UCLA Law Review

April, 1990

37 UCLA L. Rev. 693-731


The Leasing Monopolist

John Shepard Wiley Jr. * and Eric Rasmusen ** and J. Mark Ramseyer ***

* Professor of Law, University of California, Los Angeles; A.B. 1975, University of California, Davis; M.A. 1980, University of California, Berkeley; J.D. 1980, University of California, Berkeley.

** Assistant Professor of Business Economics, Anderson Graduate School of Management, University of California, Los Angeles; B.A. 1980, M.A. 1980, Yale University; Ph.D. 1984, Massachusetts Institute of Technology.

*** Professor of Law, University of California, Los Angeles; B.A. 1976, Goshen College; A.M. 1978, University of Michigan; J.D. 1982, Harvard University.

We thank Stephen Breyer, William Comanor, Frank Easterbrook, Franklin Fisher, William Klein, Thomas Krattenmaker, and Stanley Ornstein for helpful comments. Copyright 1990 Wiley, Rasmusen, and Ramseyer.

ABSTRACT

The United Shoe case banned lease-only policies by monopolists. But the court erred in believing that monopoly pricing could explain United Shoe Machinery's complex of leasing policies. At best, this explanation only accounts for a few details of the case. The bulk of the company's conduct seems simply efficient -- suggesting that the Shoe decision was wrong and its later precedential consequences pernicious. The Coase Conjecture might seem to make some sense of Shoe's ban on monopoly leasing, and suggests that the Shoe rule may have been too narrow. At the same time, however, the Conjecture dictates that the Shoe rule be confined ways the opinion did not suggest that are unacceptably costly to accomplish. Courts would do well to accept that Coase describes a problem that is real. But they would also do well to accept it as a problem not worth solving.

INTRODUCTION

Antitrust law regulates the behavior of monopolists by branding some acts as "bad conduct." To define this key notion, every antitrust casebook includes the United Shoe Machinery decision, a classic district court opinion that condemned a monopoly firm's practice of only leasing and never selling its shoe manufacturing equipment. n1 The Supreme Court has never questioned its approval of the case, and Shoe has provided the authority for later and successful government attacks on leasing policies in the computer and copier markets. n2 Nonetheless, because the Shoe court never correctly  [*694]  evaluated the economic impact of leasing, the decision remains highly unsatisfactory. Although the court condemned leasing because of its exclusionary effects, most leasing arrangements are no more exclusionary than sales.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n1 United States v. United Shoe Mach. Corp., 110 F. Supp. 295 (D. Mass. 1953), aff'd, 347 U.S. 521 (1954).

n2 For example, in 1956 IBM complied with a Justice Department consent decree requiring it to offer its computers for sale. Nonetheless, most consumers continued to rent. See J. SOMA, THE COMPUTER INDUSTRY 35, 61-62 (1976). The Department of Justice again attacked IBM's rental policies in a suit that the Reagan administration eventually dismissed. See F. FISHER, J. McGOWAN & J. GREENWOOD, FOLDED, SPINDLED, AND MUTILATED: ECONOMIC ANALYSIS AND U.S. v. IBM 191-96 (1983). In 1975, the Federal Trade Commission entered a consent decree with Xerox that declared that the firm had followed "a lease only policy" that was among its violations of federal antitrust law. In re Xerox Corp., 86 F.T.C. 364, 367-68 (1975) Cf. United States v. Am. Can Co., 1950-1 Trade Cas. (CCH) P62,679, at p. 63, 963, 63,968 (1950) (final judgment finding antitrust violation expresses policy favoring customer ownership of defendant's products "as compared with the continued leasing of them").
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

If the defendant United Shoe Machinery Corporation did not lease in order to exclude its competitors, the question remains why it did insist that its customers lease. Perhaps leasing was incidental to particular United practices that were in fact exclusionary, that bore no necessary relationship to leasing, and that caused the Shoe court to err by condemning leasing generally. Or perhaps both United and its customers preferred leases and United's various other practices for benign reasons, and antitrust law ought to discard Shoe altogether. But an idea of Ronald Coase, from an article written some twenty years after the Shoe decision, suggests that leases may enable monopolists to earn monopoly returns where sales would not. Although the Shoe court had no chance to consider Coase's later insight, prominent commentators have noted the connection between that case and the "Coase Conjecture" -- a theory that explains why leasing can be attractive to a monopolist that makes goods which are durable. n3
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n3 Coase, Durability and Monopoly, 15 J.L. & ECON. 143 (1972). We acquiesce in the conventional label for this Coase insight -- which he expressed entirely in English -- without endorsing the condescension that mathematical economists might imply by their use of the label.

For reasons different from Coase's, the durability of a good may make any competition that does exist more intense than otherwise. See Carlton & Gertner, Market Power and Mergers in Durable-Goods Industries, 32 J.L. & Econ. S203 (1989).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Coase's logic is brilliantly counterintuitive. Usually a firm's products are complements of each other because greater output enhances the value of production by creating a reputation in customers' minds. But a monopolist that makes long-lived goods -- like diamonds, paintings, or aluminum ingots -- faces an unusual situation. Customers can accelerate or delay their purchases in anticipation of changing prices, but when customers do buy, they leave the market until their good wears out. As a result, each future unit of output potentially is a substitute for every present unit.

The fact that some customers value the good more highly than others tempts the monopolist to price-discriminate through time: to sell at a high price to willing buyers today, and then to discount in the future for customers further down the demand curve who were  [*695]  unwilling to pay the initial price. The initial high-valuation customers (the "Highs") would be willing to pay a high price rather than not obtain the good at all. But they would not be willing to pay the high price if they foresaw a price drop. Hence, the monopolist cannot succeed in having a high price today and a low price tomorrow.

Unfortunately for the monopolist, while it cannot price-discriminate, neither can it credibly claim that it will maintain its high price forever. After the Highs have made their purchases, a rational monopolist has no reason not to try to sell to the low-valuation customers (the "Lows") left in the market. As the Highs have already bought, the monopolist may as well lower its price and sell to the Lows. But foreseeing that price drop, the Highs will not buy in the first place. The only outcome that does not create a logical contradiction is for the price to be low today and low tomorrow. The Coase Conjecture is a paradox: the durability of the monopolist's goods denies it its monopoly profit.

Coase also showed that the monopolist can escape this unprofitable paradox by leasing instead of selling. By selling the good at a high price, the monopolist removes the high-valuation customers from the market and encounters the temptation to lower the price. But when the monopolist leases the good at a high rental charge, the high-valuation customers remain in the market, and the monopolist has no incentive to lower the future rental charge. Leasing enables the monopolist essentially to switch from selling a durable good to selling the nondurable service flow that a durable good produces. By thus retaining an incentive to maintain high rental charges, the monopolist can convince customers willing to pay those charges to do so immediately -- because they have no reason to expect a future drop in charges -- and can thereby earn a supracompetitive profit. n4
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n4 Coase also explained that a monopolist can take two other steps to gain its monopoly profit: make the product less durable or offer to repurchase the machines at a set price. Coase, supra note 3, at 149.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

In the last ten years, the Coase Conjecture has become an important part of industrial organization theory, n5 and prominent economists have continued to produce increasingly sophisticated elaborations on it. Faruk Gul, Hugo Sonnenschein, and Robert Wilson made Coase's intuitive discussion precise with mathematical  [*696]  game theory. n6 Jeremy Bulow constructed a simpler but classic two-period model of a durable-goods monopoly and explored how durable a monopolist would choose to make the product it sells. n7 Eric Bond and Larry Samuelson discussed what happens if buyers eventually replace the "durable" goods. n8 Sam Bucovetsky, John Chilton, and Valerie Suslow asked how Coase's logic affected the way a monopolist would choose to exclude potential competitors. n9 Nancy Stokey explored what happens if a seller cannot change prices quickly. n10 Faruk Gul n11 and Lawrence Ausubel and Raymond Deneckere n12 extended the Coase Conjecture to oligopolistic durable-goods sellers. Charles Kahn asked what would happen if marginal costs increased with output. n13
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n5 Note, for example, the extensive discussion of the Conjecture in a recent text, J. TIROLE, THE THEORY OF INDUSTRIAL ORGANIZATION 72-74, 79-87, 91-92 (1988).

n6 Gul, Sonnenschein & Wilson, Foundations of Dynamic Monopoly and the Coase Conjecture, 39 J. ECON. THEORY 155 (1986).

n7 Answer: Not very. Bulow, Durable-Goods Monopolists, 90 J. POL. ECON. 314 (1982) [hereinafter Bulow, Monopolists]; Bulow, An Economic Theory of Planned Obsolescence, 101 Q.J. ECON. 729, 733 n.10 (1986) [hereinafter Bulow, Planned Obsolescence]. For a numerical example of the Bulow model, see E. RASMUSEN, GAMES AND INFORMATION 276-80 (1989). See also Rust, When is it Optimal to Kill off the Market for Used Durable Goods?, 54 ECONOMETRICA 65 (1986) (when consumers can choose when to scrap the product, the monopolist may make it of either socially excessive or socially deficient durability).

n8 Answer: The monopolist may be able to sell the product at a monopoly price after all. Bond & Samuelson, Durable Good Monopolies with Rational Expectations and Replacement Sales, 15 RAND J. ECON. 336 (1984). K. Sridhar Moorthy found the same result follows if the monopolist is able to trick consumers into believing that its actually high rate of future production instead will be low. Moorthy, Consumer Expectations and the Pricing of Durables, in ISSUES IN PRICING 99 (T. Devinney ed. 1988).

n9 Answer: By selling rather than by leasing. See Bucovetsky & Chilton, Concurrent Renting and Selling in a Durable-Goods Monopoly Under Threat of Entry, 17 RAND J. ECON. 261 (1986); Suslow, Commitment and Monopoly Pricing in Durable Goods Models, 4 INT'L J. INDUS. ORG. 451 (1986). When a potential competitor considers challenging a monopolist, the competitor first asks whether the monopolist would be willing to wage a price war. Because a leasing monopolist owns the durable goods in use, a price war would reduce the capital value of the monopolist's own assets -- hence, the lessor-monopolist will hesitate to cut prices. By contrast, the seller-monopolist will cut prices more willingly, as any decline in the value of the existing assets falls on consumers who have already bought the assets. The same point appears in Bulow, Planned Obsolescence, supra note 7, at 743-46.

n10 Answer: The monopolist may be able to charge monopoly prices to some buyers. Stokey, Rational Expectations and Durable Goods Pricing, 12 BELL. J. ECON. 112 (1981).

n11 Gul, Noncooperative Collusion in Durable Goods Oligopoly, 18 RAND J. ECON. 248 (1987). Gul's model yields a curious result: "For the durable goods model, monopoly is more competitive than oligopoly: higher total profits and prices can be sustained under oligopoly than under monopoly." Id. at 249.

n12 Ausubel & Deneckere, One Is Almost Enough for Monopoly, 18 RAND J. ECON. 255 (1987). This model yields results similar to those in Gul, supra note 11.

n13 Answer: The Coase Conjecture is less applicable. Kahn, Durable Goods Monopolist and Consistency with Increasing Costs, 54 ECONOMETRICA 275 (1986).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

 [*697]  As hostile as Chicago economists usually are toward efforts by judges to intervene in the market, Ronald Coase has apparently invented a widely accepted theory that requires just such intervention. Richard Posner, author of a classic Chicago antitrust monograph and now a distinguished federal judge, has adopted exactly this tack. After explaining flaws in the Shoe court opinion, Posner cites the Coase Conjecture as a possible motivation for the lease-only policy: "Professor Coase has argued that the lease-only policy of a monopolist of a durable good, such as United Shoe Machinery Corporation, may be designed to overcome the difficulties encountered in trying to charge a monopoly price for a durable good." n14 Yet if antitrust law aims to force sellers to price at competitive levels, then the Coase Conjecture implies that judges should embrace the Shoe ban on lease-only monopoly. Posner makes just this suggestion: "Perhaps the lease-only policy should have been forbidden on that ground . . . ." n15
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n14 R. POSNER, ANTITRUST LAW: AN ECONOMIC PERSPECTIVE 184 (1976); see also R. POSNER & F. EASTERBROOK, ANTITRUST: CASES, ECONOMIC NOTES AND OTHER MATERIALS 627 (2d ed. 1981); Froeb, Evaluating Mergers in Durable Goods Industries, 34 ANTITRUST BULL. 99 (1989).

n15 R. POSNER, supra note 14, at 184.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

In delightful irony, however, it is not Posner's Coase-based suggestion but rather the usual Chicago reluctance to interfere in the market that gives the right result in this case. True enough, in one respect the Coase Conjecture does support the Shoe rule; indeed, it suggests that courts should broaden the rule significantly. In other ways, though, the Conjecture implies that judges ought to narrow the Shoe rule, for the Conjecture applies only under stringent qualifications -- and the rule certainly was not appropriate in the Shoe case itself. These qualifications imply that courts should either bound Shoe's proscription with a large number of inevitably arbitrary lines, or scrap the Shoe rule altogether. We argue the latter. In no event should a court consider Shoe's ban on lease-only conduct to be sensible law.

