5. Price Discrimination and Vertical Agreements

The antitrust laws still forbid price discrimination and vertical business agreements (merger, resale price maintenance) that may tend to reduce competition substantially. Price discrimination can be illegal under section 2 of the Clayton Act (1914) as amended by the Robinson-Patman Act (1936). Mergers can be illegal under section 7 of the Clayton Act. Tying contracts and other restrictive agreements can be illegal under section 3 of the Clayton Act or under section 5 of the Federal Trade Commission Act (1914).

Price Discrimination

Price discrimination is the practice of selling some homogeneous product—a good of like grade and quality—to different buyers at different prices. For instance, if a firm sells homogeneous salt to different buyers at different prices, the firm has price discriminated. The price difference is itself the price discrimination, and it can be illegal (except under certain conditions outlined below) when it may substantially lessen competition or tend to create a monopoly.

In antitrust practice, the phrase “may substantially lessen competition” has come to mean that competition is reduced—and the law violated—whenever there is some adverse effect, or probable adverse effect, on other business organizations in the market. In the salt-selling example, the price discrimination may tend to adversely effect some rival salt manufacturer who loses sales and thus profits, or it may tend to injure wholesale buyers who pay the higher prices and are in competition with low-paying buyers. The seller can be found guilty in either instance, and buyers who knowingly receive illegal price discriminations can also be found guilty under the law.

The fundamental difficulty with a law that prohibits price discrimination is that it tends to treat any adverse effect upon rival firms as a reduction in competition that can violate the law. But this treatment of competition is an example of the classic error in antitrust economics. Price reductions are an essential part of any competitive process, and so is the movement of resources from higher-cost sellers to lower-cost sellers. If consumer-buyers tend to purchase more from low-cost sellers, then it is entirely appropriate that high-cost sellers lose sales or have their business adversely affected. To interfere with this process and to prosecute the firms with the lower prices—and it is only the lower prices that threaten competitors—is blatantly protectionist of the existing market structure of suppliers.

Some might argue that this criticism is too severe because sellers accused of price discrimination can attempt to demonstrate, in their defense, that they have price discriminated in good faith in order to meet competition (from some rival seller, presumably) or that the price discrimination can be fully justified by specific cost savings. In practice, however, these so-called absolute defenses have proven unsatisfactory. It is inherently unclear when price reductions are in good faith, and aggressive competitors often attempt to beat, not meet, prices charged by rivals. Further, the cost-savings defense is all but illusory since it requires a level of technical precision in cost accounting, especially in accounting overhead costs, that may simply be impossible; specific price reductions by multi-product companies on specific products can rarely be cost justified to the legal satisfaction of the FTC or the courts.¹ Thus, the bulk of the firms indicted for price discrimination—that is, for price competition—have failed to defend themselves successfully; they have lost or abandoned their cases, and then they have raised their prices to comply with the law.

There is now little professional debate over the intent and effect of the price-discrimination law: its purpose has clearly been to reduce price competition and to protect high-cost, high-priced businesses from the resource-reallocation process. Like minimum-wage laws, agricultural price supports, and National Recovery Act codes during the Great Depression, the Robinson-Patman Act was depression legislation aimed at reducing the rigors of the market by restricting price competition.² Presumably, the justification claimed for such a law in the 1930s is no longer relevant, if it ever was. Today the law’s only effect is to stifle the competitive market process.

The Borden Case

The Borden evaporated milk case is a classic example of the irrationalities associated with attempting to enforce a law against price discrimination. In 1958, the Borden Company was indicted by the Federal Trade Commission for selling evaporated milk of like grade and quality to different buyers at different prices. Borden charged a lower price for milk that it packed and sold to private-label customers than it charged for its own Borden brand (“Elsie”) milk. Since the milk in both instances was chemically the same, the FTC charged that the price differences amounted to price discrimination in violation of the law.³

The milk at the factory may well have been the same, but consumer perception of the milk at retail was demonstrably not the same. Consumers were willing to pay more for the Borden brand of evaporated milk than for milk packed by Borden but sold under various private labels. Perhaps consumers were willing to pay more because the Borden Company had established a substantial reputation for high-quality products, which generations of consumers had come to rely on. For example, Borden carefully controlled the shelf life of its own brand of milk, whereas its responsibilities for private-label milk ended when the milk was packed and sold. In addition, some of Borden’s expenses, such as advertising, transportation, and labels, did not apply to its private-label milk, and this may have made it possible for Borden to charge lower prices to the private-label distributors. In short, there were both demand and cost differences with respect to the different brands of evaporated milk that could easily have rationalized the general differences in the prices of the products.

