firm is to the Nash prediction? How might the upstream firm predict the actions of the downstream firm in order to make its pricing decision?

Some models are presented below that are designed to provide some insight into the answers to these questions.

Once the upstream firm has set a price, the downstream firm has a unique best response. The downstream firm's best response function is a linear function of the

9 upstream price and is specified as follcws :

PDBR = 350 + .5Pu. 27 In order to determine if downstream prices follow this best response path, a fixed effects rnodep o was used to estimate

where Dklt equals 1 for experiment k and 0 otherwise. Results from this regression can

be found in TABLE C.1 of Appendix C. The joint hypothesis that Q1 = Q 2 = ... = Q 7

ll = 350 and f3 = .5 is rejected • which on the surface seems to imply a rejection of the theory. However. a closer analysis reveals that the rejection is caused entirely by experiment rnm 107. because when tests are conducted on only the first six experiments, we can no longer reject the hypothesis that the intercept is 350 and the slope is .5. 12 More specifically. preliminary analysis indicates that strong serial correlation in the seventh experiment may be what is causing the rejection of the hypothesis. Hence. there does appear to be evidence that the downstream firms are approximating their best response functions.

Another issue that is worth considering is the question of whether the downstream firms are getting better at playing their best strategies over time. This hypothesis was tested using the following equation: 28

where D4kit is a dummy variable that equals 1 if it is one of the last four periods is

experiment k and 0 otherwise. was estimated. 13 Results from this regression can be

found in TABLE C.2 of Appendix C. The hypothesis that learning is occurring would

indicate that the {3's would be negative. {3,. {33' and {3s were all significantly different

from zero and negative. while {32' {34' and {36 were positive and not significantly

different from zero. {37 was significantly different from 0 and positive. again providing

indication that experiment mm 107 behaved contrary to the other experiments. It is

clear from the graph of mm 107 that the downstream price moves away from the best

response in the last four periods in that experiment. {36 was insignificant because the

downstream firm in mm 106 was very close to his best response all along and really had

no learning to do. In summary then. we see some limited evidence. except for mm107. that learning is going on in this environment in the downstream market.

Docs the upstream price get closer to the Nash prediction of 410 pesos when the downstream firms get closer to their best response? In order to answer this question. the fixed effects model below was estimated:

Coefficient estimates and standard errors from this regression can be found in TABLE

C.3 of Appendix C. Upstream prices getting closer to the Nash prediction as downstream firms get better at finding their best responses would imply a {3 > O. The hypothesis that f3 = 0 cannot be rejected in this model. This model provides no 29 evidence that the closeness of downstream price to its best response is significant in determining how close the upstream price is to the Nash prediction.

The downstream firm has a clear best response in these markets. The upstream firm. however, must predict what the downstream firm's response will be in order to choose a "best response." Three models of this predictive process are hypothesized.

A myopic prediction occurs when the upstream firm assumes that the downstream firm will post the same price this period as it did last period. The best response to this prediction is for the upstream firm to post

(1) PUt = a + {3P Ot.1 where a = - 100 and {3 = 1.

A constant markup prediction occurs when the upstream firm predicts that the downstream firm will mark the price up this period by the same amount as it did the last period. The best response to this prediction is to post

(2) PUt = a + (3( POt.1 - PUt.1 ) where a = 455 and {3 = -.5.

An OLS prediction involves the upstream firm running the regression POL = Y + oPUl +

Et each period with all the observations available up to that point in time. Once he has made an estimate of y and 0, his best response is to charge

(3) Pl 't = a + {3 (1/0) + 77 (y/O) where a = 155, {3 = 302.5, and 77= -.5. 30

Two separate fixed effects models were run to estimate (1) and (2).14 Results

from both models can be found in TABLES C.4 and C.5 in Appendix C. In both cases,

the hypotheses that the coefficients are equal to their theoretical values are rejected.

These are joint tests of the prediction model and whether the upstream firms respond

optimally to their predictions. If the upstream firms are making either myopic or

constant markup predictions of downstream behavior, there is no evidence that they

are using the best responses to these predictions to make their pricing decisions.

In order to discover if there is any evidence that the upstream firm is using an

OLS-type prediction to choose his price, a test of the equality of means between the

actual upstream prices and the best response prices given OLS predictions was

conducted. We cannot reject the hypothesis that the true means of the two

populations are equal in any of the last eight periods, suggesting some evidence that the upstream firms are pricing on average as if they are making optimal responses to an OLS prediction of downstream behavior.

An alternative way to test these prediction models is to examine which method

provides the best forecasts of upstream behavior. Using the last 8 periods of each experiment and examining mean squared errors, we find that (3) has the lowest MSE in all experiments but mm 103, where (2) does better. This is to be expected because the OLS prediction uses two parameters to estimate while the other two prediction methods use only one. Since we can never know what method, if any, upstream firms use to predict downstream behavior, the best that can be done is to find a method 31 which is consistent with the data. (1) and (2) do not seem to be consistent with the data, while there is some evidence that (3) is.

1.5 Conclusions

The experiments reported here provide support for Spengler's double marginalization theory. The vertical externality is present in laboratory markets when vertical markets are controlled by monopolists, providing evidence that a chain of monopolies is less efficient than a single monopolist. Prices in the monopoly/competition markets approached the integrated solution, and prices in the monopoly/monopoly markets supported the double marginalization prices. The profit incentive for an upstream firm to control downstream decisions was present. Many of the upstream subjects said that they felt as if they didn't have much control, and that it was the downstream firm that decidec how much profit they were going to make.

This is further evidence that the vertical externality provides the upstream firm with an incentive to control the actions ofthe downstream firm. The vertical externality is removed when the downstream market is characterized by competition and welfare is increased. The incentive for the use of vertical restraints or integration is not present in this simple environment once competition is introduced.

The experimental examination of vertical issues has many directions for future research. The experiments reported in this paper indicate that in the basic vertical 32 structure, there may be an incentive to use vertical restraints or to integrate. These experiments can be modified in several ways to examine other interesting issues. First of all, how important are the information conditions? Is it necessary for the firms to know demand or for the upstream firm to know the downstream cost? What happens when downstream firms no longer produce to order? What if a variable proportions production function were introduced in the downstream market? There are several other environments in which it might be useful to investigate the incentives for vertical control. Another basic vertical externality comes from a promotional effort issue.

Theory predicts that when the downstream firm controls the promotional effort, it tends to supply too little of it. It would be interesting to examine what types of vcrtical restraints, if any, are used to maintain vertical control when the incentive is therc to impose them. 33 1.6 Endnotes

1. We will ignore other incentives for integration such as the elimination of transaction costs or the realization of technological efficiencies here.

2. These experiments were hand-run with the aid of computerized buyers for the monopoly/competition experiments.

3. Final payments to upstream monopolists ranged from $1.00 to $39.00.

4. This practice has become common in experiments in which some portion of the subjects are given an economic advantage. Hoffman and Spitzer (1985) found that subjects were more likely to fully exploit their economic rights when those rights were earned rather than assigned randomly.

5. Of course t-statistics require the data to be normally distributed, but we have no way of verifying this. However. nonparametric tests. such as the sign test and the Mann-Whitney test, lead to identical conclusions. Therefore. we will continue to report only the results from the t-tests. All statistical tests are done using a 5% significance level.

6. It should be noted that because of the possible time dependence across periods, these t-tests may not be independent.

7. Possible dependence of tests is recognized again.

8. Producer surplus was computed without commissions.

9. Because of the discreteness of the demand function, when the upstream price ends in a five. the actual best response is 2.5 pesos higher than this. The data was adjusted to reflect this.

10. A random effects model was rejected which is not surprising with only seven individuals.

11. A test for different slope coefficients across experiments in this model was rejected at the 596 significance level.

12. This test can be conducted on pooled data instead of in a fixed effects model because there is no evidence that the first six experiments have any individual-specific effects. 34 13. Pooling the data was inappropriate here because the presence of individual-specific effects could not be rejected.

14. This method was not applied to model (3) because model (3) would require estimating y and 0 in stage 1 and using these estimates as regressors in stage 2. The resulting generated regressors problem would invalidate standard inference procedures. 35

2. A THEORETICAL MODEL OF SYSTEMS EXCLUSIVE DEALING

2.1 Introduction and Literature Review

An exclusive dealing contract exists when one firm agrees to purchase only from another. An example of this might be a retailer agreeing to purchase the product of only a particular manufacturer. It is one of several nonprice vertical restraints that have become the subject of much analysis, both academically and legally. Because of the legal implications, much ofthe literature has concentrated on whether exclusive dealing can enhance or protect a firm's market power. During the past year, the Justice

Department has been reworking the relatively lenient view taken by the 1985 Vertical

Restraints Guidelines in an effort to step up enforcement of vertical restraints.

There is a wide range of opinions on the impact of exclusive dealing contracts.

Some argue that ED contracts are always anti competitive and should be illegal. while others insist that exclusive dealing contracts enhance (or at least don't harm) competition. The view that exclusive dealing has anticompetitive effects generally stems from foreclosure arguments. ED contracts tend to foreclose a portion of the market to competitors and reduce their choices during the duration of the agreement.

If one supplier captured all of the leading distributors, other producers would be severely handicapped in their efforts to reach consumers -- even if their products were better or cheaper. This has an adverse effect on the supplier's competitors. It is also 36 argued that ED provides barriers to entry because an entrant may either have to deal with more expensive dealers, enter into both stages of production, or may simply have fewer choices for dealers.

On the other hand, there are those that argue that an exclusive dealing contract cannot create the power to exclude -- it is only a means of implementing that power if it already exists. Bork (1978) argues that, like actual vertical integration, exclusive dealing creates efficiencies. Because a manufacturer would be unable to impose an ED contract on its dealers without having the necessary market power to begin with, the contract doesn't offer the firm any advantage that it wouldn't have had without the contract. Therefore. the advantage of the contract must be the creation of efficiency.

Marvel (1982) also provides an efficiency explanation for exclusive dealing. He discusses the use of exclusive dealing contracts to create a property right to advertising done by a manufacturer. An exclusive dealing contract provides the manufacturer with a property right to its promotional investment by preventing the dealer from attempting to sell a cheaper brand to customers brought in by the manufacturer's advertising. Besanko and Perry (1993) consider this incentive in the context of a differentiated products oligopoly and find that efficiency increases as more manufacturers adopt ED.

A series of papers discussing the competitive effects of exclusive dealing agreements was published in the late 1980s. The initial paper. Coman or and Frech

(1985). examined the market conditions under which exclusive dealing impedes entry into a market. The criterion used for evaluating the effects of exclusive dealing 37 contracts is their impact on the entry costs of rival manufacturers. This criterion is used because Director and Levi (1956) suggest that a contract would be anticompetitive if it raised rivals' entry costs. In a model with one dominant manufacturer and a potential entrant, they identify a set of circumstances under which an exclusive dealing arrangement could have anti competitive implications. They find that ED is more likely to occur when product differentiation is relatively strong. They also find that a manufacturer with a product differentiation advantage could impose exclusive dealing on his dealers while also raising the wholesale price.

In a reply to this paper, Swartz (1987) suggests that these results are changed when the constraints imposed by dealer rationality are more carefully incorporated into the model. By doing so, he finds that ED is more likely when product differentiation is weak and that in order to impose exclusive dealing, the manufacturer must lower the wholesale price. He indicates that CF's results are sensitive to the way they represent product differentiation -- as the size of a constant premium that the manufacturer can fully extract through its wholesale price.

Mathewson and Winter (1987) obtain different results by considering two manufacturers selling products which are imperfect substitutes. The manufacturers sell to a large number of retailers, each of which has a local monopoly in selling to final consumers. Exclusive dealing by the dominant manufacturer eliminates its rival from the market. To observe ED in their model, the firms must be asymmetric. If the products are still close enough substitutes, the competition to attract retailers causes a decrease in the wholesale price. They find that the retail price of the good will rise 38 with exclusive dealing when the demand for the products is very asymmetric. Welfare

is affected by ED in two ways. First, variety is reduced, and second, the final price may either increase or decrease. Welfare can obviously decrease, but by using a linear demand example, they find that welfare may increase in some instances.

This paper examines the use of exclusive dealing contracts in an industry where the product sold is actually made up of a system of components. For example, the product might be a set of CD's and the associated CD player, a VCR and video tapes, or a video game console and video game cassettes. It is motivated by the state of the home video game industry in the late 1980s and early 19905. The dominant firm,

Nintendo of America. had been using ED contracts with its affiliated software companies for some time. These contracts forbade software companies who designed games for the Nintendo system to put the same game in one of its competitor's formats. The stated concern over these contracts comes from the foreclosure literature

-- that other manufacturers are left with either no software companies or those of lower quality. During this time, Nintendo was using a console that was of a lower quality than that of its competitors. (Nintendo's console used an 8-bit central processor while Sega's used a 16-bit central processor.) However, it was contended by its competitors that even with a lower quality console, it had captured a large portion of the best games with exclusive dealing contracts, and thereby held on to something like

8096 of the market. The idea is that it would not have been able to do this without the use of exclusive dealing contracts. 39 The use of systems exclusive dealing has not been examined before. An interesting question might be addressed here. Could Nintendo have held on to its huge share of the home video game market by using these ED contracts even though it was competing with a lower quality console?

2.2 Model Specification

The commodities in this model are actually systems of goods. System i is made up of one y-component and n, z-components, where ni is exogenously determined.

There are two upstream firms which sell the y-components directly to the consumers and the z-components to the downstream firms. The downstream firms then sell the z-componems to the consumers. The downstream firms are each assumed to have a local monopoly. and therefore only one representative downstream firm is examined.

These firms are assumed to be price-takers with no monopsony power. The upstream firms sell y-components of differing qualities, with the qualities indexed as S2 and SI'

The upstream marginal cost of a system is mCi = SiS + nicz' so that the marginal cost of a system increases both with the quality of the y-component and the number of z­ components available with it. The two systems are incompatible, so that consumers place zero value on a system made up of Yi and z/s where i:;e j. Prices are denoted as follows: 40 Pwi = the wholesale price of firm i's z-components,

Pzi = the retail price of firm i's z-components, and

Pyi = the retail price of firm i's y-component.

The total retail price of system i will be denoted as Pi = Pyi + niPzi' Each consumer purchases either 1 or 0 systems and is assumed to purchase all available z-components if the system is purchased. Preferences can be described by:

u = e s,n i - Pi if system i is purchased, and

= 0 if neither system is purchased.

The effective quality of a system is determined by the quality of the y­ component and the number of z-components available on that system. e is a taste for quality parameter and is uniformly distributed across the population of consumers between f2 2= 0 and e= ft + 1.

The determinants of system quality, Si and ni, are exogenously determined before the game begins. Assume that S2n2 > Slnl' This means only that SiSl > n/n2, indicating that it is not necessary for the high quality system to have both the highest quality y-component and the most z-components. Since S2n2 > Sln1' if PI > P2' no one will ever purchase system 1. Therefore. we will restrict our attention to the case where

PI ~ P2' It is also true that no one will purchase system 1 if S2n!P2 2= SIn/Pl' To see this: 41

> 0

Therefore. we will restrict out attention to the case where s2niP2 < sIn/PI and PI ~

P2' If both systems are available. then those consumers with e > (P2 - Pl)/(S2n2 - Slnl)

system 1. Those with e < P/Slnl do not purchase either system. This leads to the following demands when both systems are available:

P2-PI PI

S2n2 -Slnl slnl

and

If only one system is available. as would be the case if upstream firm i captured all the downstream firms in exclusive dealing contracts. the consumer wiII purchase the system if e ~ p/sjn j. In this case. the demand for system i would be: 42

Note that at some p, say pO, D1ED (pO) ~ D/D (pO) so that upstream firm 2 is dominant

in the demand sense.