I. THE SHOE CASES

The United Shoe Machinery Corporation may be the most popular antitrust defendant of all time. Plaintiffs have taken it from district courts to the Supreme Court six times in one century. The federal government first (and unsuccessfully) claimed a criminal violation of the Sherman Act by five individuals who in 1899 merged their companies to create the United Shoe Machinery Company.  [*698]  Holmes's unanimous 1913 opinion upheld the district court's dismissal of the attack on the merger. n16 Again using the Sherman Act, the government sued the corporate entity United Shoe Machinery Company for the same merger, n17 as well as for a number of its standard leasing clauses (which Justice Louis Brandeis had drafted in private practice). n18 In 1918, the Supreme Court upheld the merger against this second challenge and approved United's leasing practices for the first time. n19 On a third try in 1922, invoking the Clayton Act of 1914, the government finally succeeded in winning Supreme Court condemnation of specific lease clauses although not of United's general leasing policy. n20
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n16 United States v. Winslow, 227 U.S. 202 (1913). In addition to the merger, the government also sought to attack the merged defendants' practice of

ceasing to sell shoe machinery to the shoe manufacturers. Instead, they only let machines, and on the condition that unless the shoe manufacturers use only machines of the kinds mentioned furnished by the defendants, or if they use any such machines furnished by other machinery makers, then all machines let by the defendants shall be taken away. This condition they constantly have enforced.
Id. at 216.

Justice Holmes, however, ruled that that conduct was not before the Court, and "the combination was simply an effort after greater efficiency." Id. at 217; see also United States v. United Shoe Mach. Co., 222 F. 349 (D. Mass. 1915), aff'd, 247 U.S. 32 (1918) (also endorsing an efficiency interpretation of the merger). Supporting Holmes's efficiency view are the facts that the merged company combined previously separate operations into a single new plant in Beverly, Massachusetts and created new service and research departments. C. KAYSEN, UNITED STATES V. UNITED SHOE MACHINERY CORPORATION 9-10 (1956). For an interpretation of the merger that stresses its anticompetitive character (and does not discuss these facts), see Bittlingmayer, Did Antitrust Cause the Great Merger Wave?, 28 J.L. & ECON. 77, 102-03 & n.54 (1985) (arguing that early and successful price-fixing prosecution channeled cartel efforts into merger activity and supporting this claim in the Shoe instance with testimony from a lawyer involved in the merger). Of course, a merger can both increase productive efficiency and facilitate monopoly pricing. See Williamson, Economies as an Antitrust Defense: The Welfare Tradeoffs, 58 AM. ECON. REV. 18 (1968).

n17 The government could relitigate the same issue because of the preclusion requirement of mutuality which modern procedure has abandoned. See Parklane Hosiery Co. v. Shore, 439 U.S. 322 (1979); Blonder-Tongue Laboratories v. University of Ill. Found., 402 U.S. 313 (1971).

n18 C. KAYSEN, supra note 16, at 15 n.38.

n19 United States v. United Shoe Mach. Co., 247 U.S. 32 (1918). Half a century later, the Supreme Court said that the 1918 Court's reasons for vindicating United's lease terms were "not clear." Hanover Shoe, Inc. v. United Shoe Mach. Corp., 392 U.S. 481, 500 (1968).

n20 United Shoe Mach. Corp. v. United States, 258 U.S. 451 (1922). The Court rejected United's defense of res judicata by holding that the Clayton Act cause of action differed from the previous Sherman Act claims. Id. at 460. The Court enjoined United's use of seven particular types of lease clauses, concluding that "while the clauses enjoined do not contain specific agreements not to use the machinery of a competitor of the lessor, the practical effect of these drastic provisions is to prevent such use." Id. at 457. The specific clauses were the following:
(1) the restricted use clause, which provides that the leased machinery shall not . . . be used upon shoes . . . upon which certain other operations have not been performed on other machines of the defendants; (2) the exclusive use clause, which provides that if the lessee fails to use exclusively machinery of certain kinds made by the lessor, the lessor shall have the right to cancel the right to use all such machinery so leased; (3) the supplies clause, which provides that the lessee shall purchase supplies exclusively from the lessor; (4) the patent insole clause, which provides that the lessee shall only use machinery leased on shoes which have had certain other operations performed upon them by the defendants' machines; (5) the additional machinery clause, which provides that the lessee shall take all additional machinery for certain kinds of work from the lessor or lose his right to retain the machines which he has already leased; (6) the factory output clause, which requires the payment of a royalty on shoes operated upon by machines made by competitors; (7) the discriminatory royalty clause, providing lower royalty for lessees who agree not to use certain machinery on shoes lasted on machines other than those leased from the lessor.
Id. at 456-57.

The Court rejected United's defense that it offered an alternative lease free of restrictions but which required an initial cash payment. This unrestricted lease option more closely resembled a sale, but the required initial payment was so large that "no manufacturer ever chooses the unrestricted lease." United States v. United Shoe Mach. Co., 264 F. 138, 164 (E.D. Mo. 1920), aff'd, 258 U.S. 451 (1922). The Supreme Court held that "the fact that a form of lease was offered which is not the subject of controversy is not a justification of the use of clauses in other leases which we find to be violative of the act." 258 U.S. at 464.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

 [*699]  In 1947, the government sued a fourth time, now attacking aspects of leasing that had survived the third challenge. n21 After a "trial of prodigious length," n22 District Judge Wyzanski ruled in 1953 that United had violated the Sherman Act. He found that the company had monopoly power (more than seventy-five percent of the shoe machinery market) and had used many objectionable practices. Central to his holding was United's exclusive reliance on leasing for its most important machines. n23 According to the court, this  [*700]  practice "created barriers to entry by competitors into the shoe machinery field." n24 Judge Wyzanski condemned the
complex of obligations and rights [that] deter a shoe manufacturer from disposing of a United machine and acquiring a competitor's machine. He is deterred more than if he owned that same United machine, or if he held it on a short lease carrying simple rental provisions and a reasonable charge for [cancellation] before the end of the term. n25

 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n21 United States v. United Shoe Mach. Corp., 110 F. Supp. 295 (D. Mass. 1953). As the district court explained the prosecution:
Until Alcoa lost its case in 1945, there was no significant reason to suppose that United's conduct violated 2 of the Sherman Act. . . . What United is now doing is similar to what it was then doing, but the activities which were similar stood uncondemned, -- indeed, one ought to go further and say they were in part endorsed.
Id. at 348. The court did allow that the doctrine of res judicata protected the original merger. Id. at 344.

n22 Id. at 298. The trial lasted for 121 days, generating 14,194 pages of transcript and 26,474 pages of exhibits. Id. at 299.

n23 "Of the 342 machine types now marketed by United, United offers 178 on lease basis only, 42 on sale basis only, and 122 on alternate lease or sale terms at the customer's option. . . . Those of more importance for the shoe manufacturer are offered only for lease." Id. at 314.

n24 Id. at 340. The circuit court in the third prosecution specifically declared that United's leasing system in itself was not a violation of the Clayton Act. C. KAYSEN, supra note 16, at 15 (citing 264 F. 138).

n25 110 F. Supp. at 340.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

This "complex of obligations and rights" arose from the leases' ten-year terms (which United enforced in a discriminatory fashion), n26 the "full capacity clause," n27 the deferred payment charge (or "return charge"), n28 and the "free repair clause." n29 Judge Wyzanski  [*701]  further condemned United for earning different rates of return on different lines of machines, depending on whether or not the firm faced competition in a particular line. n30 As relief, the court ordered United to offer its machines for sale. It permitted United to continue leasing, but only under certain conditions: at nondiscriminatory rates; for no more than a five-year term; with no full capacity clause; and with separate repair charges. n31 On the same day it decided Brown v. Board of Education, n32 the Supreme Court affirmed the district court in a single sentence. n33
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n26 "When a lessee desires to replace a United machine, United gives him more favorable terms if the replacement is by another United machine than if it is by a competitive machine." Id. at 340. United required partial payment of full lease obligations if a lessee returned a machine to replace it with a competitor's machine. United did not make a similar demand when lessees returned machines for such other reasons as the following: they had abandoned use of that machine's particular operation; they began to perform it by hand; the United machine had not operated as anticipated; or they sought a different United machine. Id. at 320. The court said "the discrimination is designed to operate as, and does operate as, a method of excluding from the shoe factories shoe machinery competitive with United." Id. at 321.

The earlier Clayton Act adjudication had declared that the 17-year term of United's previous lease "is not prohibited by the statute." 264 F. at 168.

n27 United's standard lease provided that "the lessee shall use the leased machinery to its full capacity . . . ." 110 F. Supp. at 316. The district court reported that United applied this term if "the lessee fails to use the machine on work for which the machine is capable of being used, and instead performs such work by using a competitor's machine." Id. at 320. The government offered 90 instances of United's use of the capacity (or a similar) clause. For example, "from 1931 to 1935 Florsheim wanted to return 5 United machines covered by unexpired leases, and substitute competitors' machines, but United insisted on the lease terms, including the full capacity clause." Id. at 321. United also "bill[ed] its customers under the full capacity clause when competitive machines were used . . . ." Id. The previous government challenge also had failed against this clause. See C. KAYSEN, supra note 16, at 15 (citing 264 F. 138); see also infra notes 84-91 and accompanying text.

n28 United levied this type of charge on all lessees when they returned the machine to United, at either the conclusion or the termination of the lease. 110 F. Supp. at 320. See Equipment Distrib. Coalition v. FCC, 824 F.2d 1197, 1202 (D.C. Cir. 1987). The court found it objectionable that return charges "can in practice, through the right of deduction fund, be reduced to insignificance if [the lessee] keeps this and other United machines to the end of the periods for which [the lessee] leased them. 110 F. Supp. at 340; see infra note 65.

n29 110 F. Supp. at 340; see infra text accompanying notes 60-61.

n30 Id. at 340-41; see infra notes 62-65 and accompanying text. In addition, the district court faulted United for acquiring patents and for purchasing second-hand shoe machines for scrap. But the court made only "brief reference" to these factors after it had outlined the other "principal sources of United's power." Id. at 344. The extent of United's latter practice was "trivial." See C. KAYSEN, supra note 16, at 113.

n31 The court also barred United from distributing others' machines; from continuing to own certain subsidiaries; and from refraining to license its patents to competitors. 110 F. Supp. at 346-54.

n32 347 U.S. 483 (1954).

n33 347 U.S. 521 (1954). United Shoe made two more forays to the Supreme Court. Trip five arose from the government's ultimately successful effort to obtain additional relief, including divestiture. See United States v. United Shoe Mach. Corp., 391 U.S. 244 (1968); see also 1969 Trade Cas. (CCH) P72688 (D. Mass.) (United agreed both to sell assets to reduce its size to one-third of the market and to additional patent licensing). Trip six was a successful "me too" suit for damages by a private plaintiff. See Hanover Shoe, Inc. v. United Shoe Mach. Corp., 392 U.S. 481 (1968).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Judge Wyzanski's 1953 Shoe opinion established a rule against lease-only monopoly -- or so the Supreme Court said in 1968, seven votes to one. n34 This interpretation remains the dominant one, n35 but it has been clouded recently by two circuit courts that, while acknowledging the Shoe ban on lease-only policies, found that such  [*702]  policies were not illegal in their cases. n36 Our focus is upon the wisdom of this controversial lease-only holding, rather than on the separate concern that individual lease clauses amounted to exclusionary promises by lessees not to patronize United's competitors. n37
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n34 See Hanover Shoe, 392 U.S. at 484-87 (Wyzanski's opinion outlawed United's lease-only policy in and of itself and apart from the "certain clauses in the standard lease" that it also condemned); see also id. at 511 (Stewart, J., dissenting) (arguing that Shoe "held unlawful only particular kinds of leases with particular provisions, not United's general practice of leasing only") (footnote omitted); see also supra note 2 and accompanying text.