Even more important, however, the lower prices Borden charged to private-label distributors involved no injury to any of the parties involved: not to the private-label distributors themselves, who willingly purchased the milk from Borden; not to the customers of the private-label milk, who, presumably, bought a quality milk product at a lower price; and not to Borden’s own customers of its “premium” evaporated milk, who could have switched to cheaper milks at any time but did not. There was never any question of monopoly in private-label evaporated-milk production, since Borden never did more than 11 percent of the Midwest private-label packing, a share it had legitimately gained because of the locational advantages of its creameries.

The real issue in the FTC’s long harassment of Borden hinged, as it turned out, on the fact that some smaller, independent creameries in the Midwest had lost some private-label business and a few had even gone out of business. There was nothing in the FTC’s records to indicate that the Borden Company was involved directly in their demise; indeed, some of these creameries had disappeared prior to Borden’s entry into private-label milk packing. And yet, it is perfectly clear from the 1966 FTC decision against Borden that the loss of these independent creameries was the tendency to lessen competition that had concerned the FTC. The long legal harassment concerned Borden’s ability to provide economic advantages to willing customers and the inability—for reasons unrelated to Borden—of some of its rivals to perform in a similar manner. Thus, the thrust of the enforcement of the anti-price discrimination law was purely protectionist of an inefficient market structure of firms. The case against Borden was ultimately dismissed in 1967,⁴ but the meaning of the case and the decades of FTC enforcement of Robinson-Patman remain clear beyond all doubt: high-cost rivals are to be protected in the name of preserving competition.

The new direction in antitrust policy is, literally, not to actively enforce Robinson-Patman. Only persistent discrimination that would result in monopoly would, presumably, now concern the FTC. This is an excellent development in the administration of the antitrust laws, but there is no guarantee that it will be permanent. The case for antitrust repeal is, in fact, at its strongest with respect to the Robinson-Patman Act.⁵ A law against price discrimination, which prosecutes successful firms in the name of preserving competition in the “public interest,” has no theoretical or empirical support.

Tying Agreements

Tying agreements, such as territorial restrictions, full-line forcing, and tie-in sales, are voluntary contractual agreements between the sellers and buyers of products or services that typically restrict the activities of buyers in certain ways. For example, buyers might sign an agreement to purchase good or service X on condition that they also purchase good or service Y from the same seller. Or a territorial restriction in a contract might forbid some distributor of a product from selling the product in the territory of another distributor. Tying agreements on the sale or lease of shoe machines might include a clause restricting service on the machines. A manufacturer might lease a copier on condition that the lessee use the ink or paper supplied by the copier manufacturer or some subsidiary. Finally, a maker of brand-name blue jeans might attempt to restrict sales to certain distributors or to fix the minimum resale price of the jeans through contract.

The older, general consensus was that these restrictive practices could injure competition and final consumers and should be prohibited when any substantial volume of business was involved. The courts, up to 1977, strongly supported this consensus. In the years following, however, professional opinion on some restrictive practices shifted markedly. The newer view holds that it is not immediately obvious why a manufacturer would want to injure its own distributors or the final customers of its own product. Nor is it immediately obvious how purely vertical business restraints could lead to any horizontal output restriction and any higher market price.

It is possible that certain vertical restrictive agreements might only be an attempt to price discriminate, or preserve goodwill, or shift certain business risks, or financially strengthen certain distributors, or curtail inefficient “free riding” activity.⁶ A manufacturer of personal computers, for example, might want distributors to provide considerable presale information or post-sale service. In the absence of some restrictive agreement, customers might decide to “free ride” off the information provided by full-service distributors and then purchase their equipment from discounters. The resulting intrabrand competition might ultimately force the full-service, authorized dealers to drop the expensive presale information, which could hurt the manufacturer in interbrand competition. A restrictive agreement between the manufacturer and its distributors that territorially restricts those distributors or protects dealer profit margins through resale price-maintenance agreements, could remedy the situation and allow more efficient rivalry with other computer manufacturers.