2.3 Game Description

Stage 1: The upstream firms simultaneously decide whether to offer ED contracts.

They then choose Pyi and Pwi and present offers to the downstream firms.

These offers consist of Pyi' Pwi and perhaps an ED clause. We will define

Stage 2: If at least one ED contract is offered in Stage 1, the downstream firm

must decide which upstream firm to deal with. If not, it deals with both

upstream firms. The downstream firms set the retail price fer the Zi'S.

Stage 3: Consumers purchase according to Dl (PI' P2) and D2 (Pl' P2) if both

products are available and according to DiED(pi' Pi) if only system i is

available. 43

2.4 Equilibrium Without Exclusive Dealing

If no ED contracts are offered by the upstream firms in stage 1 , during stage 2, each downstream firm will maximize profits

where llsn is equal to S2n2 - Slnl' by setting

and . Pz2

Given these retail prices for the z-components, the upstream firms' profit functions become:

n,1 = (P,1 + n1P.1 - SIC, - nlc,l (P~:I - :~J

WI-mel --- (wZslnl - wls2nZ) 2ilsns1nl 44 and

II., = Ip" + n,p._, - s,c, - n,c,) (a - p~:11

w2-mc2 (6 = 6Asn - W + WI) , 2Asn 2

where Wi = Pyi + niP",;. The reaction functions resulting from these profit functions arc:

= .! (Slnl w + mc 1 2 s n 2 1 2 2 and

Solving simultaneously we obtain:

• 1 wl (8Asnsln1 + mCzS1n1 + 2mc ls2n2) 4sZn2-slnl • 1 Wz = (26Asns2n2 + mc ls2n2 + 2mczSznz) • 4sZnZ-sln1

Upstream profits at w,· and w2' are: 45

and

_1_ (2e~snS2n2 + mc1s2n2 + mC2(Sln l -2s2n2))2 2~sn 4s2n2-sIn}

2.5 Equilibrium With Exclusive Dealing

If one or both of the upstream firms offer ED contracts in stage 1, then only one system will be sold. So the final demand for the system which the downstream firm decides to sell will be

Regardless of which upstream firm the downstream firm chooses to deal with, he will choose Pzi in order to maximize: 46 Therefore, he will choose

1 - - (as.n. - p . + n,·pw,.) 2n., " )'I

and make profits of

1 - 2 -- (as.n. - '"') . 4s.n., , "

The downstream firm will accept the contract which provides him with the most profit. so in order for upstream firm 1 to persuade the downstream firm to accept his contract, the following condition must hold:

In other words,

This implies that upstream firm 1 must set a combination of prices such that 47

Since the best offer that upstream firm 2 can make is to set w2 = mc2• upstream firm

1 will set a price

Define Gs, = Gs,n; - mc;. By setting W1ED. upstream firm 1 will make

e (Slnljl/2 -e (Slnljl/2] s2 S n as) s2 S n ED( ED) 2 2 2 2 nu} WI

Similarly. in order for upstream firm 2 to persuade the downstream firm to accept his ED contract. it must be true that:

which implies that 48 Therefore.

Since the best offer upstream firm 1 could ever make is WI = mcl' upstream firm 2 would set

and make profits of

ED( ED) nu2 W2

In stage 1. upstream firm 1 would choose to offer an exclusive dealing contract only if conditions (1) and (2) hold.

(1)

(2) 49 Upstream firm 2 would choose to offer an exclusive dealing contract only if conditions

(3) and (4) hold.

(3)

(4)

(1') i..l/2

Upon closer examination of (2'), we see that it is a quadratic in o.

This is a parabola which opens upward. :"herefore, condition (2') wiJI hold when .§, $ o $ 6,. Using the quadratic formula to determine Q, and 6" we find that

21.. 3/2 (2i.. -1) +(i.. -1)(4i.. -Ii -(i.. -1)1/2(4i.. -1)(8i..s/2-8i..2-4;' 3/2+5i.. -1 )112 2i.. -1/2(i..-l)(4i..-l)2+2)..3/2 50 and

2A 3/2 (2A-l)+(A-I)(4A-Ii + (A-I)l/2(4A-I)(SA SI2 - 8A2-4A 3/2 +SA-l)1(2 2A -1/2 (A -1) (4A -I)2+2A3/2

Therefore, the three conditions necessary for upstream firm 1 to choose to offer an exclusive dealing contract are:

(i) (8A 5/2 - 8A 2 - 4A 3/2 + 5A - 1) ~ 0 (ii) Q. s (, S 51 1 (iii) A1/2 ~ (,

Condition (i) is always true for A ~ 1 (or S2n2 ~ s1n 1), and condition (iii) is implied by condition (ii).

For upstream firm 2, conditions (3) and (4) become:

(3') A1/2 s 8 (4') A1/2 (, _ A ~ ~ [28 A - A - (,]2 A-I 41.-1

Condition (4') is a quadratic in 6 as follows: 51 Therefore,

3 2 o = 21. / (21.. -1) +(1. -1)(41.. -li-(1.. _1)1f2(41.. -1)(81. Sf2_81.. 2-41.. 3f2+S A_1)1/2 ~ 21..1f2(21.. -1)2

and

6 = 21.. 3/2(21.. -1) +(1. -1)(41.. -1)2 +(1.. _1)1f2( 41.. -1) (8I.. Sf2 -8",2_41.. 3f2 +5'" _1)1/2 2 21. 1f2(21.. -1)2

The three conditions necessary for upstream firm 2 to choose to offer an exclusive dealing contract in stage 1 are then:

U) (81.. 5/2 - 81..2 - 41.. 3f2 + 51. - 1) ~ 0 (ii) Q ~ 0 ~ 6 2 2 (iii ') 1.. 1/2 ~ 0

Again. condition (ii') is the determining condition.

2.6 Results

These results can be summarized in FIGURE 2.1 below. The equilibrium will depend on the relationship betvveen 0 and A. In region J, upstream firm 1 would 52 choose to exclusively deal. In region II, upstream firm 2 would choose to exclusively deal. Anywhere else, neither firm would offer an exclusive dealing contract and the downstream firms would sell z-components for both systems.

10

8

6

4

o 2 4 6 8 10

Exclusive Dealing Ranges

FIGURE 2.1 53 In region I. firm 1 chooses to impose exclusive dealing. and prices and profits would be as follows:

and

In region II. when firm 2 chooses to exclusively deal. prices and profits would be as follows: 54 and

- mc/

Anywhere else, prices and profits would be:

p,"] = _1_ (8s n - 211 1 1 1 " 1- P:2 = - (8s2n2 2112

___1__ (8Asn Slnl + mc s n + 2mc s n ) 2 1 1 1 2 2 4s n - Slnl . 2 2 'W ___1__ (28Asn S2 2 + mc s n + 2mc s n ) • 2 n l 2 2 Z Z 2 4s2nZ - Slnl and 55 The ability of either firm to impose exclusive dealing is dependent on the

relationship between the quality of its system and its marginal cost. A firm can

"purchase" quality either by increasing the quality of its y-component (thereby

increasing SjCY) or increasing the number of its z-components (thereby increasing njcz).

It is therefore possible for a firm to increase the total quality of its system without

proportionately increasing its marginal cost. It can do so by increasing the system

quality using the cheaper component. It is even possible for the high quality firm. firm

2, to have a lower cost than firm 1. A necessary condition for this to occur is that the

high quality firm does not have both the most z-components and the highest quality y-componem. The high quality system could have a lower marginal cost if, for example, most of system 2's quality comes from the number of z-components and Cz is significantly less than the cy or vice versa. All that is required is that (n2-n1}/(sl-S2) <

Firm 2 will be able to impose exclusive dealing contracts if <5 ~ ~2' which means that

e - mC2 82112 ~ ~. e _ mc} }.. 8111} 56 This implies

ED D (mc ) "co -~ D 1 ED(mc l') 2 2 A

§.2 < A when A > 1, so §'.JA < 1. Therefore, the previous statement means that when both systems are priced at marginal cost, system 2 must be able to deliver at least a certain percentage (less than 100%) of the customers that system 1 could deliver if it were priced at marginal cost. FIGURE 2.2 indicates the size of §'.JA for different values of j.. It is evident as A increases, system 2 must provide a smaller and smaller percentage of the customers than firm 1 could deliver at marginal cost.

If mCjs2n2 i mc/sln p the statement above is always true since firm 2 would be

able to deliver more customers than firm 1 at marginal cost. This implies that mC.Jmc I

, s2n:l51np so if the ratio of the marginal costs of the two systems is not as great as the ratio of the qualities, firm 2 will be able to exclusively deal. This is a distinct possibility with this model because ofthe way marginal cost is determined. If mc.Js2n2

~ mc/sln!, the condition necessary for exclusive dealing can be met if mc.Js2n2 is not

"too" much greater than mc/slnl. 57

Q,. -A 0.8

0.6

0.4

0.2

o 5 10 15 20 25 30

Percentage of Customers Firm 2 Must Deliver

FIGURE 2.2

Firm 1 could use exclusive dealing contracts when

Since 'A/6) > 1, system 1 must deliver a larger number of customers when priced at marginal cost than system 2 when it is priced at its marginal cost. This will occur when mc/s)n) is significantly smaller than mc:!s2n2' FIGURE 2.3 indicates the size of 58

')../61 for different values of ')... As the systems get further and further apart in terms of quality. system 1 must be able to deliver a larger and larger percentage of the customers that firm 2 could deliver at marginal cost.

,l. 10 c\

4

2

0 5 10 15 20 25 30

,\

Percentage of Customers Firm 1 Must Deliver

FIGURE 2.3

We would also like to know when an equilibrium with ED wil! provide a lower retail price for the system that is sold than an equilibrium without ED contracts. Since 59

the downstream firms will choose Pzi = 1I2ni (esini - Pyi +ni Pwi) regardless of whether

an ED contract is offered, the final price of the system wiII be Pi = 112 (esjnj + w;l.

Therefore, we will examine when Wj' ~ wjED •

If upstream firm 1 offers an ED contract, the final price of its system will be lower under exclusive dealing when

2>" 3{2 13 >----- 4>.. - >..1{2 - 1

If upstream firm 2 offers an ED contract, the final price of its system will be lower under exclusive dealing when

13 < 4i.. - 1 - >..1{2 2>..1{2

These areas have been superimposed on FIGURE 2.1 and shown in FIGURE 2.4. The areas in regions I and II in which wholesde prices, and therefore retail prices, will be lower have been shaded.

When imposing ED contracts, firm 2 must provide the retailer as much profit as firm 1 could if the system were sold at marginal cost. If the demand for system 2 at marginal cost is large enough in comparison with the demand for system 1 at its marginal cost, firm 2 wiIl be able to do this while also raising the wholesale price.

Firm 2 will be able to raise the wholesale price, and consequently the retail price will increase. when 60

Ie ~

8

€

4

2

o 2 4 6 8 10

Exclusive Dealing and Retail Price

FIGURE 2.4 61

It is clear from FIGURE 2.4 that (4A-A1I2-1)/2 A3/2 > Q/A. Therefore. in order for firm 2 to impose exclusive dealing and raise its wholesale price at the same time. firm 2 must be able to deliver a larger percentage than that necessary simply to exclusively deal.

When imposing ED contracts. firm 1 will be able to increase the wholesale price when

4A - A1/2 - 1 ED ED(me ) > D (me) DI l 2 2' 2A 1/2

1 2 I12 Again, it can be seen from FIGURE 2.4 that (4A_A / _1)12A >A/6 1•

In order for the wholesale. and therefore retail. price to decrease, the firms have to be asymmetric enough to be able to impose exclusive dealing contracts on the retailers, but not asymmetric enough to be able to avoid competing for those contracts using the wholesale price. 62 3. DID NINTENDO USE EXCLUSM DEALING CONTRACTS TO MAINTAIN ITS MONOPOLY POSITION IN THE U.S. HOME VIDEO GAME INDUSTRY!

3.1 Introduction

Many of us grew up laughing at the hair-brained schemes of Lucy and Ethel. bemoaning Gilligan's mishaps, and learning right from wrong with Wally and Beaver.

Now, instead of watching others have adventures, children are having their own -- conquering evil forces in the universe, fighting from level to level while the danger increases and the risk of death looms directly before them, and perhaps finally beating the foe, all without ever going any further than their living room. Home video games are standard fare in the homes of American children. Many spend hours with their eyes glued to the television screen and their hands fastened to a controller.

In 1992, the home video game industry grossed more than all the U.S. movie box office features combined. 1 The largest manufacturer and marketer of those games is Nintendo. Nintendo supplies all the hardware and software needed to change the living room television into a video game player. The console, which is the size of a phonebook, is hooked by a wire to the back of the television, and a reusable game cartridge can then be inserted into the console. Nintendo also provides push-button controllers that allow input from the player. It is estimated that more than one-third of all households in the U.S. has a Nintendo system. 2 63 Nintendo entered the U.S. home video game market in 1985, and our society will never be the same because of it. Nintendo immersed its players in Nintendo products, featuring its game characters on everything from feature length films to soda pop and cereal. Nintendo became more than a popular product, it became a way of life for many children. They could watch Nintendo cartoons in the morning, carry their lunch to school in a Mario Brothers lunchbox, come home and read Nintendo Power, and of course, playa wide assortment of games. In 1990 a poll of U.S. schoolchildren showed that the hero of Nintendo's most popular game series, Mario, was more popular than Mickey Mouse. 3 The revenues from one of those games, Super Mario

Bros. 3, reached more than S700 million to make it the best-selling video game of all time." The North Pole Poll indicated that the Nintendo Entertainment System (NES) was the number one selling toy, both in terms of dollar and unit sales, for the 1987,

1988. and 1989 Christmas seasons.s Nintendo has even affected the health of our society. allegedly being the cause of some ailments and the cure for others.

Nintendo has been consistently recognized as one ofJapan's best companies by leading Japanese business journals.6 The kanji characters that make up the name of the company -- nin-ten-do -- have been commonly translated as, "Work hard, and the rest is in Heaven's hands." As one journalist put it, "they're working hard and leaving as littie to Heaven's hands as possible.,,7 Not everyone views Nintendo as the hero, however. There has been a widespread feeling that Nintendo may be an evil. monopolistic force in the universe that must be conquered if the consumers are to be 64 saved. It has been alleged by many opponents that Nintendo's dominance in the industry was the result of unfair and monopolistic trading practices.

3.2 History of Nintendo

In 1889 Fusajiro Yamauchi founded a company to produce and sell cards for the traditional japanese card game of "hanafuda." Nintendo's "hanafuda" cards became the most popular cards in the Kyoto region. In 1907 Nintendo became the first japanese company to manufacture Western-style playing cards. By the time Fusajiro was ready to retire in 1929. Nintendo was the largest japanese playing-card company. Under

Nintendo's second president. Sekiryo Yamauchi. the company became an efficiently run business with a rigid management structure. Among other things. Sekiryo developed an assembly line for producing the cards. In 1949 Fusajiro's great grandson.