Leasing was at issue in various private suits against IBM in the 1970s and early 1980s. See, e.g., Transamerica Computer Co. v. International Business Mach. Corp., 698 F.2d 1377, 1382 (9th Cir.), cert. denied, 464 U.S. 955 (1983); Telex Corp. v. International Business Mach. Corp., 510 F.2d 894, 919-28 (10th Cir.), cert. denied, 423 U.S. 802 (1975). But such cases are not pertinent to our discussion because IBM during this time operated under a consent decree that forced it to offer a sales option. See supra note 2.

n35 See, e.g., Robinson, Recent Antitrust Developments -- 1979, 80 COLUM. L. REV. 1, 4 (1980) (Judge Wyzanski found United Shoe Machinery had run afoul of monopoly law "because it employed a restrictive 'lease only' policy which excluded actual and potential competition.").

n36 In Williamsburg Wax Museum v. Historic Figures, Inc., 810 F.2d 243, 253 (D.C. Cir. 1987) (citing but without explanation ignoring Hanover Shoe), Judge Mikva allowed leasing because the defendant in fact did sometimes sell the product and because (unlike in United Shoe) there was not an "entire panoply of practices." Id. The Ninth Circuit allowed leasing in Souza v. Estate of Bishop, 821 F.2d 1332, 1337 (9th Cir. 1987) (failing to cite Hanover Shoe), because it "creates no continuing relationship between lessor and lessee" and involves no "terms that inhibit lessees from purchasing or leasing land owned by other landholders." Both opinions fail to make clear whether lease-only is impermissible by itself, or only "in combination with other practices."

n37 A monopolist may try to insert into a lease exclusionary promises from customers that they not deal with the monopolist's competitors. A simple exclusionary promise is illegal under undisputed law. See infra note 48. But leasing is neither necessary to nor sufficient for such exclusion. Monopolists can seek such promises independently, or in connection with contracts for new or continued sales, for service, or for anything else. Our main attention in this Article is on United's lease-only practice, not on United's particular lease terms or practices that some have believed amount to such exclusionary promises. We dispute some of these beliefs in Part II(B) and discuss the issue at greater length in E. Rasmusen, J.M. Ramseyer & J. Wiley, Naked Exclusion, UCLA Anderson Graduate School of Management Business Economics Working Paper #89-17 (1989), and Rasmusen, Recent Developments in the Economics of Exclusionary Contracts, in THE CENTENARY OF COMPETITION LAW IN CANADA (R.S. Khemani & W.T. Stanbury eds.) (forthcoming 1990) (working title). We make two further observations on this score.

First, if United did bundle any exclusionary clause into its lease, the clause it bundled was inexplicably generous. The Shoe court made much of United's policy of charging a lessee more if it returned the machine to replace it with a rival's than for another reason. See supra note 26. Yet these assertedly exclusionary charges were considerably less than the customer's remaining liability under the terms of the lease. See 110 F. Supp. at 320. A United bent on excluding rivals ought to have insisted on full liability under the truncated lease.

Second, Kaysen reports that "it is clear that United does not oppress shoe manufacturers in any way that makes them generally conscious of such oppression." C. KAYSEN, supra note 16, at 203. United's supposed efforts to exclude its rivals thus escaped its customers, who would bear the brunt of being the vehicle for any real exclusionary scheme.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

One also might read the Shoe opinion in a far different light: as an artifact of a bygone era in antitrust, a relic from the days when populist judges aimed the Sherman Act at efficient firms simply because they were large, rather than to achieve efficient or proconsumer ends. To support this reading, one would fasten on Shoe's embrace of Judge Hand's famous Alcoa opinion, n38 which is unmistakably of this genre and which some have argued is no  [*703]  longer living law. n39 But we doubt Shoe is dead. No judge has said so. n40 The hornbooks treat it as an important source of the law of monopoly. n41 A law firm would be courageous (or not recently in touch with its malpractice carrier) if it counseled a client with market power to adopt a Shoe-style lease policy. The horse we beat is not dead. But as the following Section shows, neither is it well shod.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n38 United States v. Aluminum Co. of Am., 148 F.2d 416 (2d Cir. 1945); see supra note 21. Judge Wyzanski had served as law clerk to Judge Learned Hand, referring to the Judge as one of "the Hand boys." Richardson, In Memoriam: Charles E. Wyzanski, Jr., 100 HARV. L. REV. 723, 726 (1987).

n39 See, e.g., Robinson, supra note 35, at 1 ("Judge Learned Hand's frequently cited but little understood Alcoa opinion was given a decent burial by the Second Circuit in Berkey v. Kodak and replaced with a more realistic and workable concept of monopolization."). To bolster this case, one might note that Judge Wyzanski confessed with disinterested candor that United Shoe Machinery's illegal practices were not only a longstanding industry custom to which its customers offered no complaint, but in some respects actually arose "to meet their preferences." 110 F. Supp. at 314, 319, 323. One further would note that the court thought it culpable that "being by far the largest company in the field, with by far the largest resources in dollars, in patents, in facilities, and in knowledge, United has a marked capacity to attract offers of inventions, inventors' services, and shoe machinery businesses." Id. at 343-44. Finally one would argue that the court appeared to regard United Shoe Machinery's research and development effort as a reason more to condemn than to praise the firm. Id. at 344-46; see also infra notes 65, 79.

n40 See supra notes 2, 34, 36.

n41 See, e.g., H. HOVENKAMP, ECONOMICS AND FEDERAL ANTITRUST LAW 137 (1985) (offering portions of Shoe's analysis as an approximation of "the prevailing legal rule"); L. SULLIVAN, HANDBOOK OF THE LAW OF ANTITRUST 114 (1977) ("The leasing policy described in United Shoe Machinery is the best example in the case law of conduct which tends to raise entry barriers.") (footnote omitted). But see H. HOVENKAMP, supra, at 148-49 (impliedly criticizing different Shoe holding that price discrimination is bad conduct).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

II. UNDERSTANDING LEASING

A. The Chicago Critique

A question different from whether the Shoe rule still reigns is whether it makes sense. One cannot fault Judge Wyzanski for "ignoring economics" because in deciding the Shoe case he had the very latest advice from the Harvard Department of Economics. After two years of pretrial proceedings, he had decided to hire as law clerk Carl Kaysen, then an assistant professor at Harvard. Kaysen taught his classes and attended parts of the trial by day, read the transcripts at night, eventually wrote his dissertation about the case, and went on to become a distinguished professor of industrial organization. n42
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n42 See Reilly v. United States, 682 F. Supp. 150, 161-62 (D.R.I. 1988) (describing Kaysen's involvement); Kaysen, In Memoriam: Charles E. Wyzanski, Jr., 100 HARV. L. REV. 713 (1987). Judge Wyzanski apparently later decided that appointing an economist as clerk improperly shielded expert testimony from examination by the parties. Id. at 715.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

 [*704]  Times change. Two years later, Aaron Director and Edward Levi launched a different brand of economics: the Chicago critique of antitrust law. n43 They did so through an article that at least implicitly attacked Shoe by challenging the notion that lease-only policies could enable monopolists profitably to exclude competitors -- a notion central to the Shoe decision. n44 By the 1970s, this attack had become explicit. Writers like Robert Bork, n45 Richard Posner, and Frank Easterbrook n46 all directly criticized the Shoe court's claim that United could have used its leasing policies to exclude rivals. They argued that it was not possible for a monopolist to exclude rivals more successfully with leases than sales. To them, a monopolist could use leases either to charge monopoly prices or to exclude rivals, but not to do both. And if using the leases to exclude prevented the monopolist from earning monopoly returns, then the monopolist would not exclude.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n43 Director & Levi, Law and the Future: Trade Regulation, 51 NW. U.L. REV. 281, 290 (1956).

n44 See 110 F. Supp. at 340.

n45 See R. BORK, THE ANTITRUST PARADOX 136-60, 164-75 (1978).

n46 See R. POSNER, supra note 14, at ch. 8; R. POSNER & F. EASTERBROOK, supra note 14, at ch. 6. Cf. United States v. Aluminum Co. of Am., 44 F. Supp. 97, 121, 123, 136-38, 143-44 (S.D.N.Y. 1941) (extensive fact survey offers no support for Easterbrook suggestion), aff'd in part and rev'd in part, 148 F.2d 416 (2d Cir. 1945); G. SMITH, FROM MONOPOLY TO COMPETITION: THE TRANSFORMATIONS OF ALCOA, 1888-1986 80, 94-96 (1988) (more mixed record).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Despite its logical clarity, the Chicago thesis remains problematic. The thesis proceeds in two apparently straightforward steps: (1) monopolists can never buy inefficient exclusionary terms for free; and (2) the price for such terms will always make the deal a bad one for the monopolist. Notwithstanding the clarity, however, at least two problems remain. First, it is far from certain as a matter of economic theory that monopolists can never acquire exclusionary terms cheaply -- an issue that we explore in another paper. n47 Second, real counterexamples do cast doubt on the Chicago thesis. n48  [*705]  For purposes of this Article, therefore, we will not rely upon the Chicago reasoning in our criticism of Shoe's rule. n49
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n47 In E. Rasmusen, J.M. Ramseyer & J. Wiley, supra note 37, we show why a monopolist may in fact be able to buy inefficient exclusionary terms at a very low price. Given certain common and plausible conditions, exclusion may well be a profitable strategy. Nonetheless, for reasons we outline in Part II(B), we do not believe the Shoe leases were exclusionary.

n48 See Lorain Journal Co. v. United States, 342 U.S. 143 (1951) (newspaper refuses advertising from customers submitting ads to local radio station); United States v. Aluminum Co. of Am., 148 F.2d 416, 422 (2d Cir. 1945) (Hand, J.) (Alcoa buys promises from power suppliers not to sell power to other aluminum companies); cf. Easterbrook, Allocating Antitrust Decisionmaking Tasks, 76 GEO. L.J. 305, 315-16 (1987) (offering a speculative efficiency justification for Alcoa's conduct).

n49 For different arguments that the Chicago analysis of excluding through leasing is incomplete, see Kaplow, Extension of Monopoly Power Through Leverage, 85 COLUM. L. REV. 515 (1985).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Lease-only policies are not so mysterious as exclusionary promises. In the following section, we discuss the source of Judge Wyzanski's misunderstandings and explain why exclusion cannot account for the bulk of United's conduct. We then outline the motives that could account for that behavior. Of these, most are innocuous; only the Coase Conjecture might support Shoe's rule against lease-only monopolies. In the last Section, however, we explain why that Conjecture calls not for a ban on leases, but for a rejection of the Shoe rule.