The Sylvania Case

The economic rationale for restrictive tying agreements was finally recognized by the Supreme Court in 1977 in the Sylvania case.⁷ Sylvania, a relatively small manufacturer of television sets, had been sued under the antitrust laws by one of its distributors, Continental, for preventing Continental from establishing a new distributorship in Sacramento, California, where Sylvania had another authorized dealer. Sylvania argued that any additional intrabrand competition would have weakened both the competing dealerships and Sylvania’s ability to compete interbrand with stronger rival manufacturers and distributors, such as Sears and Zenith. Since Sylvania was a relatively small manufacturer attempting to hold on to a declining market share, and not some near-monopolist about to crush all its competition, the Supreme Court accepted this particular dealer restriction as reasonable. And although a rule-of-reason approach to restrictive agreements is not entirely satisfactory, the repudiation of per se illegality in Sylvania was certainly a movement in the right direction; that is, toward per se legality.

Resale Price-Maintenance Agreements

Resale price-maintenance agreements—vertical agreements restricting price—still remain illegal per se, even though the economic case for permitting them is persuasive.⁸

Most of the distaste for resale price maintenance goes back to the 1930s depression and the years immediately following, when the so-called fair-trade laws existed. The fair-trade laws legalized resale price-maintenance contracts by exempting them from federal antitrust regulation. However, these laws often went well beyond simply permitting restrictive vertical price agreements between willing buyers and sellers. The notorious non-signer clauses provided that any retailer that refused to sign a fair-trade contract with a manufacturer could, nonetheless, be legally bound by the terms of agreements signed by others!⁹

Depression policymakers were extremely hostile toward price competition, believing it to be one of the major reasons for the prolonged economic stagnation of the 1930s. Chain-store taxes and non-signer clauses to limit price reductions were only two examples of that hostility. Needless to say, modern proponents of free trade do not support legally restrictive non-signer clauses. They hold only that resale price-maintenance agreements should be exempted from the antitrust law. It would then be up to manufacturers and distributors to make such agreements voluntarily, if they so desired; it should not be the function of government to prohibit such contracts or coerce any firm into them.

Vertical Merger Agreements

The ultimate vertical “restrictive” agreement between a manufacturer and a distributor is a vertical merger. A shoe manufacturer, for example, that purchases a retail shoe distributor could certainly proceed to fix resale prices in its own stores. A vertically integrated manufacturer might order its wholly-owned retailer to exclude the shoes of a manufacturing competitor. A shoe manufacturer could purchase a leather supplier and either foreclose leather supplies to a rival or direct that the leather be sold at a higher price, in order to squeeze the rival between high-input costs and low shoe prices at retail.

The arguments that such activities provide a rationale for antitrust policy is weak and unconvincing. The competitive market process cannot be injured by any of them. Rival shoe manufacturers, excluded from some retail outlets, would not be excluded from the shoe market; presumably there are other retail outlets, and more could always be created. Higher prices for leather in an openly competitive leather market would only mean lower leather sales and lower profits. And the selling of less leather—or fewer shoes—can in no way be ultimately profitable to the larger, vertically integrated company.

There is, of course, one development that may tend to exclude rival sellers: successful vertical integration that results in improved efficiency and lower costs and prices. Indeed, the only mergers or integrations that ever threaten rivals are mergers in which the merged firms benefit from integration economies and intend to pass along some of the benefits to consumers in the form of lower prices or improved services. Such mergers might well induce nonintegrated firms to plan similar cost-saving integrations, and, from a consumer perspective, the sooner such integrations occur, the better. To prohibit these mergers, in the so-called public interest, would be the height of economic irrationality.