Hiroshi Yamauchi. was appointed as the third president of Nintendo. He renamed the company Nintendo Company. Ltd. and broadened its product line. In addition to continuing Nintendo's lucrative card business. he introduced a line of individually portioned instant rice. opened a hotel. and started a taxi company. In 1959 he arranged a licensing agreement with Walt Disney. which allowed Nintendo to produce playing cards backed with pictures of Mickey Mouse and other Disney characters.

In 1969 the company started its "Ultra" series of toys -- Ultra Hand (a clasping extension of the hand). Ultra Machine (a pitching machine). and Ultra Scope (a 65 periscope that aIlowing viewing around corners, over fences, etc.}. The company also

produced a laser-light gun and shooting ranges where these guns could be used to

simulate the shooting of clay pigeons. Nintendo's first venture into arcade video games

was with "Donkey Kong" in the mid-1970s, and it eventuaIly became a recognized

leader in coin-operated arcade games. In 1977 Nintendo, together with Mitsubishi,

created the Color 1V Game 6, which was a game machine that played six versions of

light tennis. Color 1\1 Game 15 foIlowed.

In the early 1980s, systems in America were being produced that played several games on interchangeable cartridges. With this new kind of system, the hardware was

still in demand long after the interest in a particular game was gone. Yamauchi

decided to develop a similar system for Nintendo, and the Famicom (short for Family

Computer) was introduced to the japanese market in 1983. Although Nintendo ended up having to recall its first set of systems due to the malfunction of one of the integrated circuits, the Famicom was successful. Atari, Commodore, Casio, and others also began releasing game systems in japan at this time. Sega, a small japanese arcade­ game company, released the SG-1000 in the same year that Nintendo released the

Famicom. but it didn't do very well. Nintendo took the lead in the japanese home video game market.

Yamauchi decided that in order to increase the variety of games available to the customer. it would be wise to let outside companies also create games for the

Famicom. However. he placed some restrictions on these companies. Nintendo limited the number of games each licensee could produce in any given year. and after the first 66 six licensees. insisted that Nintendo manufacture all the cartridges that would be played on its systems. Nintendo had changed its manufacturing policy after some defective games reached the market when the licensees were allowed to manufacture their own cartridges. The companies also had to pay Nintendo a 20% royalty on every game cartridge sold.s One of the first licensees. Hudson. quadrupled its annual profit in

1984 because of a game it released for the Famicom.9 By 1985 there were around 20 licensees. By 1988. there were fifty. By 1990. there were seventy licensees and hundreds of games. almost all manufactured by Nintendo Company. Ltd. By 1991. almost every household in Japan with children had a Famicom system. lO

3.3 The U.S. Home Video Game Industry Before Nintendo

In 1972 Nolan Bushnell. the founder of Atari Inc.. placed "Pong." the first commercial video-arcade game. outside a tavern in California. "Pong" consisted simply of a slow-moving white blip on a black field. but it caught on. In 1973. after 6.000 games were sold for about S1000 each. Bushnell sold the rights to the game. One hundred thousand "Pong"-type games were produced in 1974 alone, with only a fraction being produced by Atari. ll In 1974 Atari made a "Pong" system for the home.

By the end of 1976. over a dozen different companies were making home video­ game systems for the American market. Atari introduced the 2600 and captured 44~o of the market share. By 1982. there were 1500 games available for it. Sales in the 67 industry were S3 billion, with Sl.7 billion of that belonging to Atari. Atari let anyone who wanted to create software for its units do so. Many independent companies made a lot of money manufacturing games for the 2600.

The market imploded in 1983 when video makers began to sell what they now admit were junky video machines and dull games. 12 It is generally felt that video game manufacturers, hoping to cash in on the boom, flooded the market with a lot of poorly developed and boring games. The market grew stagnant. There was no new technology or the introduction of new game categories. The games replicated each other in content. image. and level of challengeY Consumers grew disenchanted.

With so many games available, they were no longer able to differentiate between them to determine which ones. if any, were interesting and fun. Even though sales were down. Atari and dozens of other companies were still making games by the millions.

Inventory levels were enormous, so massive discounting began. Garnes that were meant to be sold for S35 were selling for S5. Atari even built more "Pac-Man" cartridges than there were players!14 Atari had huge inventories that it couldn't sell and nearly went bankrupt. The company was divided up and sold. The hardware divisions became Atari Corporation, and the coin-operated business became Atari

Games. At the time of the division, there was an agreement made that Atari Games could do anything except make hardware or software that competed with Atari

Corporation under the Atari name. The U.S. horne video game market had reached its peak in late 1982. with sales of S3 billion. By 1985, only S100 million worth of horne 68 video games were sold and most video game companies had withdrawn from the market or reorganized.

3.4 The Market Test

In 1979, Hiroshi Yamauchi sent his son-in-law, Minoru Arakawa, to America to break into the coin-operated arcade business, which at the time was one ofthe largest entertainment industries in the United States. They imported Nintendo's arcade games directly to America. One of those games was "Donkey Kong," from which the popular character of Mario the carpenter originated. By 1985, Nintendo Company, Ltd. was making a lot of money on the Famicom in Japan. Almost all of the revenues from the

American subsidiary, Nintendo of America, were from coin-operated games. Nintendo was ready to begin selling the Famicom in America. There were discussions between

Atari and Nintendo about the possibility of Atari selling the Famicom in the United

States and Europe. However, the deal never went through, mostly because Atari was falling apart at this time. 1s

Arakawa began an investigation of the American home video game industry to see if launching a game system in the U.S. was feasible. Arakawa found that the industry was a mess. People in the toy industry had just been burned by the whole home video game business and were not interested in being burned again. However, even though the home video-game industry was virtually nonexistent, people were still 69 playing arcade video games. Because of this evident enthusiasm for video games,

Arakawa determined that it must have been something other than a lack of interest which destroyed the horne game industry. When he interviewed people involved in the toy business, he found that the general feeling was that the industry crashed because there had been an over-supply of poor quality games. The arcade business was still alive and well because it wasn't like the home video-game business in many ways. In the arcades, people can simply stop playing a game if it's boring, and they've only wasted a quarter. The same was not true in the home video-game business. Since there was no inexpensive way to test a home video-game, customers had to go by the packaging and advertising. If the customer did not get a game that lived up to the claims on the box, he or she had wasted S40 or S50. If this happens too many times, the customer stops believing what is printed on the packages. In addition to the hands-on testing that could be done in the arcades, arcade games were faster moving and more complex than home video games because they had more memory capacity.

The games were just more fun.

In January of 1984, Nintendo showed its American version of the Famicom, the

Advanced Video System (AVS), at the Consumer Electronics Show. It didn't get a very good reception, so Nintendo decided to redesign it to make it more like a toy, less like a computer. The resulting system was simple -- no computer jargon, no keyboard, and no other function than playing games. It was renamed the Nintendo Entertainment

System (NES). and it was test marketed in New York in 1985. 70 Since retailers were reticent about getting involved with a home video-game system again. so Nintendo had to make its deals very attractive. In order to induce stores to give the product a chance. they were given the merchandise free for 90 days. after which they had to pay for what they sold and could return the rest ifthey wished.

Nintendo also committed to a large multimillion dollar advertising budget. By the end of its first year. a million systems had been sold in America. proving that the home video-game market had not been completely dead as many had insisted.

By 1988. Nintendo had seven competitors who were marketing home video­ game systems, but Nintendo had 85 to 90% of the markets both in Japan and the

United States. During this time. Atari was selling its 5200s and 7800s. and Sega sold a few billion of its Master Systems. Nintendo had gone from a small. unknown company to where it accounted for 20% of the U.S. toy industry sales. Its contracts with retailers became tougher. no longer including the money-back guarantee. instead requiring payment upon receipt. By 1989. there were NES units in one out of every four American homes. 16 Nintendo had S2.7 billion in sales. about 80% of the industry'S S3.4 billion. 2396 of all the money spent on toys was spent on Nintendo products. and Nintendo accounted for 25 of the top 30 selling toys in America. 17

Games typically sold out within days. sometimes minutes. after appearing on the shelves. Game companies sometimes had to wait as long as 4 or 5 months for

Nintendo to fill an order.

By 1992. Nintendo had a dedicated player base of 70 million people worldwide. 18 By 1993. there were 600 or so software titles available. By May of 71 1993, the cumulative worldwide sales of the seven Mario adventure games totaled 100 million, putting Mario in a very select group of entertainers. 19 In the years 1988-1993,

Nintendo had 27 different multimillion selling game titles. 20

What allowed Nintendo to succeed in an industry that was practically nonexistent? After all, there were more game titles available by the end of 1989 than were available during the video game boom in the early 1980s, and the vast majority were in the Nintendo format. 21 Perhaps because Nintendo brought hardware and software that had already been successful in the Japanese market. It had also introduced a product that was technologically superior to its predecessors. For instance. "Pong" had limited (8-kilobits) memory computer chips, which allowed only primitive graphics and limited movement capabilities. 22 Games in arcades had larger memory capacity. so the softvvare could be more elaborate and long-playing. This pro\'ided for more complex games. Nintendo moved closer to the arcade style. introducing more sophisticated games when it entered the market. Some of its games contained 1 and 2 megabits of memory. In addition, Nintendo exercised a great deal of control over the software available for its system .- it controlled both the content and packaging and the number of copies available for sale. Nintendo also monitored the market very carefully and had a wide range of customer services and customer contact, including a game counseling hotline, a system magazine, and strong advertising support. In a corporate press release, Nintendo reported that an independent research study had determined that Nintendo's success could be traced to the company's strong brand imagc.23 72

FIGURE 3.1

Home Video Game Industry Sales 1979 - 1993

$ Billions 6~------~------~ • Industry Sales o Nintendo Sales 5 ...... - _ .. - .. -

4 ...... - _. -. _ ..... - .. -- _.

3 ......

2 . - . - ...

1 _... o 1979 1982 1985 1988 1991 1993 73 Nintendo might argue that it succeeded in the industry because of its persistence and commitment to the future weIJ being of the company. Many other businesses might have given up in the situation that originalJy faced Nintendo in the

American market, with so much resistance from those in the industry. After the market crashed. nothing much happened in the industry until Nintendo came along. While

Nintendo insists that it gained its dominance fairly and squarely, others charge that it succeeded because of iIJegal practices, such as price fixing, monopolistic control, and intimidation of retailers. Complaints against Nintendo's "questionable" practices came from all directions.

3.5 The Security Chip

Nintendo did a one important thing differently than its predecessors in the industry. and it has been a source of much contention for Nintendo. When Nintendo decided to introduce its system in Americ2., it wanted to control the quality of software released for it. Nintendo Company, Ltd. had experienced some trouble with counterfeiting in the Japanese market and didn't want to experience the same problems in America. Unlike Atari, who allowed anyone to create games for its system, Nintendo insisted on controlling who designed games and what types were created. To enforce this control. it placed special electronic chips in its machines to block the playing of games not approved by Nintendo. The console wouldn't work unless a chip in the cartridge unlocked the chip in the console. This security system aIJowed Nintendo to 74 both censor the games and to stop counterfeiters. Critics of the chip suggested that

it served no useful purpose other than to block competitors, so it was illegal.

However, Nintendo was issued a patent on the chip in January of 1989 by the U.S.

Patent Office. None of its systems in Japan had a security chip.

3.6 Licensees

Nintendo believes that "The Name of the Game is Games.,,24 Nintendo says that the secret to its success was to keep the price of the machines low to sell in volume and to keep the games fun. 25 Therefore, software produces the profit. It has professed many times that it believes the key to doing business in the home video game industry is to selI fun and challenging games. Consumers buy video game hardware not for its own sake, but for the games that can be played on it. Software is where the company can differentiate its product from the products of others in the industry.26 In 1989 the average consumer who purchased a Nintendo system ended up buying an average of seven game cartridges at an average price of S40. 27

The first software that was available for NES in America was made by Nintendo.

However, Nintendo knew that there was going to be a need for more games than it could produce itself. While Nintendo has its own development staff and has produced

between 2590 and 40~o of the Nintendo software market share, the rest is created by third-party game developers who sign a licensing contract with Nintendo. These 75 licensees bear all the risk of developing and marketing new titles. It took six months

for the first licensees in America to sign up, and those that did were American

subsidiaries of Japanese companies that imported video-arcade games. These

companies set about making home versions of their arcade games. The first American company to sign on, Acclaim, did so in 1987. It had been created solely for the purpose of making games for the NES.

Considering that Nintendo feels the viability of its product is dependent on the software available for its system, it is not all that surprising that it would want to maintain as much control as possible over the games. The contracts Nintendo had with the companies that developed games for its systems contained the most restrictive terms Nintendo believed it could legally impose. Nintendo's opponents felt that the contracts were too restrictive. causing anticompetitive results. The contracts came under legal scrutiny as the possible means by which Nintendo achieved its dominant position in the market.

The licensing contracts contained several restrictions. First. Nintendo limited the number of game titles a company could create for its system. Initially, Nintendo signed contracts that limited the number of games a licensee could develop in one year to five. This was supposed to help control quality. It prevented companies from making up for poor- selling games by simply creating lots of other poor-selling games.

This limitation was to provide developers with an incentive to make each game count.

However. one of its early licensees. Konami. convinced Nintendo to allow it to form a new company so it could get around this limitation and get a second license. 76 Nintendo later modified the practice by only making agreements that covered a single game. If a developer had a sequel or new product, it had to negotiate a new contract.

Second, Nintendo reserved the right to approve the game, packaging, artwork, and commercials. Nintendo evaluated every game on a forty-point scale. The system had eight categories, each worth five points. Nintendo would then "strongly recommend" the number of copies of the game that should be issued accordingly.28

Nintendo-approved games bore its Official Nintendo Seal of Quality.

Third. the cartridges were produced by Nintendo Company, Ltd. in Japan and could then be purchased by the licensee. with a minimum order of 10,000. The

Japanese firms that were allowed to manufacture their own cassettes were not allowed to manufacture the cartridges for their U.S. subsidiaries. Depending on the memory capacity. Nintendo charged between S9 to 514 per cartridge. This included the cost of manufacturing. printing and packaging the game, and the royalty for the use of

Nintendo's licensed intellectual properties. 29

Fourth. the licensee had to agree not to release the game to be played on any of Nintendo's competitors' systems for two years. Fifth, the games could not be sold outside of the United States or Canada.

Even with so many restrictions, many companies signed on. Nintendo's vast share of the domestic hardware market was an incentive for software developers to

"hitch their wagons to Nintendo's star, ensuring a steady stream of high-quality games."30 The only companies that seemed to be complaining about Nintendo's strict licensing agreement were those that couldn't get in. As long as their earnings were 77 large, the companies didn't mind paying Nintendo a large share of them in exchange for the opportunity to make the money in the first place. 31 In 1989, there were 40

U.S. licensees, and in 1990, there were 53.