B. Fallacies of Exclusion

Judge Wyzanski condemned United conduct that he said was "exclusionary." Unfortunately, vigorous competition also excludes. A firm that offers high quality at low price necessarily excludes rivals that offer less for more. Judge Wyzanski grasped this fundamental when he "confessed at the outset that any system of selling or leasing one company's machines will, of course, impede to some extent the distribution of another company's machines." n50 But he did not heed the implication. To be undesirable, conduct must violate a more basic antitrust standard: the exclusion must be inefficient or unfair to consumers. n51 Under a proper analysis, conduct that Judge Wyzanski condemned is in fact innocuous.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n50 110 F. Supp at 324.

n51 There is controversy over the basic standard by which antitrust law ought to measure good and evil. See, e.g., Lande, The Rise and (Coming) Fall of Efficiency as the Ruler of Antitrust, 33 ANTITRUST BULL. 429 (1988). Consensus that competitive prices are socially preferable to monopoly prices, however, saves us from having to rehearse this controversy here.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

1. Leasing Only

A monopolist cannot exclude rivals more effectively by leasing exclusively than by selling. A sale is simply a lease for the life of the product. Short leases free customers from capital commitments to the monopolist's product. Leasing thus leaves those customers more able to switch to a rival. The monopolist's lease-only policy does prevent a second-hand or recycled market. n52 But it facilitates  [*706]  a more effective form of competition: entry by firms that make new competing goods. n53
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n52 Judge Wyzanski complained about this effect. See 110 F. Supp. at 325. For a discussion of this form of competition, see Carlton & Gertner, supra note 3, at S211-12.

n53 Cf. F. FISHER, J. McGOWAN & J. GREENWOOD, supra note 2, at 195-96 (unwanted leasing encourages new entry and cannot be a barrier to it).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

2. Long Terms

Judge Wyzanski thought the length of United's ten-year leases increased their exclusionary effect. n54 In fact, long-term leases are no more exclusionary than sales -- even if (as in Shoe) n55 the leases bar subletting. A monopolist may try to exclude rivals by satisfying the available demand, but its ability to do so does not depend on the form of the transaction. What counts is that consumers have patronized the monopolist -- whether by sale or by lease -- and want to consume no more. Using either form of transaction, the monopolist can reap its monopoly profit and nothing greater.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n54 110 F. Supp. at 324-25.

n55 See id. at 315.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Consider a market in which the seller has a safe monopoly in period one, but in period two faces potential entry that it would like to exclude. It can adopt one of three strategies: (1) outright sales; (2) successive one-period leases; or (3) long-term leases covering both periods. Although Judge Wyzanski was vague about the exact mechanics, an argument that long-term leases are more exclusionary than sales might run as follows. If the monopolist uses the short-term lease, then entry is easy. The entrant offers a low price in the second period and can attract the monopolist's customers. If the monopolist uses outright sale, entry is also easy. The entrant offers a low price in the second period, and the customer can resell the original purchase and replace it with one of the entrant's products. The long-term lease, however, is exclusionary because the monopolist can include a clause forbidding subletting. Thus, the monopolist commits the buyer to hold the product for the two-period duration of the lease. Even if the entrant appears with a lower price, entry is unprofitable because the customer cannot switch.

This argument errs by ignoring the prices of the lease, of the sale, and of the resale. To offer a simple example with no discounting, suppose the monopoly price per period of use is 20, and the competitive price (charged by the entrant) is 10.

Using a short-term lease, the monopolist can rent the product at a price of 20 for the first period and 10 for the second, for total revenue of 30. The high monopoly profits in the first period attract  [*707]  competitors into the industry, and those new entrants compete away the profits in the second. n56
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n56 For a discussion of when a monopolist might be able profitably to prevent entry in the second period, see E. Rasmusen, J.M. Ramseyer & J. Wiley, supra note 37.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Using outright sale, the monopolist can sell the product at a price of 30. If the monopolist sets the price any higher, it would exceed the customers' willingness to pay. After all, the monopoly rental charge of 20 for the first period indicates that customers would pay no more than that for the right to use the product during that time; and because they expect competitors to enter in the second period, they will pay no more than 10 for the right to use it during the second. In a two-period model, therefore, customers will pay a total price of only 30. If the monopolist charged more, they would switch to the next best substitute or drop out of the market in the first period and buy (or rent) from the entrant in the second. Most importantly, customers who buy in the first period do not benefit from selling their product in the second to buy from the competitor. When the competitor appears offering to sell at 10, the resale price falls to 10 as well. Accordingly, first-period buyers can charge no more than 10 for the products they bought at 30. Unfortunately for the first-period buyer, there is no point to selling at 10 to buy at 10. n57
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n57 Moreover, because it would not benefit the buyer to resell, the monopolist would not hurt the customer by including a no-resale clause in the initial sales contract.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

For the same reasons, a monopolist who used a long-term lease could charge no more than 30, although it could allocate this total over time in a variety of ways. Because the customer is willing to pay a total of 30 for the right to use the product over the two periods, the customer will not care whether it pays 20 for the first period and 10 for the second, 15 each, or 10 for the first and 20 for the second. Whatever the allocation of rent, however, the monopolist cannot do any better with a long-term lease than with an outright sale, and the customer cannot do worse. n58
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n58 Note, however, that there may be a variety of efficiency reasons for selecting one form of financing over others. See infra text accompanying notes 66-91.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

If long-term leases will not increase a monopolist's profits, neither will they automatically exclude entrants. Absent specifically exclusionary clauses added to a lease, long-term leases are no more exclusionary than outright sales. Both sales and long-term leases deny entrants access to the customers that buy or lease in the first period. If a long-term lease has a no-sublet clause, it also takes those customers out of the second-period market. But the same occurs  [*708]  with a sale. After all, the first-period sale puts units on the market that some customer has to own. Both long-term leases and sales, in short, necessarily reduce the number of customers that entrants can serve, and by the number of customers willing to pay the monopoly price in the first period. n59
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n59 In fact, even if a monopolist uses short-term leases, the entrant will not enter if in the second period the monopolist leases at 10. In other words; the monopolist can make entry yield zero profits if it is willing to accept zero profits itself.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

3. Tied Repair Service

The Shoe court faulted United for offering "free" repair because this practice bundled United's repair service with its lease. n60 Yet United could not have used such a tie to exclude an equally efficient rival. Nothing in the tie prevented the rival from offering service of its own. At core, the Shoe holding simply reflects the lament that in some industries large companies may be better able than small companies to offer the services that customers want. If antitrust law bans such arrangements, it hurts no one more than the customers themselves. n61
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n60 110 F. Supp. at 340.

n61 Cf. F. FISHER, J. McGOWAN & J. GREENWOOD, supra note 2, at 218 ("IBM's policy of protecting its own assets by maintaining them itself could not be a barrier to entry."). On the efficiency reasons for tying service contracts to leasing and sales contracts, see infra text accompanying notes 77-81.

We analyze this particular example of tying only, not the more general controversies in that field. For a discussion of these more general controversies, see Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2 (1984).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

4. Differential Rates of Return

Judge Wyzanski ruled that United had excluded competitors by earning rates of return scaled to the level of competition faced by each of its machines. n62 Yet current law permits a monopolist to charge a monopoly price. It does so because any ban on monopoly pricing would force courts to become, in effect, public utility rate regulators. n63 If charging a monopoly price is legal, however, then the onus of United's actions lay in lowering its monopoly price on  [*709]  some items to a competitive level. n64 No antitrust rule could be more perverse than a ban on competitive pricing. n65
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n62 See supra text accompanying note 30.

n63 See Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263, 294 (2d Cir. 1979) ("Such judicial oversight of pricing policies would place the courts in a role akin to that of a public regulatory commission."), cert. denied, 444 U.S. 1093 (1980).

n64 The accounting and supervision task that judicial review of pricing would involve is hardly eased because a monopolist earns a variety of different accounting profits on different products. Holding low prices illegal as exclusionary will provide a subsidy to small businesses at the expense of consumer welfare. But see Peritz, A Genealogy of Vertical Restraints Doctrine, 40 HASTINGS L.J. 511, 572-74, 576 (1989) (defending small business preservation as an appropriate antitrust goal).

n65 In addition to the four problems just discussed, Judge Wyzanski also incorrectly analyzed United's "Right of Deduction Fund" policy. Under this policy, United set aside a small percentage of each lessee's rent payments for that lessee's Right of Deduction Fund. A lessee that returned a machine early would otherwise have faced a variety of special charges. Under this policy, however, the lessee could apply the money in that Fund toward those charges. Judge Wyzanski concluded that this policy "deters, though probably only mildly, the opportunities of a competing shoe manufacturer." 110 F. Supp. at 325. In essence, however, the Fund was nothing more than a quantity discount. As such, the Fund did deter the opportunities of United competitors. But discounting always does. That is competition.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

C. Efficiency Reasons for Leasing

If United did not lease to exclude competitors, why it did lease remains an issue. A wide variety of explanations present themselves, and we discuss several of the more prominent: transaction costs, risk-bearing, financing, incentives for maintenance, and incentives for product quality. n66 None of these justifies a ban on lease-only policies. Instead, each shows how leasing often benefits both the customer and the producer. In fact, it often benefits the  [*710]  parties so extensively that neither has any interest in a sale. A lease-only policy can arise, in short, not because the producer wants to suppress sales, but because no customer wants to buy. The Shoe case apparently presented just this situation, given the complete absence of customer protest about leasing. n67
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n66 Judge Wyzanski's opinion shows that he understood (however unenthusiastically) many of these efficiency reasons for leasing:
It has been easy for a person with modest capital and of something less than superior efficiency to become a shoe manufacturer. He can get machines without buying them; his machines are serviced without separate charges; he can conveniently exchange an older United model for a new United model; he can change from one process to another . . . .

110 F. Supp. at 323.

The court also found that "leasing has been traditional in the shoe machinery field since the Civil War. So far as this record indicates, there is virtually no expressed dissatisfaction from consumers respecting that system; and Compo, United's principal competitor, endorses and uses it." Id. at 340. The court also pointed out:
[A] large number of shoe manufacturers . . . expressed their preference for leasing rather than buying machines. How deeply rooted is this preference might be disputed; but it cannot be denied that virtually all the shoe manufacturers who took the stand, and the 45 shoe manufacturers who were selected as a sample by the Court, expressed a preference for the leasing system.
Id. at 349. That Judge Wyzanski insisted on a "sample" of 45 more consumer witnesses suggests that his regard for the value of leasing did not come swiftly. Nor did his opinion fully recount United's more elaborate and compelling efficiency defense. C. KAYSEN, supra note 16, at 190-91; see also id. at 202 ("As seen by the shoe manufacturer, United's activities are clearly benevolent.").

n67 See C. KAYSEN, supra note 16, at 278 ("The testimony of the shoe manufacturers indicates unequivocally that they do not now desire a sales system."); supra note 66.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

1. Transaction Costs

Most importantly, leasing can reduce transaction costs. A short-term user's alternative to leasing is to purchase: to buy the product, use it for a short time, and then resell it. In a world without transaction costs, business travelers would be willing to buy cars at their destinations and resell them there the next day. We do not see this behavior because it is prohibitively expensive: Each traveler would have to find a car seller, arrange transportation to meet the seller, determine car quality, register ownership, pay sales tax, buy auto insurance -- and then repeat the transaction the next day as the seller. The costs to a car rental agency for the equivalent rental transaction are trivial by comparison. Even if it is a monopolist, the agency can offer a rental service that travelers would prefer to the prospect of enmeshing transactions. n68
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n68 Bankruptcy costs, a special form of transaction costs, also may provide a reason to lease rather than to sell. Victor Goldberg has suggested that leasing expedites the reallocation of resources in the event the user of the product declares bankruptcy. If bankruptcy law speeds the redeployment of assets that are leased rather than sold, the prospect of this profitable recovery and reuse of the asset could decrease United Shoe Machinery's willingness to sell, thus leading to a lease even when the customer was indifferent about the form of the transaction. V. Goldberg, The United Shoe Machinery Leases 11-13 (unpublished manuscript). Consistent with Goldberg's suggestion, United's standard lease did give it the right to terminate forthwith in the event a lessee became insolvent or bankrupt, in which case the lessee was obligated to return the machine to United in Massachusetts. 110 F. Supp. at 317-18. We remain unsure, however, how much force to ascribe to Goldberg's suggestion. It appears that bankruptcy law privileges leasing over chattel mortgage sales only in insignificant ways, and that this difference does not create any standard preference for leasing among bankruptcy lawyers. Moreover, a concern with bankruptcy recovery does not seem to explain particular features of the United leases that the government repeatedly attacked, while a great number of other possible efficiencies can account for United's general interest in leasing.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

