Marks, morals, and markets.

• Author: Sheff, Jeremy N.
• Position: IV. Toward a Contractualist Theory of Trademark Law through Conclusion, with footnotes, p. 797-815

1. Trademark as Promise (144)

   The contractualist theory of producer-consumer relationships in trademark law offers a highly attractive justification of core trademark doctrine that is at least as convincing as that of consequentialist theory, if not more so. Consider the classic case of infringement by passing off (145) (or what one might refer to today as point-of-sale confusion as to source). (146) What would a contractualist consider wrongful about one producer using a trademark on his goods that confuses consumers into thinking the goods actually originated with another producer, and how might the contractualist view differ from the consequentialist view on this point?

   The consequentialist, as noted above, explains this problem in terms of search costs. (147) If passing off were permitted--that is, if a mark used as a source identifier by one producer could be used by that producer's competitors on the competitors' goods--then consumers could be misled about the unobservable qualities of the products to which the mark is affixed. (148) A consumer might buy a shoddy widget from Producer A thinking he was buying a quality widget from Producer B, suffering injury in the amount of the value attributable to the difference in quality. Moreover, because producers know more about the unobservable qualities of their products than consumers do, a world in which passing off is permitted is one in which consumers would have to undertake their own search for relevant information to avoid the kind of injury that would result in the widget example, raising the transaction costs associated with gathering and disseminating that information and lowering aggregate welfare as a result.

   The problem with this analysis is that it fails to account for an important aspect of infringement doctrine: the relevance of product quality. Under the Lanham Act, infringement includes a use of a trademark that is likely to cause confusion. (149) The Second Circuit--one of the most active (150) and authoritative courts in trademark law--has thoroughly explored the relevance of comparative product quality to the question of likely confusion, and its conclusion is difficult to square with a search costs theory. In the Second Circuit, the more similar the quality of the defendant's goods to that of the plaintiff's goods, the more likely the defendant will be held liable for infringement. (151) Several other circuits have adopted the Second Circuit's approach, with similar disregard for search costs. (152) While these courts recognize that divergent quality may suggest an injury of greater magnitude where confusion exists despite that divergence, they (understandably) consider such an inference relevant only to the availability of particular remedies, not to the determination of liability. (153)

   This treatment of product quality is precisely the opposite of what we would expect if we believed that trademark infringement liability was designed to reduce consumers' search costs. Obviously, not all uses of a trademark by a competitor of the mark's owner will mislead as to the unobservable qualities of the competitor's products that are relevant to the consumer's search. If trademarks really are about efficiently conveying information about such qualities to consumers, we should encourage, or at least excuse, the use of a well-known trademark by someone other than its first user where such use will efficiently convey accurate information about the user's goods to consumers. (154) Thus, the more similar the products are with respect to their unobservable qualities, the weaker the case for liability under a search costs theory. To hold otherwise--as some circuits do--merely encourages wasteful duplication of search costs. On the one hand, it might require competitors of the owner of an established trademark to undertake their own investments to create a redundant source of information capital. On the other hand, it might require consumers to undertake an additional costly search themselves to identify attractive substitutes for the mark owner's products. Moreover, a consequentialist would have to account for the possibility that use of a mark by the mark owner's competitors on goods of high quality might increase the value of the mark, to the benefit of the original owner.

   Some circuits have attempted to justify the role of product quality in infringement analysis, but their efforts are not flattering to consequentialists. The Ninth Circuit, for example, while adopting the Second Circuit's position, noted in the alternative that treating similar product quality as probative of infringement is justified because "present quality is no assurance of continued quality." (155) This argument, while consistent with consequentialist theory, seems to allow a plaintiff who suffered no injury to obtain a judgment against a defendant who may (or may not) injure him someday, ignoring standing (156) and ripeness (157) doctrines that would seem to be directly applicable. Moreover, the language of the Ninth Circuit (and of the Seventh Circuit in a similar opinion) is more Lockean than consequentialist: it evinces a concern for the trademark owner's right to control the reputation he built up in an ongoing business, not the reduction of search costs or the preservation of incentives. (158) In short, consequentialism has some difficulty defending this basic feature of a core area of trademark doctrine. (159)

   A contractualist, in contrast, faces no such difficulty. Contractualism analyzes passing off in terms of the relationship between the defendant producer and his customers, and their mutual respect for one another's autonomy. Consider a producer, D, a consumer, C, and a trademark, M, that C believes is indicative of goods that originate with another producer, P. Where D uses trademark M on D's goods and C purchases those goods in the belief that he is acquiring a good made by P, D may be guilty of violating one or more of the moral principles that emerge from contractualist analysis. (160) In the easier case in which D's product is inferior to P's products, we would say that D has violated Scanlon's Principle M. (161) That is, he has caused C to take an action (purchasing the product) that he would not otherwise have taken, in the expectation that he would receive something from D (a product consistent with the quality of P's products) that D failed to provide, thereby injuring C.

   But perhaps C doesn't simply want a product consistent with the qualities he has come to expect of P's products. Perhaps C also wants to be assured that the product was in fact made by P--whether as a guarantee of the product's unobservable qualities or for any other subjective reason--and in the absence of such assurance, C would not have purchased the product. In such a circumstance, we would say that D has violated Scanlon's Principle F--the principle of fidelity. (162) Importantly, this conclusion does not in any way depend on the actual qualities of D's product--it depends on our respect for C's autonomy and for the moral value of C's power of choice.

   Principle F, unlike Principle M, cannot be satisfied by delivering something of equivalent value or quality. (163) If C wants and expects a product made by P, then D (knowing of this desire of C's) may not defend his delivery of a product made by D instead on the grounds that it is just as good. In this view, the use of a trademark is tantamount to a promise--an assurance concerning the nature of the good that the seller must perform unless the buyer explicitly releases him. To hold otherwise would fail to respect the consumer as an end in himself. It would fail to respect his autonomy-based right to make purchasing decisions on the grounds that seem best to him, and would instead substitute a producer's (or a court's) judgment that the consumer ought to be satisfied with what he received. This view of trademark as promise offers, in my view, a much more plausible descriptive account of the courts' treatment of product quality in infringement cases than the search costs theory of consequentialism.

2. Contractualism Versus Consequentialism: Products or People?

   Turning from the descriptive to the prescriptive, recall the argument made in Part III that the distinction between contractualist and consequentialist approaches to the regulation of information transfers among parties engaged in market exchange is a matter of competing priorities. The consequentialist prioritizes the efficient creation and dissemination of information about the subject matter of exchange. The contractualist, however, refuses to elevate that goal above the obligations that market participants possess to respect one another's autonomy--particularly the obligation not to knowingly take advantage of asymmetric access to information about the subject matter of exchange. To apply this insight to trademark law, we must identify a circumstance in which the law purports to regulate trademark-related transactions on the basis of the information conveyed through those transactions to third parties. I believe there is only one area of trademark law that meets this description, and it is one that I have previously written about at some length.

   The doctrine of post-sale confusion--a subset of infringement--is designed to prevent confusion among the general public arising from the conspicuous consumption or downstream sale of a product bearing an unauthorized trademark, even where the original purchaser of that product was not in any way confused at the point of sale. (164) I will focus here on two theories of post-sale confusion, each of which involves a defendant seller, D; a plaintiff trademark owner, P; a nonconfused purchaser of D's product, C; and an observer of C's consumption of D's product, O.

   The first theory--which I have elsewhere called "bystander confusion" (165)--refers to a situation in which D sells a shoddy product bearing P's trademark to C, and...