3.7 Marketing Practices

Nintendo was one of the great marketing successes of the 1980s.32 Much of the source of Nintendo's success may be attributed to its unique marketing program that keeps it in close touch with its customers. Nintendo publishes a magazine for its system, has a telephone hotline that customers can call with questions, prints brochures for parents, and keeps very close track of retailer sales.

3.7.1 Nintendo Power

When Nintendo entered the U.S. market with the NES, everyone who sent in a warranty card automatically became a member of the Nintendo Fun Club and began receiving newsletters. There were over one million Fun Club members by early

1988?] Nintendo realized there was a market for game information and tips, so rather than allowing outside magazines to make the money (as they had in Japan),

Nintendo decided to start a magazine devoted to NES in America. In 1989, the first issue of Nintendo Power went out to all of the Fun Club members. The subscription price became S 15 for 12 monthly issues. At the end of 1989, Nintendo Power had 78 become the largest-circulation magazine for kids in America. 34 By 1990. it had 6 million readers.35 In 1992. it was the highest paid subscription publication for kids in the 8-15 year-old category.36

The magazine has been called "a marketing person's dream: a monthly brochure for Nintendo games that the customer pays for."37 The magazine had a lot of persuasive power with the kids. If readers saw a game featured in Nintendo Power. they would be ready to buy it. Nintendo started designing direct contact with the consumers into its product by leaving out some game features from its manuals that weren't necessary for play. but enhanced the ability to solve the game most efficiently and creatively. If the player wanted tips on how to do this. he/she could look in

Nintendo Power or call the Nintendo Hotline.

3.7.2 Nintendo Hotline

Part of Nintendo's success was due to being able to make decisions based not on retailers' opinions about buying trends. but on direct contact with the customer.

Because it began receiving a lot of calls from kids about their systems or games. a game counseling phone service was started in 1986. It allowed Nintendo to gain much more insight into what consumers wanted. and it allowed the consumers to bond with the company in a unique way. Children were more apt to have a friendly feeling towards the company.

Initially. it was an 800-line service. but was changed to a 900-line service in

1990 because it became too expensive. Nintendo hired game counselors to field the 79 calls and dole out advice from 4 am to 1 D pm. By April of 1993, Nintendo had made its 3D-millionth contact. Nintendo claims that "No one in the industry comes close to this level of connection with real customers. We know what will work because we just talked to the consumer."38 Nintendo also has consumer representatives to answer questions about repair, assembly, and even the availability of certain games in local stores.

3.7.3 Keeping Parents Happy

Parents were complaining that they couldn't get their children away from the

T.V. set. that their kids were addicted to video games. This made them angry with

Nintcndo. They were concerned with the amount of time their children were playing games instead of participating in learning activities. As an overt attempt to counteract this image. Nintendo funded an MIT faculty member 53 million to develop educational software for video games. Nintendo also made brochures available to parents, helping them understand video games and their language. giving helpful suggestions on how to help their children develop a balanced lifestyle which included, but not exclusively. video games, and advocating playing video games as a family.

Nintendo even affected the health of Americans. Diseases were discovered and named after Nintendo. For example, Nintendinitis is an inflammation ofthe tendon on the back of the thumb caused by prolonged pushing of the buttons on the control panel. In addition to being a health detriment, Nintendo products have also been part of the cure of certain ailments. Video games and computers have been used in the 80 rehabilitation of stroke and brain injury patients, improving motor control, reaction time, and scanning ability. They have also been found to be helpful in combatting some of the effects of old age. Of course, kids also found them helpful in counteracting boredom during long hospital stays?9

3.7.4 Game Management

Nintendo marketed its games like movies -- releasing them carefully, rationing them so demand outpaced availability, and withdrawing them at the first indication of a slowdown. Nintendo's merchandizing team carefully monitored toy shelves, seeing to it that the product was well displayed, checking sales and in-store inventory.

Nintendo would switch new titles for old ones so that they didn't go bad on the shelf.

Blaine Phelps. who trains game counselors for Nintendo, says, "Our analogy is

Hollywood. We are not here to talk the world into buying more theaters. We are here to be the Steven Spielbergs who create the stuff that makes them want to come inside."4(J

Nintendo intentionally did not fill all of its developers' and retailers' orders for games. In 1988 Nintendo sold 33 million cartridges. Surveys showed that it could have sold 45 million. Retailers had requested 110 million. 41 This is interesting because it indicates that Nintendo could probably have raised the price of its cartridges and sold the same number, thereby increasing its profits. Obviously it must have had another goal in mind. By creating a shortage, the product became a scarce commodity that people are willing to wait in long lines for. Other customers see the shelves in the 81 stores empty, with signs indicating the game has been sold out. This created a much

different atmosphere in the industry than that which was present during the first craze where there were more copies of more games available than the buyers wanted. It seems to have been an attempt by Nintendo to keep customers from getting bored with its products. Since Nintendo had control of the number of cartridges available, it was able to adjust the supply of software while watching the movement of the market. Since they seemed to have a good handle on what consumers wanted because of their direct contact, the system seemed to work well for them.

Nintendo heavily promoted the games it licensed. It would spend millions of dollars on advertising in a year. In addition, consumers were encouraged to participate in the Nintendo World Championship video game competition. Players were also able to test new games at this exhibition that traveled to 35 cities in 1990.

Nintendo's game characters branched into other areas of entertainment, both into cartoons and the big screen. In September of 1989, a series based on Mario, the

"Super Mario Bros. Super Show" was ranked third among all nationally syndicated childrens' programs. 42 Another Nintendo character cartoon, Captain N, was the number one Saturday morning show among 6-11 year-olds.43 The Wizard, a movie based on the story of a young video game player participating in competitions like

Nintendo's, was released in 1989, and a movie about the Mario Brothers was released in 1993. 82 3.8 Competitors and Advances in Technology

Nintendo's largest competitors in both the U.S. and japanese markets, although

only a distant second and third, were also japanese companies. NEC was a computer

company, while Sega was a video arcade game company. Until about 1991, together

they shared the 10 to 15 percent of the home video-game market that Nintendo didn't

control. NEC was a successful computer company, and Sega, which was started in the

1950s by an American businessman in japan, was relatively successful in the video

arcade business. Sega's Master System competed with the Famicom and the NES, but

it never gained much of either market. The Famicom, NES, and Master System all

used 8-bit technology. NEC never introduced an 8-bit system. Bit size measures how

much information the processors can handle at once. An 8-bit processor can work with

strings of information that are 8-bits long, while a 16-bit processor can process

messages made up of 16-bits. A 16-bit processor can process more messages, making

a 16-bit machine twice as powerful as ar. 8-bit machine. 44 A 16-bit processor allows

more complicated and intricate games. It provides longer play, more colors, cleaner

graphics, and animation that looks more like motion pictures.

Sega introduced the Genesis system, the first video-game system powered by

a true 16-bit processor, to the United States late in 1989. NEC introduced its

TurboGrafx system in November of 1989, but it was not a true 16-bit machine. It had

a 16-bit graphics processor, but only had an 8-bit central processor. 83 For software, Sega already had arcade hits that it could translate to run on the

Genesis. It also released a converter, priced at about 535, which allowed the 8-bit

Master System games to be played on the Genesis. "Bring the arcade experience

home," was the tag line for Genesis. Sega's new system had clearer graphics and more

sophisticated effects than the NES, and it also had stereo sound. Still, Genesis wasn't

an overnight success. There were relatively few usable 16-bit games on the market

initially, so the players had little reason to purchase it. Sega began contracting with

third party licensees in 1989. However, it was difficult to get companies to write

games for it because many were tied to Nintendo. Nintendo's games were clearly

inferior from a technological point of view, but the games were still appealing because

their style was different. Yamauchi claims that Nintendo characters come to life

because the players could relate to them. 45 NEC and Sega weren't able to match that

style. The initial Genesis games weren't as much fun as the best Nintendo games. 46

The programmers may have been trying to exploit all the new possibilities of the system and not concentrating as much on the content of the games. Initially, the consensus was that the games looked better than they played.

At the end of 1990, Nintendo still had about 85% of the industry's $4 billion in sales. However, more than 60% of Sega's 16-bit buyers were Nintendo trade-ups.47

Sega had about 60 games for the Genesis system at the end of 1990. In 1991 Sega sold about 1 million Genesis systems.48 By the end of 1991, Nintendo reported still having a large share of the market, 53.5 billion of the industry's $4.4 billion (roughly

7990); however, TVv'ICE magazine disputed this, putting the figure at 70%, leaving Sega 84 with 20%, and others at 10%.49 The NES was selling for about S100 with an average price of S40 per game,50 and Genesis was selling for S189 with games priced between

S50 to S60. 51

Nintendo indicated it had had a 16-bit system in the works for a long time, and actually began selling one, the Super Famicom, in Japan in November of 1990 , yet it did not introduce one in America until close to 18 months after Sega introduced

Genesis here. Nintendo has always professed to bank more of its success and profits on new games than on hardware. "Super Mario 3," an 8-bit game, did $400 million in sales in 1990, despite the more advanced system that was available. One industry analyst explained Nintendo's reluctance to introduce its 16-bit system in America with,

"You don't obsolete your own product at the very instant when it is getting such a tremendous boost from its software."52 Peter Main, vice-president of marketing for

Nintendo, explained that customers need to be led through the buying cycle, so they get all they can out of the current player before replacing it. Just like games, where it is important to take players through each version of a game in proper order to make sure the players enjoy their experience, players need to master the initial system first.

A player can't enjoy some games without having first played earlier versions. 53 He said that Nintendo would stress the development of more games instead of more bits, since the games rather than the machines are the secret to success. "The name of this game is the games."54 Yamauchi agreed, saying "We are confident that our strategy of maintaining a sharp focus on software will continue to win in the market place. We will employ new hardware technologies only when they have demonstrated a potential 85 to improve the playability of games."55 Nintendo's philosophy is that software

developers must first demand different hardware, rather than be forced to design for

new hardware. Nintendo also wanted to keep the millions of 8-bit Nintendo owners

happy and playing. Nintendo officials didn't believe they had adequate game software for an upgraded player. 56

Even without the help of a 16-bit machine, Nintendo kept roughly 85% of the

market in 1990. Part of these were sales from Game Boy, Nintendo's handheld game system, which vastly outsold Sega's Game Gear. Laura Buddine of Tiger Media, Inc. said, "As long as Nintendo can control the developers, it can keep the consumer buying obsolete 8-bit technology."57

3.8.1 The Super Nintendo Entertainment System

Nintendo finally introduced its Super NES 16-bit console nationwide in America in September 1991 with a price tag of 5199.95. It also began a huge advertising campaign at that time. SNES offered better sound and a broader color palette than

Sega's Genesis. However, SNES was not compatible with the NES games. Nintendo decided that the cost of doing so would have added too much to the price. 58

Nintendo claimed that in the last four months of 1991, SNE5 outsold the combined full year sales of Sega. 59 However, less than a week later, Nintendo announced plans to cut the price of SNES to 5180. Up until this point, Nintendo had never had to discount. Retailers said Sega was selling at a faster rate than Nintendo, but Nintendo contended that although that might be true at some outlets, SNES was 86 sold by more outlets than Sega. Nintendo led Sega in retail outlets 16,000 to 10,000.

One research organization, the NPD group, concluded that the video game market didn't grow much in 1991, so the gains in Sega's sales had to have come at Nintendo's expense.60

At the end of 1991, their roles were reversed, and Nintendo was the system that had far fewer games available -- 25 as opposed to Sega's 160 Genesis titles. Sega had the game variety and price lead at this time, with Genesis selling for S129, while SNES was priced at S149.95. At Christmas time, only 4% of the buyers were not system owners already.61 In 1992 Nintendo's sales of SNES were estimated to be S743 million while Genesis sales were S440 million. 16-bit sales were roughly even between the two in 1993, with Nintendo still the overall leader in the industry because of NES and Game

Boy. At the end of 1993, more homes still had video game machines made by

Nintcndo than Sega. 62 The consensus in the industry seemed to be that while Genesis had the speed, Nintendo had the more complex and involved games. In 1991, when a panel of four boys. ages 6 to 10. was C"unvened by U.S. News and World Report to determine which system they liked better. Genesis won. 63

3.8.2 The Continuing Market for NES

Once Nintendo entered the 16-bit market. members of the industry claimed it validated the new technology. However, sales of the NES still continued strong. Even though some have questioned the continued viability of the 8-bit video game market in the face of new technology. Peter Main indicated that Nintendo would always work 87 diligently to produce creative software for the NES, and that it was a top priority of the

company to have the NES remain a challenging video game system.64 The idea is that

new players can start with the NES and work up. Sales of 4.5 million NES systems in

1991 brought in S2 billion, more than Sega's and Nintendo's 16-bit sales combined.65

Nintendo relaunched the NE5 in early 1992, with a price tag S10 less than before,

589.95, and new and improved software. It sold 2 million units in 1992 and continued

to do a healthy business in 1993.66

3.8.3 Licensee Benefits

The winners in this increased competition between 5ega and Nintendo may be the software companies. For SNES, Nintendo allowed licensees to make 3 games a year, without an exclusivity clause. If a game earned more than 30 points on the evaluation scale, it didn't count as one of those three games. Some licensees could manufacture their own cartridges, but of course they still had to buy the security chips from Nintendo for S1 each and pay the royalty of 20% based on the wholesale price. 67

There were concerns in the industry that even though there was no exclusivity clause in the contracts, that Nintendo might somehow punish companies that produced for both. However, 5ega announced in October of 1991 that one of Nintendo's former licensees, Atari Games subsidiary Tengen, would develop 40 games for Genesis over a period of two years. With Sega's market share increasing, Nintendo seemed to have a little less bargaining power. Many companies that were once forced to pay high 88 licensing fees and to deal exclusively with Nintendo were now able to negotiate and begin selling their games for both systems.

The video game market began growing more slowly, with both Nintendo and

Sega having to cut prices. The number of game systems sold (mainly NES) peaked in

1989 at about 9 million. In 1990, roughly 7.9 million systems (both NES and Genesis) were sold. In 1991. only about 6.5 million SNES, Genesis. and NES systems were sold.6~ In 1994 Nintendo is projected to receive 44% of the video game sales, while

Sega gets 5490•69 Nintendo's market position and the fortunes of its licensees appeared less secure. As machines sales slowed. the number of advanced players increased. increasing the need for more sophisticated games. Companies that enjoyed fast and easy growth designing and marketing home video games found that survival became trickier. Compared with the rage of the late 80s, things changed considerably.

The video game aisles were deserted. signs indicating that particular games were sold out were conspicuously absent, customers were able to get through on their first try on the hotline. and less innovation and diversity in the games was occurring. As the video game market operates at near saturation, new avenues are being explored. The portable Game Boy player was advertised strongly as being a businessman's toy, to appeal to adults. Both NEC and Sega have come out with systems using CD-ROM technology. and Nintendo is planning on introducing a 64-bit machine in 1995. 89 3.9 Lawsuits

During the late 80s and the early 90s, Nintendo was the object of a series of lawsuits, which accused it of using product designs and licensing policies to lock others out of its business and keep prices artificially high. The Bush administration, in an effort to shift towards tougher antitrust enforcement, was keen to investigate cases involving vertical restraints. So not only did Nintendo have to contend with private lawsuits, they were also investigated by the F.T.C. One of Nintendo's employees put it aptly when he said that Nintendo's business was ''video games and litigation."io

Nintendo also was involved in several smaller lawsuits over things like the addition of new circuitry to their machines when the old circuitry was circumvented, the introduction of a product which modified the way Nintendo games played by giving players more lives and changing the characters' abilities, and the counterfeiting of games. It had its share both of winning and losing these cases.