2. Risk Allocation

Risk allocation also often accounts for leasing. Suppose that a small business, Smithco, decides that it needs a computer given the present level of sales, but that sales might decline. If Smithco buys  [*711]  the computer, it bears the risk that future sales indeed will decline and that the computer will go unused. If it leases the computer, it can shift that risk to IBM. If the producer can more cheaply diversify the risk than the user, leasing is efficient. n69 Special provisions of the lease can shift the risk still further. By making its lessee's monthly rental charge largely contingent on the number of shoes made with the machines, United's leases did just that. n70
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n69 See F. FISHER, J. McGOWAN & J. GREENWOOD, supra note 2, at 191-93 (discussing differences in allocation of risk in computer lease and sale transactions); see also United Shoe, 222 F. at 391 (lessees of United's machinery "cannot be supposed desirous of surrendering" the advantage "of paying for the privilege of their use only according to the amount of use").

n70 If shoe demand is low, few shoes are made, and the total rental fee is low. If the entire fee were a flat monthly payment, on the other hand, the renter would have just as high payments in bad times as in good. The form of the lease acts as partial insurance against low demand for shoes.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The parties can also use leases to allocate a variety of other risks between producer and user. The risk of technological obsolescence is one such risk. n71 United's leases reflected a variety of others. The company charged its lessees four fees: (a) a monthly rental, (b) a payment per pair of shoes made with the machine (with a flat minimum if too few shoes were made), (c) a payment upon termination or expiration of the lease, and (d) a payment in case of loss by fire or accident. n72 Under charge (d), the lessee did bear part of the loss in case of fire -- presumably to give some incentive to smoke carefully and to extinguish fires. But the lease acted like fire insurance with a deductible; the lease stated that the charge was only "partial reimbursement . . . for such destruction, and the lessee [was required] forthwith [to] return whatever remain[ed] of the machinery so destroyed to the United Corporation at Beverly, Massachusetts." n73 Moreover, by requiring the return of the machine's wreckage to verify loss, United could eliminate a lessee's incentive to avoid future lease payments by paying $ 200 and claiming total loss of a usable machine. True, had the customer bought the machine instead of leasing it, it could have bought its own separate  [*712]  fire insurance. Yet given the ease with which insured parties can file fraudulent claims on movable machinery, insurance for its loss might be more expensive. n74 Leasing may have provided that insurance more cheaply.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n71 Compare F. FISHER, J. McGOWAN & J. GREENWOOD, supra note 2, at 191-92 & n.16 (competitive pressure forced IBM to reject its treasurer's 1964 suggestion that IBM refuse to rent any System/360 equipment so as to place risk of technological obsolescence on customers) with Carroll, Hurt by a Pricing War, IBM Plans Write-Off and Cut of 10,000 Jobs, Wall St. J., Dec. 6, 1989, at A1, col. 6 ("That security blanket unraveled in the 1970s as technology began moving so fast that IBM decided it had to start selling machines, lest its rental equipment suddenly become obsolete overnight.").

n72 110 F. Supp. at 314-18.

n73 Id. at 315-16. The fire-loss payment for a typical machine was $ 200, which is low compared with the minimum yearly rent of $ 87 and the expiration payment of $ 75. Id. at 314.

n74 United was in a better position to avoid such fraud because of its extensive knowledge of its customers' operations. See infra note 77; C. KAYSEN, supra note 16, at 23, 29, 47-50. United's "Outside Machine Installation Reports" tracked customers' use of rival machines "with all the attention a doting mother devotes to an only child." Id. at 114.

United had once required lessees to obtain insurance for its machines but for 25 years had waived the requirement. 110 F. Supp. at 321-22.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

3. Financing

When producers can raise capital more cheaply than their customers, financing considerations can also cause them to lease. Suppose that neither IBM nor Smithco has present cash for the computer -- IBM must borrow to produce it, and Smithco must borrow to buy it. If Smithco buys the computer, IBM can use the sales price to retire its loan, but Smithco's loan will remain outstanding. If Smithco leases the computer, IBM must continue its loan, but Smithco can simply pay an annual rental charge out of its revenue. If IBM, being a bigger and better known company, can borrow more cheaply than Smithco, then leasing is more efficient. In effect, IBM plays banker to Smithco; the arrangement makes sense because IBM, by monitoring the use of the leased machine, may know more about Smithco's business health than a bank would. n75
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n75 The use of leasing to replace borrowing is a standard chapter in finance texts. See R. BREALEY & S. MYERS, PRINCIPLES OF CORPORATE FINANCE 521-45 (1981); T. COPELAND & J.F. WESTON, FINANCIAL THEORY AND CORPORATE POLICY 536-58 (2d ed. 1983). This is consistent with the 1917 Shoe opinion:
The testimony also shows that the advantage of the leases was and is that manufacturers of not large means were able to obtain machinery which they were without capital to buy. They helped, indeed, the big and the little. One manufacturer whose output was 5,000 pairs of shoes a day testified that if his company had been compelled to buy outright the machinery necessary to equip its factory, it could not have developed as it had.
247 U.S. at 63; see also 222 F. at 352 (leasing "enabled manufacturers of small and moderate means to embark in manufacturing to an extent which would have been impossible for them, if they had been obliged to purchase machinery, because many machines are so expensive as to lock up capital and render it dead for practical purposes of financing shoe manufacturing."); Carroll, supra note 71, at A8, col. 1 ("Because IBM can borrow more cheaply than its much smaller rivals, it can offer customers very attractive leasing deals.").
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

 [*713]  4. Maintenance

In markets for durable goods, maintenance considerations can also cause the parties to lease. For complicated equipment like shoe machinery, the producer will often be the party best able to service the machine. Frequently, however, the producer and customer face a classic "moral hazard" problem: Customers buying repair services on a per-problem basis give the producer an incentive to charge a high price and fix only the symptoms of the problem. n76
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n76 On moral hazard issues generally, see E. RASMUSEN, supra note 7, at chs. 6 & 7. The problem is most acute in durable-goods markets because customers are less likely to return to the producer for repeat purchases. Hence, reputation considerations are less likely to eliminate the moral hazard problem described in this Section.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

By leasing the machine, the parties can reduce this moral hazard. The price of the repairs will no longer be an issue, and the lessor will have an incentive to fix the machine properly. In fact, if the rental charge increases with use (as in Shoe), the lessor also will want to fix it promptly. n77 Although the lessee has less incentive to maintain the product, the lessor can reserve its right to check maintenance and look for abuse. n78
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n77 See 110 F. Supp. at 302 ("And breakdowns of such machines should be promptly attended to, because, in the shoe manufacturing industry, stoppages are particularly costly."); id. at 322 ("Calls for this service [repairing breakdowns] are frequent: in some factories 2 to 10 men are needed every day."); id. ("In all respects, the service rendered by United is uniformly of excellent quality. It is promptly, efficiently, and courteously rendered."); see also 247 U.S. at 56 (1918 Shoe decision speaks glowingly of "a service force . . . estimated at 6,000 men, to repair immediately breaks or deterioration without extra charge. And these men are kept at convenient places to repair machines and replace worn-out parts, and depots of supplies are maintained."); id. at 64 ("The breakdown of some of these machines will in many of the factories block the entire flow of the work.") (quoting testimony of United's president).

n78 The United lease did so provide. 110 F. Supp. at 315. Note, too, that the court stated, "In fact, United has at all times assumed the burden of keeping its leased machinery in good order. It has made no separate charge to the lessee for such services, but has charged him for such parts as are required." Id. at 322. It is, in fact, much easier for customers to know that they have received new parts than proper service.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The parties also would have eliminated much of this moral hazard problem if the consumers could have bought service contracts in a competitive market for maintenance services. Judge Wyzanski apparently had this in mind when he ordered United to charge separately for service. n79 In the process, however, he was likely to have created the opposite moral hazard problem: Customers  [*714]  who can pay separately for the service contract gain an incentive to buy one of inefficiently low quality. True, customers have some incentive to buy a service contract of high quality because poor quality service will reduce their output and profit. Yet because customers pay rental fees based on output, part of the loss from poor quality repairs would fall on United. As a result of Judge Wyzanski's order, the lessee paid the full cost of repair under the service contract but shared with United the resulting benefits. It is possible that customers thereby gained the incentive to keep repair expenses at inefficiently low levels. Bundling the lease with the service contract can eliminate this problem by implicitly letting United and its customer split the cost of the service contract. Forced to unbundle its services and leasing, however, United was now less likely to offer the output-based leasing that its customers enjoyed. n80
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n79 The court's goal was the development of large independent repair companies, which in turn would free United's competitors from the need to offer service with their machines. See 110 F. Supp. at 325. The court conceded, however, that "no [service] system that has been suggested would be likely to be superior from a technological viewpoint. Shoe manufacturers, in general, are well satisfied with the technical service." Id. at 322. The court thus seemed more interested in rivalry than efficiency.

n80 Cf. C. KAYSEN, supra note 16, at 190 ("Since United is paid on performance, including high rate of output and dependability of production, the design of rugged, high performance machines is stimulated.").
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

This analysis is not exhaustive. One can reply, for instance, that United could have double-checked the quality of another firm's service by direct inspection. n81 A problem with judicially engineered "less restrictive alternatives," however, is that they are not always less expensive; in fact, the judge will have out-managed the company if they are. Some doubt judges' capacity on this score. Rather than exhaust this question, however, our aims here are simpler: to underline the complexity of principal-agent relations that lurk within apparently simple tasks like maintenance and to stress the consequent attraction of allowing firms the flexibility to adopt and adapt arrangements like leasing to cope with such complexities.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n81 See supra note 78 and accompanying text.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

5. Product Quality

The unobservability of product quality in some markets can also cause parties to lease. Suppose that (a) customers cannot determine quality until after they have bought the product, and (b) low quality is cheaper to produce than high quality. In markets of this sort, sellers will often produce only low quality, and customers will only pay a low price. For nondurable goods, this problem need not be severe -- the seller may decide to produce high quality at the outset to preserve its reputation and to charge a high price in the future. n82  [*715]  But in a durable-goods market, customers do not return for new purchases, and the seller has a lessened incentive to maintain its reputation.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n82 This argument is spelled out in Klein & Leffler, The Role of Market Forces in Assuring Contractual Performance, 89 J. POL. ECON. 615 (1981); see id. at 620. For a shorter explanation, see E. RASMUSEN, supra note 7, at 96-99.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Nonetheless, by leasing instead of selling, producers in these markets sustain an incentive to maintain product quality and firm reputation. Under leasing, customers make payments at regular intervals. They thus know that if a producer cheats them with low quality once, they can refuse to pay a high price thereafter, and that this threat can keep the producer honest. n83
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n83 This argument, unlike the Coase Conjecture, does not rely on different customers having different valuations of the product. But both arguments focus on the ability of a lease to transform a single long-term market into a series of short-term ones. Again, unlike the Coase Conjecture, the quality argument applies to both competitive and monopolized markets, and it implies that leasing can be good for both suppliers (who get higher prices) and customers (who get higher quality).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

* * * *

These explanations show how leasing can be efficient -- indeed, extremely valuable. n84 Similar considerations may also explain some  [*716]  of the more arcane clauses of particular leases. For example, we have mentioned that the United practice of making the rental charge increase with heavier use of the machine might have been designed to share risk. Using such a system of charges also leads, however, to other problems that can complicate the lease. Judge Wyzanski condemned United for its use of a "full capacity clause," which gave United the right to cancel the lease and recover the machine unless "the lessee shall use the leased machinery to its full capacity." n85 Yet United may have merely been trying to prevent customers from using extra United machines as cheap insurance against the possibility that the principal machine would break down. In essence, such customers may have been exploiting United's pay-by-use fee schedule.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n84 This list is far from exhaustive. Particular cases can reveal a host of other particular efficiency explanations. For example, leasing can make patent infringement more difficult by making it harder for the purchaser to take the product apart. The consensus today is that strong enforcement of patents is a good thing for consumers because it provides a valuable incentive for valuable innovation. Leasing can also be used to protect the lead-time monopoly returns on unpatented inventions. Whether the law ought to facilitate that monopoly pricing is considerably more controversial. See generally infra note 120.