The Brown Shoe Case

The Brown Shoe case of 1962 nicely capsulizes all that is wrong with legal regulation of vertical integration.¹⁰ Brown Shoe, a manufacturer of shoes, bought the Kinney chain of retail shoe stores in 1956. By the court’s own admission, the merger would have allowed Brown to realize certain economies and efficiencies, such as faster style changes and lower shoe prices, which it might have been able to pass along to shoe customers. Such a development would have put competitive pressure on nonintegrated shoe manufacturers and retailers and encouraged them to vertically integrate, too. But this trend toward concentration—already evident in the shoe industry, according to the court—was allegedly destructive of competition and certainly contrary to the congressional intention. Thus, despite its obvious consumer benefits, the merger was declared illegal and Brown was ordered to divest itself of Kinney.

From the standpoint of consumer welfare, there was absolutely no reason for the judgment against Brown Shoe. Brown was a relatively small manufacturer of shoes in 1956, with 4 percent of domestic output, and Kinney owned only 845 retail outlets out of an industry total of more than 70,000. The shoe manufacturing and shoe retailing markets were easy for new firms to enter. Concentration in the shoe industry was not increasing, despite the court’s insinuations to the contrary. And finally, the court’s treatment of business efficiency as an exclusionary restraint of trade stands antitrust precisely on its head. If Brown Shoe was not the worst decision in antitrust history (there is, after all, a lot of competition),¹¹ it certainly takes a high rank.

Conclusions

Antitrust has come a long way since Borden and Brown Shoe. The dominant opinion today is that price discrimination and vertical agreements do not generally present any serious threat to consumer welfare and that such activity ought not to be legally restrained because of some adverse effect on rival sellers. Antitrust critics agree, of course, but some would go further to abolish the Clayton Act altogether, to be sure that such abominations as Borden and Brown Shoe never occur again.¹²

---

¹ Herbert F. Taggard, Cost Justification (Ann Arbor: School of Business Administration, University of Michigan, 1959).

²The Robinson-Patman Act (1936) was reportedly drafted by the U.S. Wholesale Grocers Association. See Richard Caves, American Industry: Structure, Conduct, Performance, 2d ed. (Englewood Cliffs, N.J.: Prentice-Hall, 1967), p. 86.

³In the Matter of the Borden Company, 381 FTC 130 (1958).

⁴Borden Company v. FTC, 381 F. 2nd. 175 (1967).

⁵Wesley J. Liebeler, “The Robinson-Patman Act: Let’s Repeal It!” Antitrust Law Journal 44 (April 1976): 18–43.

⁶See, for example, the discussion in Richard A. Posner, Antitrust Law: An Economic Perspective (Chicago: University of Chicago Press, 1976), pp. 171–84. See also idem, “The Next Step in the Antitrust Treatment of Restricted Distribution: Per Se Legality,” University of Chicago Law Review 48 (Winter 1981): 6–26. See also Howard P. Marvel and Stephen McCafferty, “The Welfare Effects of Resale Price Maintenance,” Journal of Law and Economics 28 (May 1985): 363–79.

⁷Continental T.V. Inc. v. GTE Sylvania, Inc., 433 U.S. 36 (1977).

⁸Terry Calvani and James Langenfeld, “An Overview of the Current Debate on Resale Price Maintenance,” Contemporary Policy Issues 3 (Spring 1985): 1–8. See also Thomas R. Overstreet, Jr., Resale Price Maintenance: Economic Theories and Evidence, Bureau of Economics Staff Report to the Federal Trade Commission (November 1983). For a dissenting view, see Robert Pitofsky, “In Defense of Discounters: The No-Frills Case for a Per Se Rule Against Vertical Price Fixing,” Georgetown Law Review 71 (December 1983): 1487–95.

⁹The non-signers clause was declared constitutional, in Illinois, by the Supreme Court in Old Dearborn Distribution Co. v. Seagram Distillers, Corp., 299 U.S. 183 (1936).

¹⁰United States v. Brown Shoe Company, 370 U.S. 294 (1962).

¹¹Ward Bowman nominates Utah Pie Company v. Continental Baking Company, 386 U.S. 685 (1967). See Ward Bowman, “Restraint of Trade by the Supreme Court: The Utah Pie Case,” Yale Law Journal 77 (November 1967): 70–85.

¹²Unfortunately, the exceedingly narrow approach to defining the relevant market in Brown Shoe may have returned. See FTC v. Staples, Inc., 90 F. Supp. at 1075 (1997).