3.9.1 Atari Games Corporation

In 1987 Atari Games, the software descendent ofthe old Atari Inc., established a subsidiary, Tengen, to become an authorized game developer for Nintendo since it couldn't do so under the Atari name. It designed three games for Nintendo. In January of 1988, a law firm hired by Atari Games obtained a copy of Nintendo's copyrighted code for its security system from the U.S. Copyright Office. It did so by indicating that the code was the subject of litigation, even though no lawsuit had been filed. In 90 December of 1988, Tengen began independently manufacturing games that played on

the NES, claiming that it had been able to get around the security system in the NES

by taking an NES cartridge apart and, using ingenuity, simply figuring out how it

worked. At tl'te same time, it sued Nintendo for S100 million, accusing the company

of monopolistic practices in violation of the Sherman Act. It alleged that Nintendo

achieved its dominance in the home video game market by creating its game cartridges

in a way that was specifically designed to thwart its competitors. It said that the

security chip Nintendo used in its system had no other use than to block competitors, and because of this security system and its dominant role in the market, Nintendo was able to control the video game market. Atari Games also accused Nintendo of wrongly limiting the number of game cartridges that it supplied its licensees, and thereby hurting their profits. Atari Games claimed that it suffered S30-35 million in lost sales because Nintendo wouldn't sell it enough cartridges to meet demand. A Tengen executive claimed that its sales increased by 40% when it began shipping the unauthorized cassettes.71 It eventually p!'oduced four new unlicensed games for the

Nintendo system.

Nintendo, in addition to filing a countersuit, alleging breach of contract, violation of state and federal trademark laws, and unfair competition, claimed that controlling the volume and quality of its video games was essential in maintaining the viability of its overall product. It also defended itself by announcing that it allocated the cartridges fairly to all licensees, and that a shortage of computer chips restricted supplies. Nintendo executives and licensees claimed that Nintendo's business strategy 91 was not monopolistic, but simply shrewd and effective, and the main reason for its enormous success.72 Nintendo asserted that Tengen had used its position as a licensee to gain technical information.

In May 1993, after a series of injunctions and appeals involving both Atari's and

Nintendo's right to sue the retailers involved and Tengen's right to sel1 the cartridges while litigation was occurring, a federal judge ruled that Atari Games had copied a

Nintendo of America computer program, and thereby infringed on a patent and copyright held by Nintendo. In a later lawsuit, a jury found that Nintendo's patent was valid when Atari Games infringed upon it.

3.9.1 Atari Corporation

Atari Corporation, the hardware descendent of Atari, Inc., also filed suit against

Nintendo. In January 1989, Atari Corp. sued Nintendo in U.S. District Court for S160 million for violation of antitrust laws. It claimed that the exclusivity clause in the licensing agreement was an unfair restrJint of trade. The suit accused Nintendo of monopolizing both the video game and player markets using exclusive contracting. In the suit, Atari Corp. said that the developers were forced to choose between selling games only to Nintendo's customers or not sel1ing at all. As a result, the software firms yielded to Nintendo's coercion, leaving Atari Corp. and other manufacturers of game consoles unable to obtain many popular games for use on their own systems. 73

Because exclusive dealing contracts had been evaluated in past cases using market share as one of the important criteria. Atari lawyers felt they had a good case because 92

Nintendo had an 80~o to 90% share of the home video game market.74 Atari's

complaint was that although exclusive licensing arrangements aren't necessarily illegal.

that they became illegal because of the market power that Nintendo possessed. Atari

believed that the practice "locked up the entire capacity of each publisher to make hit

games."75

Nintendo argued that its exclusive contracts had legitimate business justifications, like ensuring quality, and reminded the public that Atari was the company

that destroyed video games the first time around. Nintendo had survived just because

it had been stronger and smarter. Nintendo called the suit, "simply an attempt to

excuse Atari's poor competitive performance in the market place."76 Atari Corp had failed to jump back into the industry quickly enough when it rebounded. This was at the heart of their defense in court as they attempted to show that Atari's problems were self-inflicted. They tried to prove that Atari Corp. had not made any effort to be a serious competitor in the home video game business when it was re-emerging. They also stressed that Nintendo's approach LIt every point was to increase the kinds of service it provided to consumers.

Atari Corp. was unable to convince the jury that Nintendo had illegally attempted to monopolize the home video game market in the U.S. in the late 1980s.

In May of 1992, the jury found that Atari Corp. had not been damaged by Nintendo's practices, and while finding that Nintendo possessed monopoly power, the jury deadlocked on two issues: whether Nintendo had willfully acquired monopolistic power through exclusionary practices in violation of antitrust laws and whether 93 Nintendo's practice of requiring game publishers to license their games exclusively for two years constituted an illegal restraint of trade. Some concerns about Nintendo's practices had been eased over the previous year as Sega made a significant dent into

Nintendo's share ofthe market. On May 14,1992, the judge entered judgment for and awarded costs to Nintendo.

3.9.3 Federal Trade Commission

In 1989 Atari Garnes executives spoke to Representative Dennis Eckart of Ohio, the chairman of the Subcommittee on Antitrust, Impact of Deregulation and

Privatization of the House SmalJ Business Committee, about Nintendo's business practices. The jurisdiction of that committee was primarily concerned with protecting economic opportunity for smalJer enterprises. Eckart then wrote a letter of complaint to James Rill, the Justice Department's antitrust chief. In it, he said that Nintendo impeded competition and was able to raise prices by using the security chip, allowing it to obtain favorable licensing agreements with software makers. 77 Rill passed the matter on to the F.T.C. In collaboration with the attorneys general of several states, the F.T.C. began concurrent investigations into price-fixing charges and the lock-out technology. Eckart denied starting the investigation at the prompting of Atari Garnes.

He said it was done in response to consumer complaints that prices were too high and supplies too low. In his letter, Eckart indicated that because all the cartridges were manufactured in Japan, it all but eliminated any domestic base for the cartridges and thereby increased the price U.S. consumers had to pay for the cartridges. He also 94 stated that his staff was aware of a number of examples where Nintendo aggressively

exercised its monopoly power with retailers. Along with retailer intimidation, there was concern that since Nintendo's rationing of the cartridges was based on the rating

the game received, that it could be using the ratings as punishment to the game

creators. Even though Nintendo had defended the shortages of cartridges by claiming

a chip shortage, when one of its licensees found another chip supplier, Nintendo

refused to use the new chips, indicating concerns about quality as the reason. Eckart

indicated in his letter that Nintendo's practices did not seem to raise significant concerns when it first entered the U.S. market, but things changed. "One out of five

Americans in 1989 will have a Nintendo game console. Such market dominance is, of course, not by itself an indication of illegal behavior on the part of any company. It docs. however. create a business environment fraught with the potential for anticompetitive behavior and heighten the need for careful scrutiny."78

Nintendo argued that since the U.S. Patent Office issued a patent for the security chip. it recognized the validity of it. What competitors say are abuses under the

Sherman Act, Nintendo says is the secret to its revival of the video craze. It did. however. discontinue two of its questionable practices in late 1990 and early 1991. It allowed some licensees to manufacture their own cartridges, although they still had to purchase the proprietary chips from Nintendo. There was also no restriction on the number of copies of games, so licensees could order as much as they would like.

Nintendo also discontinued the use of all exclusivity provisions. There were some who fcared that even if it wasn't explicitly stated in the contract. that Nintendo had ways 95 of getting back at companies who made games for other systems -- for example, not

featuring the game in the most influential advertising spot, Nintendo Power. The

companies that were able to do their own manufacturing tended to be the biggest

licensees who were closely attached to Nintendo and wouldn't risk releasing too many games. 79 Nintendo denied loosening the restrictions as a result of government criticism. instead claiming that it believed that outsiders could finally produce defect­ free games. bO The easing of the restrictions carne when the video game business was slowing. so a glut of games was a possibility that industry watchers became somewhat concerned about.

The first F.T.C. investigation, made public in April 1991, concentrated on the price-fixing aspect. There were claims that Nintendo fixed the retail price of its consoles to S99.95. Nintendo was told that the F.T.C. was prepared to bring action against it unless it signed a settlement agreement. The agreement consisted of

Nintendo providing up to S25 million worth of coupons to some 9 million consumers nationwide -- anyone who had purchased an NES from June 1,1988 until December 31,

1990. The coupons were good for S5 off any Nintendo purchase. Nintendo had to redeem a minimum of S5 million and a maximum of S25 mi1lion worth of coupons. It also had to pay"'the states S4.75 million in damages -- $3 million to be used for antitrust enforcement and S1.75 mi11ion in legal costs. Nintendo had to agree not to raise the prices of its products during the coupon redemption period and to tell its dealers that they were free to independently set their own prices. Nintendo denied any

8 wrongdoing. stating it was settling simply to avoid lengthy court proceedings. ! The 96 F.T.C. had found that company sales representatives enforced the prices set by the

manufacturer, and retailers who resisted were threatened with a slowdown of

shipments, a reduction in the number of consoles delivered, or a complete lockout. 82

It was the first joint federal-state prosecution of resale price maintenance. The

attorneys general of all 50 states and the District of Columbia sued Nintendo, although the New York and Maryland prosecutors were the ones who arranged the settlement.

The second F.T.C. investigation, looking into allegations that Nintendo had attempted to monopolize the home video game industry by using product designs and licensing policies to lock others out of the business and keep prices artificially high, was concluded in 1992 without any action being taken.

3.10 Legal Treatment of Exclusive Dealing

The lawsuit filed by Atari Corporation against Nintendo alleged that Nintendo had violated antitrust laws by its use of exclusive dealing contracts. The exclusive contracts Nintendo had with its game developers can be thought of as exclusive dealing contracts, even though they were not of the type usually discussed between a manufacturer and a retailer. By agreeing to the contract, the developers agreed not to purchase, and then resell, particular game cassettes from any other manufacturer.

The courts have traditionally been fairly hostile to exclusive dealing contracts.

However, in recent years, they have begun to recognize that exclusive dealing 97 arrangements can serve legitimate purposes and that their anti competitive effects must be balanced against their efficiencies.

The main contention against exclusive dealing has been market foreclosure.

When a manufacturer has captured a set of retailers with exclusive dealing contracts, those retailers are foreclosed from its competitors. Rival manufacturers must either find other retailers or enter the market at both levels. Exclusive dealing can create both barriers to entry and to expansion, which is the reason for the concern that exclusive dealing contracts will impede competition. In order for foreclosure to be of concern however, the number of satisfactory retail outlets must be limited for some reason.

Manufacturers may have an incentive to impose exclusive dealing for a number of reasons. Many of these have to do with product promotion. First of all, the use of exclusive dealing can prevent a manufacturer's rivals from free-riding on its customer drawing power from product quality or advertising. It prevents retailers from switching customers brought in by one manufacturer's advertising or brand name to a competing product on which they make a bigger profit margin.

Second, because the retailer only has one brand to sell, he will tend to promote that product more vigorously. This assures manufacturers that their product will be merchandized with the maximum amount of effort. When a supplier's product needs special selling or servicing techniques, the supplier will be more interested in exclusive dealing than otherwise. 98 Third, exclusive dealing contracts protect the manufacturer's investment in strengthening its dealers through technical and sales training. etc. A manufacturer will be more likely to assist his distributors in their marketing efforts. In this same vein. it protects the trade secrets of the manufacturer. The manufacturer may give distributors a wealth of trade secrets and technical information. and exclusive dealing prevents them from using this information to assist the sales of a rival.

Finally, exclusive dealing arrangements may decrease transactions costs for the manufacturer because of fewer, consistent dealers. The dealers may also experience lower costs because they have a continuous. assured supply of the product.

Exclusive dealing is unique among vertical restraints in that it limits purchasing rather than resale. It may limit interbrand competition, but not intrabrand competition.

It might even increase interbrand competition because retailers may promote the brand more seriously and receive more marketing help from the manufacturer. However. counteracting this is the possibility that fewer retail outlets may sell competing products.

There is evidence that Nintendo had many of these incentives to impose exclusive dealing. With the powerful marketing and customer network it maintained. it may have needed the contracts. First. it promoted its systems vigorously during this time. through Nintendo Power. the Nintendo World Championships, television. the

World of Nintendo displays in retail outlets. etc. It also promoted the individual games in Nintendo Power and through the hotline. Exclusive dealing contracts prevented Sega

(or other competing system manufacturers) from free riding on that advertising. Ifthe 99 game had been available on competing systems, those manufacturers would have been

able to benefit from this promotion.

Second, Nintendo provided valuable technological help to software developers if they asked for it. Licensees had access to Nintendo's marketing department, development, and customer services. A licensee could get advice from game evaluators if they needed it. Licensees also had access to some of Nintendo's trade secrets, and exclusive dealing protected those secrets. That was at the heart of Nintendo's complaint against Atari Games' Tengen unit. It was a licensee that got all the assistance and trade secrets, who then produced their own unlicensed cartridges and attempted to compete with Nintendo and its licensees.

Third, Peter Main indicated that the competition between game developers was fierce:" Since their survival depended on Nintendo and its success, they had a strong incentive to promote both their particular games and the Nintendo system. This benefitted Nintendo by increasing the quality of the games and their promotion.

The main standard for liability has traditionally been the size of the market foreclosed. This came from the concern that the higher the market share under exclusive dealing, the greater the likelihood of harm from foreclosure. A description of the legal history of exclusive dealing will be discussed later. Atari Corp. felt it had a good case against Nintendo's exclusive contracting because of the size of its market share. However, Steuer (1983) claims that the practical effect of foreclosure will be dependent on several things besides the market share of the firm and its dealers. The key to assessing the effects of foreclosure is on whether the manufacturer is able to 100 reach the consumer. The level of distribution at which the contracts are imposed will be a determining factor. If a retailer only carries one brand, his customers must purchase that brand or go to another retailer. On the other hand, if a wholesaler se1\s only one brand, his customers (the retailers) may still have easy access to other brands because several wholesalers generalJy visit each retailer. If customers are loyal to a particular retailer. an exclusive dealing contract with that retailer will also capture the customer.

The type of product is also important to the amount of actual foreclosure that would occur. If the product is a convenience product, consumers will tend to buy it from their regular retailer. However. for products which consumers spend time shopping for. the brand name wilJ be more important than retailer loyalty. Also. alternative methods of distribution are sometimes available. and even new distributors can be established. Therefore. it does not directly fo1\ow that if a manufacturer has exclusive dealing contracts with 5090 of the retailers in an industry he has captured all the pre-exclusive dealing sales of those retailers. Foreclosure should be measured by the extent to which competitive suppliers actually have been foreclosed from reaching consumers. The percentage of dealers captured may not be the best indicator of foreclosure. but it does provide a starting point. The numbers should then be examined in light of consumer loyalty to distributors and to the brand. as well as alternative means of distribution.