Producers also might lease to internalize spillover effects. For example, most private clubs do not give their members transferable property rights in the club -- in effect, they lease the use of the club to their members. Were they to sell, members who transferred their membership to the "wrong" persons would depreciate the value of membership. Selling a membership to the future "club bore" spills over harm to all other members.

Similarly, leasing might aim to protect the capital value of a trademark. The father of one of the authors once plausibly (or at least illustratively) claimed that Rolls Royce sold its cars only to those with sufficient social standing and thus refused sales to some "undesirables" with the requisite cash. Because the cars were transferable, however, the scheme soon broke down. Had it adopted a lease-only policy, the company could have avoided the problem. Antitrust law accepted a similar justification for tying (rather than leasing) in United States v. Jerrold Elec. Corp., 187 F. Supp. 545 (E.D. Pa. 1960), aff'd per curiam, 365 U.S. 567 (1961) (defendant allowed to tie CATV components and service because cable failures would have reflected adversely on the infant cable industry).

By placing it in closer contact with those making daily use of its machines, leasing may also facilitate the lessor's research effort to improve its product. United claimed this advantage. See C. KAYSEN, supra note 16, at 171, 190.

n85 110 F. Supp. at 317. Wyzanski noted that:
Despite its express terms, the full capacity clause is not considered by United to have been violated unless the lessee fails to use the machine on work for which the machine is capable of being used, and instead performs such work by using a competitor's machine. In other words, it is not treated as a violation if the lessee fails to use United's machines because he performs the work by hand, or because he discontinues the type of operation for which the machine is capable of being used.
Id. at 320; see also supra note 27 and accompanying text.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

That problem arises precisely because United charged its lessees according to how heavily they used the machine n86 so that a lessee with low use incurred a relatively low fee. If a shoe company needed, say, one machine operating at all times, then United's schedule would tempt lessees to rent two machines: one for use and one as an emergency backup. If United charged a flat rental fee, it would not mind this strategy. But many customers might prefer the lower risk of a use-sensitive fee. n87 Because United did charge a rental fee that increased with output, it would be reluctant to offer free insurance of this kind. In fact, it might not even collect a profitable rent on the machine in use; the customer could rent a competitor's machine for normal production and United's machine for a backup. n88
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n86 See supra text accompanying note 72.

n87 See supra note 69 and accompanying text.

n88 A similar story could be told about a customer who wanted to keep one machine for normal use and an extra machine for "peak-load" use during periods of extraordinary demand. Sometimes customers legitimately need extra machines for this use, and United had special short-term leases for such customers. "United's practice is to furnish a machine temporarily and only for a limited period to take care of periods of peak production in a shoe factory." 110 F. Supp. at 321.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The full-capacity clause thus may have facilitated a mutually desirable allocation of risk by preventing opportunistic behavior by  [*717]  lessees. n89 Two facts support this efficiency interpretation of the fullcapacity clause and contradict the usual argument that the clause was nothing more than United's effort to buy a customer's promise not to deal with competitors. First, one of United's constituent companies used this lease clause before the original United merger -- presumably before it gained monopoly power. n90 But a firm lacking monopoly power has difficulty insisting on a simple exclusionary promise. Imagine your reaction, for instance, if a local supermarket demanded your promise that you never buy groceries from competing grocers. Second, United offered to waive the fullcapacity clause "for a 'reasonable time' if a lessee wishes to try out a competitive machine." n91 Willingness to waive the clause in just the situation where its exclusionary effect would be greatest suggests an exclusionary interpretation is incorrect.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n89 Compare Klein, Crawford & Alchian, Vertical Integration, Appropriable Rents, and the Competitive Contracting Process, 21 J.L. & ECON. 297, 297-98 (1978) with O. WILLIAMSON, MARKETS AND HIERARCHIES: ANALYSIS AND ANTITRUST IMPLICATIONS 26-30 (1975).

Two earlier Shoe courts excused the full-capacity clause with explanations consistent with our "free insurance" hypothesis. See 247 U.S. at 62 ("Without [the full-capacity clause] defendants say that as lessors, they would have no assurance of compensation for their machine."); 222 F. at 389 ("The lessor might well accept for the use of its machine a smaller royalty per pair, based upon the understanding that the machine would turn out as many pairs as its capacity permitted, than it could afford to accept without any such assurance.").

n90 "This clause was in the lease of the Consolidated Company prior to the formation of the United Company." 247 U.S. at 62.

n91 C. KAYSEN, supra note 16, at 70.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

D. Monopoly Reasons for Leasing

These efficiency explanations for leasing have no necessary link with monopoly although they apply to transactions in monopolized markets as well as in competitive markets. Antitrust errs if it deters such conduct, whether by a monopolist or by a competitive firm. In the case of two other motivations for leasing -- price discrimination and the Coase Conjecture -- the implications for antitrust policy are not so clear.

1. Price Discrimination

Price discrimination would appeal only to a firm with market power and the freedom to depart from cost-based pricing. Suppose a monopoly supplier faces two customers: a highly profitable manufacturer of boots and a marginal maker of sandals. The monopolist would prefer to charge each its willingness to pay. A policy of sales,  [*718]  however, will not accomplish this goal if the sandal maker can engage in arbitrage by buying machines low and reselling high to the boot manufacturer. By leasing rather than selling, the monopolist can retain control over the machines, prevent arbitrage, and thus succeed in price discrimination. n92
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n92 Victor Goldberg discusses this possibility and concludes that it might explain United's lease practice. See V. Goldberg, supra note 68, at 6-10. Certainly United's lease policy of charging by the number of shoes its machines produced, see supra note 72 and accompanying text, looks like a classic case of price discrimination -- unless the use charge perfectly mirrored depreciation from use.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

To note that monopolists may sometimes use leases to price-discriminate, however, fails to justify Shoe's ban on lease-only policies. Even if no efficiency reasons supported leasing, n93 price discrimination alone would not warrant such a ban. Judges could ban discriminatory leasing without banning all leasing. And notwithstanding the hostility that many courts have shown toward the practice, n94 it is far from clear that price discrimination necessarily harms anyone. Instead, a ban could cause the seller to reduce its output because it might give up selling to the customers (like the sandal maker) who pay the low price. An economist's typical conclusion is that "there may be a useful role for government regulation of discriminatory pricing, but . . . a flat ban could have adverse welfare consequences." n95 In this Article, therefore, we treat current hostility in antitrust law toward price discrimination generally as insufficiently justified to support Shoe's ban on lease-only policies.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n93 But see supra text accompanying notes 66-91.

n94 See Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 14-15 & n.23 (1984).

n95 Katz, The Welfare Effects of Third-Degree Price Discrimination in Intermediate Good Markets, 77 AM. ECON. REV. 154, 165 (1987). Katz lists the relevant recent literature, which is based on research from the 1930s. Id. at 167. See also Hausman & MacKie-Mason, Price Discrimination and Patent Policy, 19 RAND J. ECON. 253 (1988) (generally beneficial to permit price discrimination by monopolist that gained market power through a patented innovation); H. HOVENKAMP, supra note 41; supra note 69.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

2. The Coase Conjecture

The second motivation for leasing that might require antitrust regulation is the Coase Conjecture. According to the Conjecture, even if a durable-goods monopolist cannot charge a monopoly price for its sales, it may be able to do so through a lease. Because leases enable the producer to sell the nondurable stream of services generated by the durable good rather than the good itself, they enable the producer to earn monopoly rents. For all the attention economists have paid to the Coase Conjecture, however, few lawyers have  [*719]  noted its implications for government policy. With their suggestion that Coase supports Shoe, Posner and Easterbrook remain among the few who have tried to integrate the Coase literature into their analysis of antitrust law. n96 As we shall see, however, Coase's logic supports the Shoe rule only in a quite modified form -- if at all.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n96 See supra note 14 and accompanying text.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

III. FITTING SHOE TO COASE'S LAST

By Coase's logic, the monopolist can use a lease to obtain a monopoly instead of a competitive return -- an antitrust evil. n97 Consequently, the Shoe rule seems like sensible industrial policy. In fact, however, Coase's logic dictates a rule broader in some respects and narrower in others than the one in Shoe.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n97 See supra note 51.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

A. Broadening the Shoe Rule

United Shoe Machinery leased and did not sell its most important machines. As a remedy to this supposedly bad conduct, the court ordered it merely to supplement -- not replace -- its leasing policy with a sales option. n98 Apparently, the firm could have escaped liability by offering the sales option in the first place.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n98 110 F. Supp. at 349-51.

Kaysen had recommended that "United should be enjoined from marketing machines by any other method than sale." C. KAYSEN, supra note 16, at 275. But Judge Wyzanski decided otherwise after he read his wife his 100-plus page draft opinion, and he found himself unable to answer her question, "But why, if the terms are economically equivalent, shouldn't a customer, if he wants to lease, be able to lease?" See Wyzanski, John A. Sibley Lecture -- An Activist Judge: Mea Maxima Culpa. Apologia Pro Vita Mea., 7 GA. L. REV. 202, 213 (1973).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

According to the Coase story, however, a sales option ought not to rescue a lease option so long as a significant number of consumers actually lease rather than buy. n99 The crucial element of a monopoly-enhancing lease is that the monopolist has an incentive to maintain high prices in the future. If a significant number of highvaluation consumers lease rather than buy, the monopolist does retain this incentive. For if the monopolist cuts its sales price in the future, the high-valuation customers who lease at the high price will cancel their leases and buy at the low price. The high-priced leases thus guard against future sales discounts and eliminate the incentive for the high-valuation customers to withhold present patronage.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n99 In this respect the 1922 Supreme Court Shoe rule is more in accord with Coase's logic than is Judge Wyzanski's opinion. See supra note 20.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

 [*720]  Thus, Judge Wyzanski's remedy of requiring a sales option is an ineffective cure to the problem that Coase explained. Instead, Coase's logic dictates a broader remedy: a complete ban on leasing. Forced to sell and sell alone, the monopolist will be unable to charge its monopoly price.