Exclusive dealing can be found illegal under Section 3 of the Clayton Act. Section

5 of the F.T.C. Act. or Section 1 of the Sherman Act. The central piece of legislation 101 is the Clayton Act, which finds exclusive dealing illegal if its effect is to "substantially

lessen competition or create a monopoly."84 Exclusive dealing has been subject to the

rule of reason under case law. It has never been per se illegal. A brief summary of the

legal case history of exclusive dealing follows. 8s

In Standard Fashion Co. v. Magrane-Houston Co. (1922), Standard Fashion. a pattern

maker, sued Magrane-Houston, a retail dry goods store. for violating their contract by

selling McCalls patterns. The court found that the 40% of the pattern outlets bound

by the exclusive dealing contracts amounted to a substantial lessening of competition

and tended to create a monopoly under the Clayton Act.

Some of the harshest treatment of exclusive dealing came from the Supreme

Court decisions in Standard Oil v. United States (1949) and F. T.C. v. Brown Shoe Co. (1966).

These cases used the rule of "quantitative substantiality," which predicted illegality

strictly on the percentage of distributors foreclosed. In Standard, the supplier had a

0 23 0 market share and had as exclusive dealers 16% ofthe independent service stations

in the relevant market. In Brown Shoe, the court found that under the F.T.C. Act, an exclusive dealing violation could be found if a significant number of outlets were foreclosed, without any showing of the extent of actual foreclosure in the relevant market as a whole. This case involved exclusive dealing between the second largest shoe manufacturer and 650 retail outlets. The outlets were free to terminate the contact at any time. The 650 retail stores made up only 1% of all shoe stores in the

United States at the time. These cases used no consistent number or percentage to determine when exclusive dealing would result in harmful foreclosure. 102 The results from Tampa Electric Co. v. Nashville Coal Co. (1961) seemed to be that

more economic analysis would be required in exclusive restraints. The court refused

to nullify a contract in which Tampa agreed to purchase its total coal requirements

from Nashville for 20 years. Nashville tried to back out after the price of coal went up.

If Tampa purchased alI of its requirements from Nashville, this in effect required it to

exclusively deal with Nashville in coal. Therefore, the court made the assumption that

the contract involved exclusive dealing for analysis purposes. The case used the rule

of "qualitative substantiality," which put less emphasis on only market share data and gave more weight to the probable effect the contract would have on competition. In other words. the opportunities for other traders to enter into or remain in the market had to be significantly limited. Only showing that the contract involved a significant amount of sales was not enough.

In the Beltone Electronics Corp. (1982) decision, the Federal Trade Commission uphc\d the exclusive dealing contracts that the hearing aid manufacturer, Beltone, had with its distributors, saying that the supplier's interest in protecting its advertising investments provided a legitimate reason for imposing exclusive dealing. It also didn't hurt that the contracts were terminable on 30 days notice and there was no evidence that new entrants were finding it difficult to find distributors. The court held that while market foreclosure is important, there are other things to be considered such as

"reasonable justifications" for the restraint and the duration of the contracts. The availability of less restrictive alternatives is not an issue either. 103

Ornstein (1983) suggests that harm inflicted upon rival firms as a result of exclusive dealing contracts isn't the relevant issue. Consumer welfare is.

Disadvantaging rivals is what competition is all about. If exclusive dealing reduces output, consumers are harmed. In the vein of Bork (1978), exclusive dealing cannot create power; it can only be a means of implementing it if it already exists. If a firm wants to exclusively deal. it must purchase that right in the form of lower prices or a provision of greater efficiencies. If exclusive dealing contracts are to pose a problem, it must be true that a seller who already has the ability to exclude insists upon a system of the contracts. Suppose a seller attempts to monopolize its market by exclusive dealing. If there are no efficiencies, the seller's rival will match whatever inducements it offered. Under these conditions, exclusive dealing cannot be exclusionary.

Congressman Eckart, in his letter to Rill, said, "Software producers have little choice but to seek agreements with Nintendo to the exclusion of other hardware game systems since other systems cannot otTer a viable potential market for their products."b6 Nintendo had the power to exclude because it could provide its developers with greater efficiencies in the form of a stronger brand name, along with customer services, etc. Yamauchi said in defense of these contracts, "If our terms are so strict they make people unhappy, they're always free not to accept them, but so many companies are satisfied because they see the merits of doing business with us. It's always their choice, not ours."B7 104 Although the emphasis is still on market foreclosure, as evidenced by Beltone,

more attention seems to be given to business justifications of exclusive dealing. The law on exclusive dealing is based on a judicial concern for vertical foreclosure of markets and restrictions on freedom of choice. Blair and Kaserman (1985) argue that there is freedom of choice. It just takes place at the time of the contract. This is the point that Yamauchi's quote in the previous paragraph makes. Striking a balance between the efficiency gains of vertical restraints and anti competitive effects forms the core of the public policy debate.

The 1985 Vertical Restraints Guidelines seem to take a more lenient view of vertical restraints than was present in the standing judicial precedents. The Guidelines indicate a three step procedure for determining if exclusive dealing is illegal. First, if the firm has less than 1096 of its market, less than 60% of each market involved is covered by the restraint, and iffirms using the restraint don't have market power, then the contracts are legal. If it fails that test, then the court should look to see if entry is easy in both markets. Failing that. the restraint could still escape if the study can find some way to excuse it. In principle. the greater the market share and the longer the duration of the contract, the more likely that exclusive dealing contracts will be found to be illegal. Recently. the Justice Department rescinded these guidelines. indicating that they "unduly elevate theory at the expense of factual analysis."8s

Ornstein (1983) did a study which examined several industries in which exclusive dealing had been declared illegal and subsequently discontinued. He found little evidence that much had changed in those industries after the exclusive dealing had 105 been stopped. He concluded that these cases did not offer much evidence of monopolization or entry being blocked due to exclusive dealing.

Nintendo would have certainly failed the first test described in the Vertical

Restraints Guidelines and perhaps also the second. Entry into the software market would have been fairly easy, but entry into the console market would have been a little more difficult. There is much evidence that firms felt very intimidated by Nintendo's connections with software developers. There was a general feeling that a new firm wouldn't be able to get good game designers for its system. That might make it neccssary to enter on both levels.

The F.T.C. did not file a suit against them and the jury in the Atari Corporation suit did not find them guilty. Therefore, ease of entry or some business justification must havc been found for the contracts. Either that, or because it discontinued its practices and the market seemed to be getting more and more competitive, the authorities decided just to let well enough alone. It would be difficult to argue the fact that Nintendo's marketing practices brought the industry back to life. Protecting the benefits of that accomplishment for itself and its dealers was perhaps taken into account. 106 3.11 Theoretical Evaluation

In the previous chapter, the question was posed, "Could Nintendo have held on to its monopoly even in the presence of a higher quality console using its exclusive dealing contracts?" Using data from the home video game industry for 1990, we will investigate this question.

FIGURE 3.2 is copy of FIGURE 2.4 from that chapter and is shown on the following page. Remember that). = s2nisln}t so it is the ratio of the effective qualities

of the tvvo systems. 6 = ejes1 ' so it is the ratio of the surplus the highest-valued consumer would receive from each system if it were priced at its marginal cost. In region I. firm 1 would exclusively dea\. In region II, firm 2 would choose to use exclusive dealing contracts. Anywhere else, neither firm would find it profitable to exclusively deal. Thus, to determine if Nintendo could have used exclusive dealing to hold on to its monopoly position. we need to determine the values of ). and 6.

Note that for convenience, the theoretical model was developed using the simplifying assumption that e = {t + 1 and the number of consumers was normalized to 1. Without these assumptions, the demand functions must be multiplied by N/(B-{t). where N is the number of consumers. This does not change any of the results except for the final profit figures, which also need to be multiplied by N/(B-{t). These assumptions do not hold in the home video game industry, so demand functions will be shown with this additional term. 107

10

8

4

o 2 6 8 10 ).

Exclusive Dealing Ranges

FIGURE 3.2

Sega's Genesis has a 16-bit processor while Nintendo's NES has an 8-bit processor. The Genesis is twice as powerful as NES, so 51 is normalized to 1 and 52 =

2. At the end of 1990, there were 300 titles available for NES, while there were only 108 60 titles available for Genesis. 89 However, with a converter that could be purchased for S35, the games available for Sega's 8-bit Master System could be played on Genesis.

There were about 100 games available for that system. 90 The question then becomes whether those games should be included as part of the Genesis system. The Master

System games could not take full advantage of the 16-bit console, so they didn't play like a 16-bit game. From talking to a Sega technical support employee, I got the impression that Master System games are hard to find and that the players who purchased them were those with 8-bit systems. They were simply an added bonus if you already had a Sega system and traded up. I argue that they should not be included because they are not part of the 16-bit system and do not use the increased quality of the console fully. Because it makes quite a difference to the results whether they are included or not, I will first discuss the case where they are not included and then briefly mention the results when the Master System games are included.

The marginal cost of each system, mC j = SjS. + njcz' was a little more difficult to determine. Information from Congressman Eckart's staff indicated that wholesale prices of 8-bit ~ames were around S14 and that Nintendo made a profit of S5 on each cartridge. 91 That would put the cost at S9. Other sources put the wholesale prices betvveen S9 and S14.92 The Atari Corporation Damages Study, which was used to calculate the damages Atari suffered because of Nintendo's practices for their lawsuit, puts the cost of an 8-bit cartridge at roughly S8. 93 The range of wholesale prices of

16-bit cassettes was 512 to 516, a few dollars higher than the 8-bit cartridge range. 94

Since the results are only very slightly affected by this choice. I will report only the 109 results using Cz = S9. S9 was used simply because it is at the upper-end of the cost estimates of an 8-bit cartridge and at the lower-end cost of a 16-bit cartridge.

The cost of an 8-bit console was determined to be roughly 540 in the Damages

Study,95 which is consistent with the claim that many people were making at the time that Nintendo priced its console close to cost to sell as many as possible. If we use the rule of thumb that retailers double the wholesale price, then an item with a manufacturing cost of S40 being sold close to cost is consistent with a S99.95 retail price tag. The cost of a 16-bit console then becomes S80, which is consistent with a retail price of S149 to S189.

Using this information, sln 1 = 120 and S2n2 = 300, making Sega the low quality firm 1 and Nintendo firm 2. Even though Sega had a higher quality console, Nintendo had the high quality system because of the significantly larger number of games available for its system. Therefore,

A ::: 2.5 o ::: 3008 - 2740 . 1208 - 620

With A = 2.5,

.9. = .6083 1 &1 = 1.2869 ~ ::: 1.9426 &2 ::: 4.1101 110

We assume that e 2! 9.1333. In other words, that e is large enough for the highest- valued consumer to purchase a high quality system priced at marginal cost. c5 will be:

Below region I when 9.1333 s e s 10.4089.

In region I when 10.4089 S e S 13.3413.

Between I and II when 13.3413 S e S 20.2175.

In region II when e ~ 20,2175.

Above region II never.

Therefore, Nintendo would choose to exclusively deal, monopolizing the market by driving Sega out, when e ~ 20.2175.

Since the average retail price of a Nintendo game was around S40,96 the total price of the Nintendo system was P2 = 5100 + 300 (S40) or S12,100. The average price of Sega's games was around 555,97 with the console priced near 5190 at this time. Therefore, the total price of Sega's system was PI = 5190 + 60 (S50) or S3,190.

According to the model, the demands for Sega and Nintendo are:

and 111

The 1990 census indicates that there were approximately 65 million households in the

United States in 1990. During that year, Nintendo sold approximately 7.4 million systems98 and Sega sold about 800,000 16-bit systems. 99 Substituting these values into the demand functions and solving for e and fl, we find that e = 261.52.

Therefore, c5 is in region II. The model predicts that Nintendo could have held on to the dominant share of the market, at least for the length of the contracts. The final price of the system would decrease with exclusive dealing only if c5 ~ 2.3460, and this is true only ife ~ 69.5606. Therefore, price would increase with exclusive dealing, and consumers would be worse off. Hence, concentrating on software seems to work in this case. The claim made that Nintendo could keep consumers buying an obsolete system by controlling the developers seems to have some validity.

If the Master System games are included, the scenario is a bit different. Because a converter is required in order to use the games, the console cost is increased to S89 for Sega.lO(J In this case,

Slnl = 300 S2n2 = 320 A = 1.0667 o = 3208 - 1529 3008 - 2740

Right off, we see that the lower quality system has a higher cost. We find if we consider the Master System games as part of the 16-bit system, Sega will always choose 112 to exclusively deal when e ~ 24.9489. Nintendo would never choose to exclusively deal. This doesn't seem to have any descriptive power for the actual events that occurred in the industry.

3.12 Conclusions

At least on the strength of this model, it appears that the court might have been able to find that Nintendo's exclusive dealing contracts were illegal because consumers arc harmed by higher prices. However, this is a static model which only examines the foreclosure effects of exclusive dealing contracts and doesn't explicitly consider the issues of game quality, promotion, and technical investment in the software firms.

Also, the contracts were only for 2 years, which is about how long it was before

Nintendo introduced a 16-bit system of its own. It also ended the questionable practices right after this time period. Even if it had not discontinued its exclusive dealing contracts, Nintendo might not have been able to continue capturing the software developers as time went on. As it happened, Sega did increase its market share as time went on and its product became more popUlar. This is probably due to several factors. Because Nintendo no longer required exclusive dealing, software developers were able to create for both systems. Therefore, Sega was able to introduce more games for its system. Some software firms also wanted to explore the 113

16-bit technology, and therefore had to create for Sega initially. Sega also allowed more violence and adult themes than Nintendo did, which appealed to the older consumers and gave game developers more freedom. With the help of Game Boy and

NES, Nintendo continues to be the leader in the industry today. However, Sega has become a strong competitive force that must be reckoned with. 114 3.13 Endnotes

1. Nintendo Company Ltd. 1993 Annual Report. 1993. p. 9.

2. 1992 Nintendo Media Kit. 1992. p. 63.

3. Susan Moffat. "Can Nintendo Keep Winning?" Fortune. 5 November 1990. p.131.

4. 1992 Nintendo Media Kit. p. 60.

5. 1990 Nintendo Media Kit. p.l.

6. 1992 Nintendo Media Kit. p. 16.

7. Christopher Boehme. "Nintendo Employees: Video Garnes are Their Life." COMPASS Readings (Northwest Airlinesl. December 1990. p. 25.

8. David Scheff. Game Over. How Nintendo Zapped an American Industry. Captured Your Dollars. and Enslaved Your Children (New York: Random House. 1993). p. 61.

9. Ibid.

10. Ibid .. p. 70.

11. Ibid .. p. 140.

12. L. Gordon Crovitz. "Congress Zaps Mario and Luigi With Killer Lawyers." The Wall Street Journal. 17 January 1990. p. A21.

13. 1990 Nintendo Media Kit. p. 51.

14. Scheff. p. 150.

15. Scheff. p. 155.

16. Scheff. p. 172.

17. Scheff. p. 203.

18. 1992 Nintendo Media Kit. p. 2.

19. Nintendo Company Ltd. 1993 Annual Report. p. 15.

20. Ibid .. p. 21. 115 21. Dennis Eckart, Letter to James Rill, 7 December 1989.

22. 1990 Nintendo Media Kit, p. 50.

23. 1992 Nintendo Media Kit, p. 45.