This point has two implications, one soothing and one troubling. The soothing implication is that this broader remedy would save courts from having to specify exactly what price the monopolist ought to charge when it sells. A court requiring the company to sell and lease must say something about its relative prices. Otherwise the monopolist can simply charge a sales price so high that all consumers continue leasing. For the court, the obvious choice of sales price is the capitalized value of the monopoly rental charge. Even at that price, however, consumers would still prefer to lease. For as Coase described, if many consumers were to buy at this high price, their action would tempt the monopolist to lower the sales price in order to sell to lower-valuation buyers. When that occurred, the buyers would suffer a capital loss, but the lessees would not. Lest they suffer that capital loss, consumers would lease. Ironically, because all other consumers make the same choice and lease, the consumer who leases thereby helps ensure that the monopolist will continue to charge monopoly prices. Thus to ensure that any significant number of customers bought, the court would have to order that the sales price be more attractive than the lease price -- just what the government requested, and the court refused in the Shoe case. n100
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n100 110 F. Supp. at 350.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Ultimately, setting an appropriate sales price is not the sort of issue courts determine well. Even calculating the capitalized value of the rental charge may require more information than a court will have. And we have just shown that to make any difference, the sales price must be lower than this capitalized value. Deciding how much lower is the kind of question public utility commissions spend months deciding in rate-of-return hearings. Judge Wyzanski foresaw some of these administrative problems and responded with simple and unsatisfying bluster. n101 A simpler response -- with greater logical integrity -- would have been a complete ban on leasing.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n101 See id. at 351; see also id. at 349 (similar relief impossible "without turning United into a public utility, and the Court into a public utility commission"); C. KAYSEN, supra note 16, at 271-72, 277.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The troubling implication of a sales-only rule is the perverse incentive it creates for the monopolist. Courts that ban leasing may  [*721]  encourage the monopolist to engage in planned obsolescence by making goods less durable. n102 Courts heeding Coase thus must act more boldly than did Judge Wyzanski, but this boldness will simplify some matters only by inevitably complicating others.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n102 See supra note 7 and accompanying text; but cf. Malueg & Solow, On Requiring the Durable Goods Monopolist To Sell, 25 ECON. LETTERS 283 (1987) (simplified model tentatively suggests that a policy of requiring sales more often will lead to beneficial than to harmful effects).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

B. Five Limits on Coase's Logic

Coase's logical support for the Shoe rule also bounds it. His explanation is the exceptional case; it does not explain why, for example, firms lease short-term services: why Hertz rents cars or Sheraton rents rooms. Neither does it explain why firms lease durable goods in competitive markets. A firm trying to lease such a good at a high price would simply lose customers. It does not even explain why they lease in monopolized markets if, before introducing the lease-only policy, a firm had sold enough goods outright that it faces competition from its past customers. n103 But even if a good is durable and the seller a genuine monopolist, the following five qualifications show that a rule founded on Coase's justification would have to be focused still further. Indeed, they show that its proper range of concern not only excludes the facts of the Shoe case, but is in fact too narrow and intractable for courts to worry about at all.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n103 Competition from past customers is likely when the merger of existing sellers created the monopoly power. This competition can make mergers in such markets of less concern than otherwise. See Carlton & Gertner, supra note 3.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

1. Leasing is Innocuous When Demand is Constant

Coase and most later researchers n104 discussed the Conjecture in the context of markets with a fixed number of buyers. Yet such lumpy markets are scarce. The more usual case is one where new consumers regularly enter the market. A stream of new consumers, however, makes the Conjecture irrelevant. n105
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n104 With the exception of Conlisk, Gerstner & Sobel, Cyclic Pricing by a Durable Goods Monopolist, 99 Q.J. ECON. 489 (1984).

n105 The applicability of the Coase Conjecture is limited if "durable" goods eventually give rise to replacement sales. See Bond & Samuelson, supra note 8; Gul, Sonnenschein & Wilson, supra note 6; Suslow, supra note 9. In addition, the prospect of repeat sales may eliminate the viability of the Coase Conjecture. See Gaskins, Alcoa Revisited: The Welfare Implications of a Secondhand Market, 7 J. ECON. THEORY 254 (1974); Gul, Sonnenschein & Wilson, supra note 6.

This discussion assumes that the Low purchasers do not remain potential buyers forever. If they do, then the Lows will "pile up," and the seller may find it profitable to hold periodic discount sales. See generally Conlisk, Gerstner & Sobel, supra note 104. This strategy would work only if a large number of Highs did not anticipate this sale and found it profitable to withhold their purchases at the high price. To the extent that they wait for the discount sales, the Coasian dilemma reappears.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

 [*722]  To understand why the Conjecture plays no role in markets with new consumers, consider once again its logic. The monopolist would like to sell its durable goods to the Highs at a high price. As soon as it did so, though, it would have an incentive to lower its price and sell to the Lows. Because the Highs recognize this temptation, they wait for the discount sale and refuse to pay the initial high price. Rather than wait forever for the Highs to buy, the monopolist discounts the goods from the outset.

Fortunately for the monopolist and unfortunately for the public, this logic self-destructs if new consumers regularly enter the market. Suppose that new High and Low buyers appear regularly. If the monopolist were to use the Coasian price discrimination strategy (pick off the initial Highs at a high price, then sell cheaply to the Lows), it necessarily would also sell cheaply to the Highs who appeared later. The monopolist can sell to the initial-period Lows only by lowering the price, and if it lowers the price, it abandons the monopoly rents it would otherwise earn by charging a high price to the later Highs.

As a result, a durable-goods monopolist in a dynamic market will price its products by comparing the profits from more sales at a lower price with those from fewer sales at a higher price. It will price low at the outset only if (a) the present value of its future profits from selling at marginal cost to many buyers (both Highs and Lows) exceeds (b) the present value of its monopoly rents from selling at a higher price to fewer buyers (only Highs). Readers will recognize that tradeoff, however, as the standard tradeoff all monopolists face.

Consequently, if new buyers regularly enter and leave the market for a durable good, they transform the durable-goods monopolist's otherwise peculiar position into exactly the one all other monopolists face. Suppose the durable-goods monopolist sells in a market where the same mix of High and Low buyers appears each year, and buyers disappear if they do not buy. If -- following Coase -- the monopolist would prefer to sell fewer goods at a higher price in year 1, it necessarily will prefer the same strategy in each of the later years. If so, however, its threat to maintain high prices in subsequent years becomes credible. With that credibility, the Coase Conjecture disintegrates: if the monopolist loses by lowering its price after the initial Highs buy, those Highs gain nothing by waiting;  [*723]  accordingly, if the monopolist charges a high price from the start, the Highs will pay.

At root, the Coase Conjecture is an artifact of a model where new buyers fail to appear after year 1. That model may fit an economy that never grows. In economies that do grow, however, new manufacturers regularly buy durable equipment. Considerations of business cycles aside, the demand for equipment stays relatively constant. Suppose, therefore, that a producer monopolizes the market for a durable machine. Because of the steady queue of new buyers, the Coase Conjecture would not prevent selling the machines (notwithstanding their durability) at monopoly prices. Absent any pent-up demand for the equipment, the producer will face the same number of Highs in years 2 and 3 as in year 1. Necessarily, if it paid to hold prices high in year 1, it will pay in years 2, 3, and so forth. Everyone would have accepted the monopolist's threat to keep prices high after the year 1 Highs have bought, and the Coasian dilemma will disappear.

If the Coase Conjecture ever did describe a market, it probably describes one shaped by radical technological innovation. Recall that the Conjecture applies only if substantially more buyers appear in year 1 than in later years. That situation generally occurs only in new markets, and new markets generally emerge only as a result of technological change. For reasons we explain in the next Section, however, those industries where technological change creates an unusually large initial demand for a product are often precisely the industries where the goods rapidly become obsolete -- and Coase's point does not apply.

More generally, the point is this: With truly durable goods, demand is likely level; with lumpy one-period demand, the goods are seldom durable. The Coase situation is thus the exception rather than the rule. A sensible part of a plaintiff's case against a lease-only policy, therefore, would be proof that the market at issue fulfills the exceptional requirements of the Coase Conjecture: at the time they executed the leases in question, a one-time demand existed that the defendant and its customers did not expect would continue. Absent any evidence of this lumpy demand, courts ought to leave such behavior alone; absent any evidence of radically lumpy demand for shoe-making machines, the Shoe court should have left United alone. n106
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n106 But see 110 F. Supp. at 343; accord, Kaysen, Foreword to F. FISHER, J. McGOWAN & J. GREENWOOD, supra note 2, at xi (United's market was "nearly static").
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

 [*724]  2. Leasing is Innocuous if Products are Not Durable

At least two reasons exist why high-valuation consumers will often repeat their purchases through time: Either the item is not physically durable, or changing technology or tastes make the physically rugged thing obsolete.

a. Physical durability

No one leases carrots, matches, or listening live to a violinist. No one can. If the good is not durable at all, use consumes it and makes a sale inevitable. But even if a good is "somewhat but not very durable," then Coase's explanation does not apply. Consider beach towels and mats, which committed beachgoers can wear out quickly. Renting could not raise monopolist profits very much for a Santa Monica beach monopolist because beach fanatics (the consumers with a high valuation) soon will be back in the market. So some reason other than Coase's must motivate a monopolist that leases a somewhat but not very durable good. n107
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n107 For the beach mat example, convenience would be the obvious reason why occasional beach users rent rather than buy a beach mat at dawn and try to resell it over happy hour margaritas. See supra text accompanying note 68.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

b. Economic durability

Advancing technology can turn even physically durable goods into economic perishables. For this reason, no defendant should be condemned for leasing an apparently durable good if the item's economic life is short. Consider slide rules. As recently as 1972, these sturdy gadgets promised to last engineers a lifetime. But indestructibility does not a durable good make. Within a year, the market for slide rules disappeared. The calculators that replaced the slide rules in 1973 promised engineers savings in time and accuracy, and engineers eagerly paid $ 150 and up for machines that did no more than simple arithmetic. Big and solid, these machines seemed almost as enduring as their sleek and sliding predecessors. Once again, though, the machines were far less durable than they seemed. Technological change soon made these machines obsolete, and within a few years they were abandoned for the thirty-function, $ 30 programmables.

We ordinarily think of innovation in the context of better machinery in high tech industries. The principle, however, is the same for "soft" innovations like new songs or new car shapes. For such soft innovations, the producer has a monopoly on the new product,  [*725]  but only for a very short time before a still newer product enters the market. Hence, if perchance a creative agency leased rather than sold its short-lived musical or design creations, it is unlikely to be due to the reason Coase gives.

So antitrust ought not apply the Shoe rule in markets where physically robust goods in fact have short economic lives. This factor would not appear to affect the Shoe case, in which there were "no sudden changes in the style of machines." n108 But it qualifies the sweep of the rule in a way that Judge Wyzanski did not suggest. n109
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n108 110 F. Supp. at 343; accord Kaysen, Foreword to F. FISHER, J. McGOWAN & J. GREENWOOD, supra note 2, at xi (United's market was "nearly static" and "showed only modest technical progress" during a period of nearly thirty years); C. KAYSEN, supra note 16, at 184 ("Progress in shoe machinery has not been made by leaps and bounds; rather it is glacial in its character."). But see United Shoe, 222 F. at 360 (United claimed its current machines made obsolete the models of 1899).

n109 One perhaps might object that this discussion assumes that technological change, like time, is impassively inevitable. In truth, of course, it is not -- a monopolist may have the power to control the pace of innovation. Yet this fact leads to ambiguous policy conclusions. According to Swan, a monopolist that can commit to future prices or that can lease will innovate at an efficient rate. See Swan, Optimum Durability, Second-Hand Markets, and Planned Obsolescence, 80 J. POL. ECON. 575 (1972). Hence, permitting a durable-goods monopolist to lease may ensure that the monopolist does not inefficiently alter the pace of technological change; unfortunately, if the Coase Conjecture applies, it does so at the cost of monopoly pricing. On the other hand, a monopolist that cannot lease (one subject to the Shoe rule) faces a distorting incentive to change products routinely in order to transform durable into less durable goods. As a result, if one cost of permitting durable-goods leasing is monopoly pricing, one cost of banning it may be planned obsolescence. Cf. Bulow, Monopolists, supra note 7 (if the Coase Conjecture held, a monopolist would plan obsolescence for its products to transform otherwise durable goods into perishable ones); Bulow, Planned Obsolescence, supra note 7 (same).

In theory, judges could review innovation by monopolists to see if it resulted from this inefficient desire to realize monopoly profits from durables. But antitrust in general has been extremely chary of regulating innovation, for fear of mistaking good innovation for bad and thereby dampening the process that Schumpeter praised and that has gained particular national prestige in the last decade. See Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263 (2d Cir. 1979), cert. denied, 444 U.S. 1093 (1980); Sidak, Debunking Predatory Innovation, 83 COLUM. L. REV. 1121 (1983). As a result, amplifying the Shoe ban on lease-only monopoly with an additional ban on "inefficient" technological change is not likely to be a satisfactory solution to the Coase problem.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

3. Leasing is Innocuous if Terms are Long

Coase pointed out that his logic dictates that the monopolist adopt short-term leases. n110 Short-term leases insure that the monopolist's intent to maintain a high rental charge is credible. If the monopolist uses a one-month lease and then increases output and  [*726]  drives down the price, it will suffer the consequences a month later when it tries to renew the leases. If the lease has a ten-year term, then the monopolist has locked in the lease charge for ten years and has much more incentive to flood the market after the lease is signed. A hundred-year lease is nearly an outright sale. Leasing avoids the Coase Conjecture only if it reduces the monopolist's temptation to increase output and thereby convinces customers that output will remain low (and prices high). Thus, from the point of view of the Coase Conjecture, leasing works best when the term of the lease is very short and best of all when the customer can cancel the lease at any time. n111 Consequently, antitrust should discount the Coase Conjecture as an explanation of -- and reason to condemn -- leases that are long.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n110 Coase, supra note 3, at 145 ("Another way in which essentially the same result could be obtained would be for the landowner not to sell the land but to lease it for relatively short periods of time.") (emphasis added).

n111 The point that the Coase Conjecture depends on the time interval between price changes (and, thus, on the term of leases) is developed in detail in Stokey, supra note 10.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

In the context of Shoe, this point is vital. The Shoe court complained pointedly about the long term of United's ten-year leases. n112 The case also makes clear, however, that United's customers desired that long term. Before its antitrust challenges, United usually leased its machines for seventeen-year terms. After an initial federal challenge, United shortened this term to seven years. At its customers' requests, however, it lengthened this term to ten years. n113
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n112 110 F. Supp. at 324 ("The 10-year term is a long commitment."); see also id. at 340, 343-44.

n113 Id. at 319.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

There are many efficiency reasons why suppliers and customers may prefer their relation to be on a long-term rather than a short-term basis. To guarantee a set price or the assured availability and quality of perishable goods, customers commonly enter long-term contracts with sellers willing to trade a price break for a predictable market. n114 Customers and suppliers of durable goods can likewise opt for a long-term lease when other considerations n115 lead them to reject an outright sale. Thus, the demand of United's customers for leases with long terms is unremarkable -- unless one entertains Coase's explanation for United's leasing.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n114 For listings and discussion of the relevant literature, see Goldberg & Erickson, Quantity and Price Adjustment in Long-Term Contracts: A Case Study of Petroleum Coke, 30 J.L. & ECON. 369 (1987); Joskow, Price Adjustment in Long-Term Contracts: The Case of Coal, 31 J.L. & ECON. 47 (1988); Polinsky, Fixed Price vs. Spot Price Contracts: A Study in Risk Allocation, 3 J.L. ECON. & ORG'N 27 (1987).

n115 See supra notes 66-91 and accompanying text.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

These facts of the case contradict the Coase account if we assume that United would never consent to customers' requests that  [*727]  eliminated its monopoly profit. Instead, United should have wanted a short-term lease to convince its customers that it would not flood the market at discount prices. Conversely, if the customers wanted cheaper machines and somehow succeeded in forcing United to give them long leases, then United would have been forced "in the twinkling of an eye" to lower its prices. n116
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n116 Victor Goldberg also infers from the length of the leases that the Coase Conjecture is not a plausible explanation of United's leasing policy. See V. Goldberg, supra note 68, at 3-4.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

As a result, were the Coase Conjecture at work, the Shoe remedy would be completely backwards. The court ordered that the lease term be cut in half. n117 If Coase's explanation applied, this remedy would have enhanced rather than destroyed the monopolist's ability to charge a monopoly price. Moreover, this part of the remedy also contradicts the court's insistence that United offer its machines for sale. Because a sale is simply an infinitely long lease, the court, in effect, told United Shoe that it could have short-term leases and infinite-term leases but nothing in between. This remedy could accomplish nothing but consumer injury.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n117 110 F. Supp. at 349, 352. Other provisions of the court remedy effectively may have reduced the lease term to one year. See id. at 352.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

4. Leasing is Innocuous When a Monopolist Could Engage in Static Price Discrimination

Suppose a monopolist has a mechanism besides leasing that enables it to engage in static price discrimination: it can simultaneously charge different consumers according to their differing willingnesses to pay. If the monopolist also chooses to lease, Coase's explanation cannot be the reason why. A monopolist engaged in present and effective price discrimination sells to the Lows at a discount price today. It has no reason to cut prices tomorrow. The Highs thus have no reason to delay purchases. Coase's explanation cannot be the reason that such a monopolist leases, and antitrust should not intrude.

Recall our earlier example about the marginal sandal maker and the profitable boot manufacturer. n118 If United faces only these two customers, it can stop the arbitrage that defeats its price discrimination scheme by altering the basic machine so that one version can make only sandals and the other version only boots. In this way, it can charge a high price to the boot manufacturer and a low price to the sandal maker without fear that the latter will resell  [*728]  machines to the former. But this accomplishment of static price discrimination erases United's problem with sustaining a high price to the boot manufacturer. If United still leases, the reason must not be Coase's.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n118 See supra text accompanying note 92.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

It would be sensible for antitrust to intrude if a durable-goods monopolist relied upon leasing as its only way to price-discriminate, were this easily done. We say this despite our earlier observation n119 that there exists no theoretical warrant generally to ban all leasing that aims to accomplish price discrimination. Suppose the monopolist must lease to implement static price discrimination (perhaps because it must meter use of the product). By this assumption, the monopolist must lease to discriminate between users; without leasing, it will have no alternative other than to sell at a single price. Yet the monopolist will face the problem that Coase describes if it sells at a single price. In this situation, then, antitrust can force the price down to the competitive level by banning leasing. This result contrasts with the usual effect of banning price discrimination which is a single price at the monopoly level. Whether the move from price discrimination to a monopoly price is good is unclear, but the move to a competitive price is indisputably good. Hence, if leasing is necessary to price-discriminate in the market for a durable good, the use of price discrimination should be no excuse for leasing.
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n119 See supra note 95 and accompanying text.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Especially when so qualified, this fourth limitation sweeps less broadly than the previous three. Effective price discrimination is generally difficult and therefore not the norm because a monopolist can rarely prevent the arbitrage that creates a nondiscriminatory secondary market. Nonetheless, institutional peculiarities sometimes do permit sellers with some market power to employ price discrimination. If a firm leases in these situations, Coase's explanation is not the right one. Antitrust again should stay its hand.

5. Leasing is Innocuous When Patents Entitle a Monopolist to Its Monopoly Profit

Patents are legal monopolies designed to encourage innovation by allowing the innovator a price above production cost. The premise behind patent law is that the reward of monopoly power is appropriate because consumers would have no surplus at all had the inventor not created the innovation in the first place. But the Coase Conjecture shows that without leasing, an innovator in a durable-goods  [*729]  market could not effectively use its patent to earn profits as a reward for innovation. Hence, a rule against lease-only contracts would eliminate the patent incentive in markets for durable goods. Antitrust law would thwart patent law. n120
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n120 This argument differs from another liked better by economists than lawyers: antitrust law should encourage all innovation, patentable or unpatentable. The other argument says that if a firm innovates, but does not patent its innovation, it should still be able to use a lease-only policy during the period before imitation occurs to allow temporary monopoly profits as an incentive for innovation. For a similar argument as to why price discrimination should always be allowed in an innovative market, see Hausman & MacKie-Mason, supra note 95.

Judge Wyzanski, however, considered and rejected this economist's argument:
To this defense the shortest answer is that the law does not allow an enterprise that maintains control of a market through practices not economically inevitable, to justify that control because of its supposed social advantage. . . . It is for Congress, not for private interests, to determine whether a monopoly, not compelled by circumstances, is advantageous.
110 F. Supp. at 345 (citation omitted). Judge Wyzanski seemed to mean to include judges in his reference to "private interests." Cf. Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 109 S. Ct. 971 (1989) (preempting effort by state government to offer protection for an unpatentable innovation); Fashion Originators' Guild of Am. v. FTC, 312 U.S. 457 (1941) (outlawing boycott effort to protect unpatented fashion creations).
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

C. Shoe: A Wagging Dog

The Coase Conjecture supports the Shoe rule, but only in a greatly altered form. The case raises a genuine antitrust problem: a monopolist that injures the public by insisting on a policy of renting. Coase's logic shows that the only effective way to combat the problem is to forbid leasing altogether, not simply to order the monopolist to offer a sales option. This rule is simple, but it can give firms the unfortunate incentive to make flimsy products so that their sales more nearly resemble the forbidden leases. More importantly, antitrust errs if it responds by banning all monopoly leases because this form of transaction often offers great benefits without any threat at all.

We have outlined the ways in which antitrust law should tailor any ban on lease-only policies. Having done so, we doubt that a court should adopt that ban. Our analysis shows that a sensible prohibition must make dichotomies of three continua. First, the rule should permit monopolists to lease "somewhat but not very durable" goods, but not durable goods. Second, the rule should permit them to lease when technological change is rapid, but not when it is slow. Third, the rule should permit them to lease for long terms, but not for short. Moreover, the analysis shows that courts should recognize further exceptions in the presence of "effective"  [*730]  price discrimination and when monopoly is due to (and not just accompanied by) patents.

Theorists are comfortable with vague terms like "durable," "rapid," "long," "effective," and "due to." But courts must assess particular facts and conclude either "liable" or "not liable." More importantly, judges must invent future economic policy by deciding past antitrust disputes. They rightfully worry that the rules they craft might channel corporate conduct in ways that, on balance, will hurt consumers. The inevitable arbitrariness of a particular solution is not, of course, a reason automatically to reject it. n121 But arbitrariness does not just look bad. It also risks deterring valuable conduct. Given the frequent benefits from leasing, rational decision making advises that we should ignore its potential evils if they are too inconsequential for note. n122
 
- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n121 Law routinely confronts such problems and pushes on, with bravery or modesty or resignation. Antitrust doctrine offers its share of such instances. See United States v. Aluminum Co. of Am., 148 F.2d 416 (2d Cir. 1945) (the famous "Rule of Hand" for translating market share statistics into monopoly power conclusions: 30% means no power; 60% is maybe; 90% is certainly monopoly power); Wiley, Reciprocal Altruism as a Felony: Antitrust and the Prisoner's Dilemma, 86 MICH. L. REV. 1906, 1928 (1988) (merger law and federal Merger Guidelines depend on numerically precise, but analytically arbitrary, thresholds).

n122 Easterbrook makes an argument of this form about antitrust treatment of predatory pricing. Easterbrook, Predatory Strategies and Counterstrategies, 48 U. CHI. L. REV. 263 (1981). Indeed, Judge Wyzanski himself was quite eloquent about the need for judges to be modest about their ability to transform economic theory into legal rules. See 110 F. Supp. at 347-48.

Our conclusions here apply only to the support that Coase offered for the full breadth of the Shoe rule and not for the far more limited ban on exclusionary conduct that we discuss supra in note 37.
 
- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

We view the monopolist that leases durable goods in this light. Leasing usually is desirable. Only quite exceptionally will it facilitate monopoly pricing. Therefore, only with the loss of much wheat can antitrust cast out the chaff. Better that the Sherman Act enter the 1990s with more fiber in its diet.

CONCLUSION

* The Shoe case bans lease-only policies by monopolists. But the court erred in believing that monopoly pricing could explain United Shoe Machinery's complex of leasing policies. At best, this explanation only accounts for a few details of the case. The bulk of the company's conduct seems simply efficient -- thus suggesting that the Shoe decision was wrong and its later precedential consequences pernicious.

 [*731]  As Posner correctly noted, the Coase Conjecture might seem to make some sense of Shoe's ban on monopoly leasing -- under some circumstances. Indeed, the Conjecture even suggests that the Shoe rule may have been too narrow. At the same time, however, the Conjecture dictates that the Shoe rule be confined in other ways that the opinion did not suggest and that are unacceptably costly to accomplish. Courts would do well to accept that Coase describes a problem that is real. But they would also do well to accept it as a problem not worth solving.