24. Christopher Boehme, p. 25.

25. Cravitz, p. A21.

26. Eckart, p. 3.

27. Scheff. p. 213.

28. Boehme. p. 25.

29. Scheff. p. 215.

30. Boehme. p.25.

31. Scheff. p. 235.

32. Carla Lazzareschi. "No-No Nintendo. Expensive Video-Game Systems May Fall to Rivals. Tight Budgets. 1I Los Angeles Times. 26 November 1991, p. 011.

33. Scheff. p. 178.

34. Scheff. p. 179.

35. Ibid .. p. 180.

36. 1992 Nintendo Media Kit. p. 67.

37. Boehme, p. 25.

38. Nintendo Company Ltd. 1993 Annual Report, p. 16.

39. 1992 Nintendo Media Kit. p. 48-49.

40. Daniel B. Wood, IINintendo's Quest: Staying Popular," The Christian Science Monitor. 3 December 1990, p. 12.

41. Scheff. p. 194.

42. 1990 Nintendo Media Kit, p. 41. 116 43. Ibid., p. 42.

44. joseph Pereira, "For Video Games, Now It's a Battle of Bits," The Wall Street Journal, 9 january 1990, p. B1.

45. Nintendo Company Ltd. 1993 Annual Report, p. 1.

46. Scheff, p. 353.

47. Daniel B. Wood, "SmalJ Firms Challenge Nintendo in New Games," The Christian Science Monitor, 22 january 1991, p. 8.

48. Jonathon Weber, "Nintendo Lowers Price, Will Offer CD Device," Los Angeles Times, 10 January 1992, p. D2.

49. Daniel B. Wood, 'Video-Game Wars Part II: The Battle for Awareness," The Christian Science Monitor, 31 january 1992, p. 8.

50. Pascal Zachary, "Nintendo to Ease Restrictions on U.S. Game Designers," The Wall Street lournal. 22 October 1990, p. B8.

51. Pereira, "Battle of Bits," p. B1.

52. John Burgess, "Making a Mint on Mario," The Washington Post, 15 December 1991, p. H5.

53. Carla Lazzareschi, "Lawsuits Against Nintendo Go to Heart of the Way japan Competes," Los Angeles Times, 19 March 1989, p. 6.

54. Frank E. james, "If You Thought Nintendo Was just a Game, You Lose," The Wall Street lournal. 6 June 1989, p. B1.

55. Nintendo Company Ltd. 1993 Annual Report, p. 1.

56. jim Carlton, "Sega, Aided by Hedgehog, is Gaining on Nintendo," The Wall Street lournal. 5 Novermber 1993, p. B1.

57. Laura Buddine, "Forced to Play KilJer Nintendo's Game," The Wan Street lournal. 9 March 1990, p. A13.

58. Scheff, p. 357.

59. "The Hedgehog Takes Hold," The New York Times, 5 january 1992, p. F12. 117 60. Leslie Helm, "Sega Muscles in on Nintendo," Los Angeles Times, 25 May 1992, p. D2.

61. Joseph Pereira, "Nintendo, Sega Zap Prices as Video-Game War Heats Up," The Wall Street lournal. 5 May 1992, p. B1.

62. Carlton, p. B1.

63. Daniel B. Wood, 'Video-Game Wars Part II: The Battle for Awareness," The Christian Science Monitor, 3~ January 1992, p. 8.

64. 1992 Nintendo Media Kit, p. 9.

65. Wood, "Awareness," p. 8.

66. John Burgess, "Sega's Sonic Boom," The Washington Post, 19 December 1993, p. H4.

67. Scheff, p. 365.

68. Lazzareschi, "No-No Nintendo," p. D1.

69. Carlton, p. B1.

70. Scheff. p. 259.

71. Douglas C. McGill, "A Nintendo Labyrinth Filled With Lawyers, Not Dragons," The New York Times, 9 March 1989, p. D23.

72. Ibid.

73. Denise Gellene, "Atari Games Gets Zapped by Nintendo Countersuit," Los Angeles Times, 6 January 1989, Sect. 4, p. 5.

74. "Nintendo Sets Jury's Sights on Atari in Antitrust Trial," Prentice Hall Law and Business, June 1992.

75. Richard B. Schmitt, "Nintendo Suit Filed by Atari Is Going to TriaL" The Wall Street lournal. 13 February 1992, p. B3.

76. "Nintendo, U.S. Unit Sued: Antitrust Violations Alleged," The Wall Street lournal. 2 February 1989, p. B4.

77. Eckart, p. 5. The remainder of the paragraph summarizes this letter. 118 78. Ibid .. p. 2.

79. Scheff. p. 271.

80. Zachary. p. B1.

81. Paul M. Barrett. "Nintendo's Latest Novelty is a Price-Fixing Settlement," The Wall Street lournal. 11 April 1991. p. B1.

82. Ibid.

83. Paul M. Barrett. "Nintendo-Atari Zapping Contest Goes to Washington." The Wall Street lournal. 8 December 1989. p. B6.

84. 15 U.S.c. Paragraph 14 (1982).

85. This history is taken from Steuer (1983). Ornstein (1989). and Blair and Kaserman (1985).

86. Eckart. p. 6.

87. Karl Schoenberger. "Nintendo to Fund Learning Project at MIT," Los Angeles Times, 6 May 1990. p. D11.

88. William A. Beltz, "Text of Assistant Attorney General Anne K. Bingaman's Address to ABA's Antitrust Section," Antitrust and Trade Regulation Report, 12 August 1993. p. 251.

89. Wood. "Small Firms," p. 8.

90. There are currently approximately 118 Master System games available. Sega stopped making Master System games at the end of 1991. During 1991, they only produced 3 or 4 games. Taking out the titles duplicated in the 16-bit cartridges and subtracting 4, there were approximately 100 games unique to the Master System at the end of 1990.

91. "Nintendo: The Quaking 'Kingdom of Software,''' Entrepreneurial Strategy Case Study from Dennis Eckart's Office. 1989, p. 5.

92. Scheff, p. 215.

93. Michael Flynn. "Atari Corporation Damages. U.S. Home Video Game Industry Revised Damages Study," 1992, p. 34.

94. Scheff, p. 369. 119 95. Flynn, p. 29.

96. Zachary, p. B8.

97. Pereira, "Battle of Bits," p. B1. Prices were given as being between S50 and S60, so the midpoint was used.

98. ''Video-Game Firm to Unveil Its New System This Year," The Wall Street Journal, 8 January 1991, p. B7.

99. Michael Lev, "As the Craze Cools, a Youth-Based Industry Matures," The New York Times, 29 April 1990, p. FlO indicates sales of 400,000 for Sega during 1989. Sega indicates that its sales doubled in 1990 in its 1994 Media Kit.

100. Since the retail price of the converter was S35, the rule of thumb that the price is doubled at each level of distribution is used to come up with a cost of S9. 120 APPENDIX A: SAMPLE EXPERIMENTAL INSTRUCTIONS

General Instructions

This is an experiment in the economics of market decision making. Various research foundations have provided funds for this research. The instructions are fairly simple. and if you follow them carefully and make wise decisions. you may earn a considerable amount of money which will be PAID TO YOU IN CASH. You will be paid in private by the experimenter at the end of the experiment.

vVe are going to create a market in which there will be two types of agents and a set of simulated buyers. There will be one PRODUCER and three TRADERS in this market. The PRODUCER has the ability to produce units of a fictitious commodity which the simulated buyers value. However. the PRODUCER cannot sell these units directly to the buyers. He can only sell ~o the TRADERS. A TRADER cannot produce units of the commodity. but he can sell units to the buyers. Therefore if a TRADER wishes to sell a unit to the buyers. he must first purchase it from the PRODUCER. Each time an agent sells a unit. the experimenter will pay the agent a 4 peso commission.

Each of you has been given a schedule called the Buyer Demand Schedule. This schedule indicates how the simulated buyers will purchase units from the TRADERS.

At each unit price level. the "Number of Units Demanded" indicates the number of units the buyers want to purchase at that price. For instance. at a unit price of 495 pesos. 121 all the buyers together will purchase 22 units. At a unit price of 460 pesos. the buyers will purchase 29 units. Remember. the three TRADERS face these buyers together.

In your folder. you will find with your Buyer Demand Schedule. a set of Record

Sheets, several Offer Forms. and several Purchase Forms (if you are a TRADER). Each of these is labelled in the upper left-hand corner with your agent type (and number if you are a TRADER) for this experiment. From this point on. all references to money will be made in terms of experimental "pesos." Your exchange rate between pesos and

U.S. dollars is written in the upper right-hand corner of each of your Record Sheets.

Note that you will maximize your cash payment at the end of the experiment by earning as many pesos as you can.

Instructions Specific to Traders

You will be a TRADER in this experiment. Remember. you cannot produce any units. The only way you can obtain unit.:s to sell to the buyers is to purchase them from the PRODUCER. The PRODUCER will have no direct contact with the buyers. His only contact will be with the TRADERS.

On the top of each of your Record Sheets. you wi11 notice that there are two items. Your "Unit Selling Cost" is the cost you must pay each time you sell a unit to the buyers. The Unit Selling Cost is the same for every unit you sell. Your "Trading

Limit" indicates the maximum number of units you can purchase from the PRODUCER and re-sell to the buyers each period. 122 Each time you sell a unit to the buyers. you will receive the unit price you charge the buyers (called the Unit Selling Price) plus a 4 peso commission less your Unit

Selling Cost. Each period you will earn (Unit Selling Price + 4 pesos - Unit Selling

Cost) x number of units sold. However. from this amount you must pay the PRODUCER for the units purchased from him: So, your NET EARNINGS each period will be calculated as:

NET = Selling + 4 pesos - Selling) x Units Purchase x Units EARNINGS Price Cost Sold Price Purchased

Therefore, you can make money in this experiment by buying and selling units such that (Unit Selling Price + 4 pesos - Unit Selling Cost) is greater than the price you pay to purchase the unit. If you purchase a unit and do not re-sell it to the buyers, you earn no money, but you still have to pay the Unit Purchase Price. Therefore. you decrease your NET EARNINGS by purchasing units that you do not sell. You cannot carry units over from one period to the next.

Let's look at an example. Suppose that during one period you were able to purchase 3 units from the PRODUCER for 250 pesos. and you were able to sell those

3 units to the buyers for 400 pesos each. if your Unit Selling Cost were 50 pesos, your NET EARNINGS for that period would be [400 + 4 - 50] x 3 - 1250 x 3] or 312 pesos. The amount of U.S. currency you would be paid is determined by the exchange rate listed on your Record Sheets. If your exchange rate were such that 3 pesos equals

1 cent, then your NET EARNINGS for that period would be 312 pesos / 3 or 104 cents. 123 Market Organization

The market for this commodity will be organized in a series of trading periods.

Each of these trading periods will proceed as follows: First, the PRODUCER will set a

unit price and indicate the maximum number of units he is willing to sell to the

TRADERS at that price. Next, each of the TRADERS will set a unit price and ~ndicate the maximum number of units he would be willing to selJ to the buyers at that price.

Then the buyers wilJ place orders with the TRADERS, and the TRADERS will purchase the necessary units from the PRODUCER to filJ those orders.

1t At the beginning of the period, the PRODUCER wilJ make an offer. This means he will set a unit price and indicate the number of units he wishes to make available to the TRADERS at that price. The ofTer will be submitted on an OfTer Form and will be collected by the experimenter. The PRODUCER will have two minutes to submit an ofTer each period. There will be two restrictions placed on the unit price set by the

PRODUCER. First, he will not be allowed to set a unit price below his Unit Production

Cost. If the Unit Production Cost is 50 pesos, for example, an admissible unit price would be 50 pesos or above. Second, the unit price can only be set in 5 peso increments. In other words, a unit price of 250 or 255 pesos is admissible, while a unit price of 252 or 254 pesos is not. If an inadmissible offer is made, the PRODUCER will be asked to revise it. After the PRODUCER has submitted his offer to the experimenter, the unit price he set will be written on the board for the TRADERS. At this time, the

TRADERS will be randomly ordered and this order will be written on the board. The experimenter will discuss the process for doing this before the experiment begins. 124 The TRADERS now know what it will cost them to buy each unit from the

PRODUCER. At this time, each TRADER will submit an offer to the buyers. This offer will consist of a unit price and the maximum number of units he would be willing to

sell to the buyers at that price, if he can obtain the units from the PRODUCER. Again, the unit price set must be in 5 peso increments, and any TRADER not complying with this will be asked to revise his offer. The offer will be written on an Offer Form and will be collected by the experimenter. The TRADERS will have two minutes to make their offers.

Since the TRADERS have not purchased any units at this point, the simulated buyers will place "orders" with the TRADERS, indicating the number of units they wish to buy from each TRADER. if the TRADER can obtain them from the PRODUCER. The buyers arc simulated using a computer program, and together they behave like the

Buyer Demand Schedule indicates. There are ten buyers who will be ordered randomly by the computer. Any buying order for them is equally likely. They are programmed to place their orders in the following manner. Each buyer will always order from the

TRADER with the lowest price who still has units left. The buyer first in line will begin ordering from the TRADER with the lowest unit price. If that TRADER has not offered enough units to satisfy the buyer's demand, the buyer will order alI the units that

TRADER has offered and then move on to the TRADER with the lowest price who still has units h~ft. The next buyer will do the same, and so on. If more than one TRADER with units left has offered the lowest price, the buyer will randomly choose which

TRADER to order from. Each of these TRADERS has an equal chance of being chosen 125 by the buyer. The TRADERS will each be informed of the number of units ordered from him after the buyers are through. A TRADER is under no obligation to fill any of his buyers' orders. The orders indicate how many units the buyers want to purchase from each of the TRADERS. However, a TRADER does not have to sell all of these units if it either cannot or does not wish to.

Once each of the TRADERS knows how many units he can sell to the buyers, the

TRADERS will be able to make purchases from the PRODUCER. Each TRADER will indicate the number of units he wishes to purchase from the PRODUCER on a Purchase

Form. These will be collected by the experimenter. If the PRODUCER has offered enough units to satisfy all the purchase requests, each TRADER will be able to purchase the number of units indicated on his Purchase Form. However, if the PRODUCER has not offered enough units to supply all the TRADERS, then the TRADERS' purchases will be processed in the order previously written on the board. The TRADER who is first in line will be able to purchase his units first, etc. Each of the TRADERS will then be notified of the number of units they were able to purchase.

Since the TRADERS do not value units of the commodity, it will be assumed that any unit purchased from the PRODUCER will be sold to the buyers if possible.

Therefore, the number of units each TRADER purchases from the PRODUCER is the number of units he selIs to the buyers that period, unless the TRADER purchases more units than were ordered from him. If this is the case, that TRADER will sell only the number of units ordered from him. Note that a TRADER decreases his NET EARNINGS by purchasing more units than he can sell. 126 Once the TRADERS have been notified of how many units they were able to purchase. the PRODUCER will be informed of the number of units he sold. the price otTers of the TRADERS wiII be written on the board. and you wiII be given a chance to compute your NET EARNINGS.

There is a short worksheet attached to the back of your instructions. Please work through it and ask any questions you might have as you go along. Once everyone is through with the instructions and worksheet. we wiII go through two practice periods before the actual experiment starts. You will not be paid for these periods. but it will be to your advantage to participate. Are there any questions?

NOTE:

1. Remember. a 4 peso commission will be paid to both the PRODUCER and the

TRADERS each time either one sells a unit.

2. It is important that the decisions that you make be yours and yours alone.

Therefore. we ask that you do not talk to other subjects during the experiment and that you do not look at their Record Sheets. 127 APPENDIX B: EXPERIMENTAL RESULTS

MC101 Upstream

Price

560- ......

510 ......

460 ...... -0- ICtUlt p<1C8 + Nllh predlclIon

360 . 0 • 0 ••••••• 0 • 0 •••••••• 0 ••••

1 2 3 4 5 6 7 B 9 10 11 12 13 14 15 Period

MC101 Downstream

Price

560 . 0 ••••••• 0 •• 0 0 ••• 0 ••• 0

510 o~ 0 • ~o or... 0 • 0 0 0..0-

~ "'uader 1 o trader 2 460 ~trad.r 3 '* compelillve price 410- + Nash predlclton

360 ...... 0 ••••• 0 0 0 • 0 • 0 0 •• 0

310~-r~--r-+--r-+~~+-~-r~--r-+-~ 1 2 3 4 5 6 7 B 9 10 11 12 13 14 15 Period 128

MC102 Upstream

Price

560 ------

510 ------

460 ------0- aelull price + Nllh predlellon

410 ------

360 ------

2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MC102 Downstream

Price

560 ------.. - - . - - - - .

....U.der 1 Otreder 2

460 - - - - - ... - . - - - - - ... - . - . - - - - -H- trader :I "* eompellilve price 410 - - - - - ..... ------.. ------. + Nash predlcllon

360 - - - ..... - . - - . - - ... - . ------

310~-r-+~--r-+-~-r-+~--7-+--r-+~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period 129

Me 103 Upstream

Price

560

510

<)- .C1UII ptic. + Nash predlciion

360

2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MC103 Downstream

Price

..... t'ader 1 460 ...... O"ader 2 ~"ader 3 *' competlllve price 410 ...... + Nash predlcllon

360 ......

2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period 130

MC104 Upstream

Price

560 ......

510 ......

-0 ICIUII PliCt 460 + Nun pl.docllon

360 ......

2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MC104 Downstream

Price

-trader 1 <> ,,"del 2 460 ...... ~Irad.' 3 "* cempelilive price 410 ...... +Nash predicllon

360 ......

310~~-+~--+-4-~-+~--+-4-~-+-4~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period 131

MM101 Upstream

Price

560 ------

510 ------

-0- ICIUII ptlel + N.. h prediciion 460 ------

360 ------.

2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MM101 Downstream

Price

460 ., - ...... --- . --- ..... - . -0- Ictuel price '* Dest r•• ponse 410 -- ...... - ... - ... - . - . - .. - .. + Nash prediction

360 -- ...... - ... - ... - ......

310~-r-+-+~--r-+-~-r-+~--r-r-+-~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period 132

MM102 Upstream

Price

560 - - - . - . - ... - .. - . - - ... - .... - .

510 ...... - - ...... - ... .

-0- lelUl1 ptlcl + N.Sh predleUon 460 ...... - ......

310~-r-+-+~--r-+-~-r-+-+~--~~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MM102 Downstream

Price

510 . - - .. - . - ...... - ......

460 ...... - - ...... -0 aCluel pllce "* be., r.sponse 410 ...... - ...... + N.. ~ pledle"on

360

2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period 133

MM103 Upstream

Price

560 ------.. - - - . - - - .... - .

510 - - . - . ------.... - . - - - .... -- -0 ICIUII puc. + N.s~ predlelion 460 - - . - . ------.... --- ...... --

360 _ .. ------. - ... - . - . - - - ... ---

3104-~-+-+~r-r-+-~-r-+-4--r-r-+-~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MM103 Downstream

Price

510 . - - .. --- ... - ... - - . - . - .... --

460 -- .... - - - - . - ... - .. - -0- lelull p<1eI "* beSl r ••pon .. 410 -- ... ------.... - . - + N.. ~ predlellon

360 - .... - - - - - . - .... - . - ......

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period 134

MM104 Upstream

Price

560 ------

510 ------.g. Ie lUll ptlcl + NISh p,odleUon 460

360 ------

310;-~-+~~r-+-~-r-+-+~--+-~~~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MM104 Downstream

Price

460 - - - - . - - - - . - - ... ------. - - . - - -0- lelull ptlca "* De,1 ,upon,. 410 - - ... ------. - - - . ------. - - + N.,n p,odlellon

360 . - - ... ------. - - - - - . - - - - - . - -

3104-~-+~~~4-~-+~r-+-~~-+~~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period 135

MM105 Upstream

Price

560 ------

510 ------. - . - - -

~ .CIUI' PfICI + Nash predlclion 460 ------. - - - - - . - . ------.. - . -

360

310+-~-r-+~--~+-~~-+-+~r-r-+-~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MM105 Dc., ,stream

Price

-ooC1UJI ",Ice

*t>e51 r •• pon •• 410 - - . - .. ------... ------..... - + Nosh predlclion

360 . - ... - .... -- ... --- ...... --

123456789101112131415 Period 136

MM106 Upstream

Price

560 ......

510 ...... -o.cIUII puc. + Nuh prediction

2 3 4 5 6 7 B 9 10 11 12 13 14 15 Period

MM106 Downstream

Price

510 ......

460 ...... g. IctU.t Pftcl * De" r.spon •• 410 ...... + Nllh prediction

360 ......

1 2 3 4 5 6 7 B 9 10 11 12 13 14 15 Period 137

MM107 Upstream

Price

560

510

-o.ChJII prICe 460 + Nasn predlCllon

360

3104-~-+-+~--~+-4-~-+-4~~+-+-~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period

MM107 Downstream

Price

~ .ctua' price *' best response + Nun pr.dlcllon 410 - - - - . ------. ------. - -

360 -- ... - . - - .. - . - .... - . - . - ... -

3104-~-+-+~~+-+-4-~-+-+~~+-+-~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Period 138 APPENDIX C: REGRESSION RESULTS

TABLE C.1

Variable Coefficient Standard Error

0 1 264.567 35.236 O2 283.200 34.840 0 3 257.762 35.293 0 4 272.798 34.868 Os 278.369 32.043 0 6 264.934 35.633 0 7 243.465 34.868 PL 0.707 0.087

TABLE C.2

Variable Coefficient Standard Error

0 1 24.546 4.012

O2 17.727 4.012 03 14.546 4.012

0 4 14.091 4.012 05 61.364 4.012 0 6 5.000 4.012 0 7 22.273 4.012

0 41 -20.796 7.768

042 3.523 7.768 0 43 -13.296 7.768

044 5.909 7.768 0 45 -36.364 7.768 046 2.500 7.768 0 47 22.727 7.768 139

TABLE C.3

Regression Results: I PUt - 410 I = a,D'it + ... + a 7D7it + {3 I POit-, - POBRit-, I + fit

Variable Coefficient Standard Error

D, 14.353 5.336 D2 18.701 5.316 D3 14.364 4.956 D4 23.381 5.101 D5 55.069 8.200 D6 21.204 4.786 D, 36.065 5.892 I POt., - POBRt -' I 0.175 0.122

TABLE C.4

Regression Results: Pcit = a,D 1it + ... + a 7D7it + {3P Oit -, + fit

Variable Coefficient Standard Error

D, 309.885 50.334 D2 303.267 51.643 D3 312.021 49.537 D4 309.332 50.845 Ds 273.800 48.739 D6 320.144 50.590 D7 309.767 48.197 POt-I 0.159 0.091 140

TABLE C.5

Variable Coefficient Standard Error

0, 389.795 18.748 O2 384.250 20.868 03 390.137 17.958 0 4 389.235 20.058 0 5 349.622 21.684 0 6 400.377 18.639 0 7 386.442 16.618 (PDl" - PLl.,) 0.044 0.113 141 REFERENCES

Barrett. Paul M. "Nintendo-Atari Zapping Contest C:)C!5 to Washington." The Wall Street lournal. 8 December 1989. pp. B1, B6.

Barrett. Paul M. "Nintendo's Latest Novelty is a Price-Fixing Settlement." The Wall Street Journal. 11 Apri11991, pp. B1, B6.

Beltz. William A. "Text of Assistant Attorney General Anne K. Bingaman's Address to ABA's Antitrust Section." Antitrust and Trade Regulation Report, 12 August 1993. pp. 250-52.

Besanko. David and Martin K. Perry. "Equilibrium Incentives for Exclusive Dealing in a Differentiated Products Oligopoly." RAND lournal of Economics, 4, Winter 1993. pp. 646-67.

Blair. Roger D. and David L. Kaserman. Antitrust Economics. lIIinois: Richard D. Irwin. Inc.. 1985.

Boehme. Christopher. "Nintendo Employees: Video Garnes are Their Life." COMPASS Readings (Northwest Airlines). December 1990. pp. 22-30.

Boric Robert. The Antitrust Paradox. New York: Basic Books. 1978.

Buddine. Laura. "Forced to Play Killer Nintendo's Game." The Wall Street Journal. 9 March 1990. p. A13.

Burgess. john. "Making a Mint on Mario." The Washington Post. 15 December 1991. pp. H1. H5.

Burgess. john. "Sega's Sonic Boom." The Washington Post. 19 December 1993. pp. H1. H4.

Carlton. jim. "Sega. Aided by Hedgehog. is Gaining on Nintendo." The Wall Street lournal. 5 November 1993. pp. B1, B14.

Comanor. William S. and H.E. Frech. "The Competitive Effects of Vertical Agreements." American Economic Review. 75. june 1985. pp. 539-46.

Crovitz. L. Gordon. "Congress Zaps Mario and Luigi With Killer Lawyers." The Wall Street lourna\, 17 january 1990. p. A2l. 142 Eckart. Dennis E. Letter to james Rill. 7 December 1989.

Flynn. Michael. "Atari Corporation Damages. U.S. Home Video Game Industry Revised Damages Study." 1992.

Fouraker. L.E. and S. Siegel. Bargaining Behavior. New York: McGraw-Hili. 1963.

Gellene. Denise. "Atari Games Gets Zapped by Nintendo Countersuit." The Los Angeles Times. 6 january 1989. sect. 4. p. 2.

Goodfellow. jessica and Charles R. Plott. "An Experimental Examination of the Simultaneous Determination of Input Prices and Output Prices." Southern Economic lournal, 56. 1990. pp. 969-983.

"The Hedgehog Takes Hold." The New York Times. 5 january 1992. p. F12.

Helm. Leslie. "Sega Muscles in on Nintendo." The Los Angeles Times. 25 May 1992. pp. D1. D2.

Hoffman. E. and M.L. Spitzer. "Entitlements. Rights. and Fairness: An Experimental Examination of Subjects' Concepts of Distributive justice." lournal of Legal Studies. 14. june 1985. pp. 259-297. james. Frank E. "If You Thought Nintendo Was just a Game. You Lose." The Wall Street lournal, 6 june 1989. p. B1.

Lazzareschi. Carla. "Lawsuits Against Nintendo Go to the Heart of the Way japan Competes." The Los Angeles Times. 19 March 1989. Sect. 4. pp. 1. 6.

Lazzareschi. Carla. "No-No Nintendo. Expensive Video-Game Systems May Fall to Rivals. Tight Budgets." The Los Angeles Times. 26 November 1991. pp. D1. D11.

Lev. Michael. "As the Craze Cools. a Youth-Based Industry Matures." The New York Times. 29 April 1990. p. FlO.

Mar.'el, Howard. "Exclusive Dealing." !ournal of Law and Economics. 25. April 1982. pp. 1-25.

Mathewson. G. Frank and Ralph A. Winter. "The Competitive Effects of Vertical Agreements: Comment." American Economic Review. 77. December 1987. pp. 1057-62. 143

McGill, Douglas C. "A Nintendo Labyrinth Filled With Lawyers, Not Dragons." The New York Times. 9 March 1989, pp. A1, D23.

Moffat. Susan. "Can Nintendo Keep Winning?" Fortune, 5 November, 1990, pp. 131-32. 136.

Nintendo Company Ltd. 1993 Annual Report. 1993.

"Nintendo: The Quaking 'Kingdom of Software.'" Entrepreneurial Strategy Case Study from Dennis Eckart's Office, 1989.

"Nintendo Sets Jury's Sights on Atari in Antitrust TriaL" Prentice Hall Law and Business. June 1992.

"Nintendo, U.S. Unit Sued: Antitrust Violations Alleged." The Wall Street Journal. 2 February 1989. p. B4.

Ornstein. Stanley I. "Exclusive Dealing and Antitrust." Antitrust Bulletin. 34. 1989, pp. 65-98.

Pereira. joseph. "For Video Games, Now It's a Battle of Bits." The Wall Street !ourna!. 9 january 1990. pp. B1, B6.

Pereira, joseph. "Nintendo. Sega Zap Prices as Video-Game War Heats Up." The Wall Street !ournal. 5 May 1992. pp. B1, B8.

Plott. C.R. and J.T. Uhl. "Competitive Equilibrium with Middlemen: An Empirical Study." Southern Economic !ourna\' 47(4). April 1981. pp. 1063-1071.

Schmitt. Richard B. "Nintendo Suit Files by Atari is Going to TriaL" The WaH Street !ournal. 13 February 1992. pp. B1, B3.

Schwartz. Marius. "The Competitive Effects of Vertical Agreements: Comment." American Economic Review, 77, December 1987, pp. 1063-68.

Sheff. David. Game Over. How Nintendo Zapped an American Industry. Captured Your Dollars. and Enslaved Your Children. New York: Random House. 1993.

Smith. V.L. "An Empirical Study of Decentralized Institutions of Monopoly Restraint." Essays in Contemporary Fields of Economics. Ed. G. Horwich and J. Quirk. West Lafayette: Purdue University Press, 1981. pp. 83-106. 144 Spengler. J. 'Vertical Integration and Anti-Trust Policy." Journal of Political Economy. August 1950. pp. 347-352.

Steuer. Richard M. "Exclusive Dealing in Distribution." Cornell Law Review. 69. 1983. pp.101-134.

Tirole. J. The Theory of Industrial Organization. Cambridge. Mass: The MIT Press. 1988.

'Video-Game Firm to Unveil Its New System This Year." The Wall Street lournal, 8 January 1991, p. B7.

Weber. Jonathan. "Nintendo Lowers Price. Will Offer CD Device." The Los Angeles Times. 10 January 1992. p. D2.

Wood. Daniel B. liNintendo's Quest: Staying Popular." The Christian Science Monitor. 3 December 1990. pp. 12-13.

Wood. Daniel B. "Small Firms Challenge Nintendo in New Games." The Christian Science Monitor. 22 January 1991. p. 8.

Wood. DanicJ B. 'Video-Game Wars Pan II: The Battle for Awareness." The Christian Science Monitor. 31 January 1992. p. 8.

Zachary. G. Pascal. "Nintendo to Ease Restrictions on U.S. Game Designers." The Wall Street lournal, 22 October 1990. pp. B1. B8.

1990 Nintendo Media Kit. 1990.

1992 Nintendo Media Kit. 1992.

1994 Sega Media Kit. 1994.