Should There Be a Safe Harbor for Above-Cost Loyalty Discounts? Why I Believe Wright’s Wrong.
It’s not often that I disagree with my friend and co-author, FTC Commissioner Josh Wright, on an antitrust matter. But when it comes to the proper legal treatment of loyalty discounts, the Commish and I just don’t see eye to eye.
In a speech this past Monday evening, Commissioner Wright rejected the view that there should be a safe harbor for single-product loyalty discounts resulting in an above-cost price for the product at issue. A number of antitrust scholars—including Herb Hovenkamp, Dan Crane, and yours truly—recently urged the Supreme Court to grant cert and overturn a Third Circuit decision refusing to recognize such a safe harbor. Commissioner Wright thinks we’re wrong.
A single-product loyalty discount occurs when a seller conditions a price cut (either an ex ante discount or an ex post rebate) on a buyer’s purchasing some quantity of a single product from the seller. The purchase target is often set as a percentage of the buyer’s requirements, as when a medical device manufacturer offers to pay a 20% rebate on all of a hospital’s purchases of the manufacturer’s device if the hospital buys at least 70% of its requirements of that type of device from the manufacturer. Because a loyalty discount tends to encourage distributors to carry more of the discounting manufacturer’s brand and less of the brands of the discounter’s rivals, such a discount may tend to “foreclose” those rivals from available distribution outlets. If the degree of foreclose is so great that rivals have to cut their output below minimum efficient scale (the minimum output level required to achieve all economies of scale), then the discount may “raise rivals’ costs” relative to those of the discounter and thereby harm consumers.
On all these points, Commissioner Wright and I are in agreement. Where we differ is on the question of whether a loyalty discount resulting in a discounted price that is above the discounter’s own cost should give rise to antitrust liability. I say no. I take that position because such an “above-cost loyalty discount” could be matched by any rival that is as efficient a producer as the discounter. If, for example, a manufacturer normally charges $1.00 for widgets it produces for $.79 each but offers a 20% loyalty discount to retailers that buy 70% of their widget requirements from the manufacturer, any competitor that could produce a widget for $.79 (i.e., any equally efficient rival) could stay in business by lowering its price to the level of its incremental cost. Thus, any rival that loses sales because of a manufacturer’s above-cost loyalty discount must be either less efficient than the manufacturer (so it can’t match the manufacturer’s discounted price) or unwilling to lower its price to the level of its cost. In either case, the rival is unworthy of antitrust’s protection, where that protection amounts to prohibiting price cuts that provide consumers with immediate benefits.
Commissioner Wright disputes (I think?) the view that equally efficient rivals could match all above-cost loyalty discounts. He maintains that loyalty discounts may be structured so that
[a] distributor’s purchase of an additional unit from a rival supplier beyond the threshold level can result in a loss of rebates large enough to render rival suppliers unable to attract a distributor to purchase the marginal unit at prices at or above the marginal cost of producing the good.
While I’m not entirely certain what Commissioner Wright means by this remark, I think he’s making the point that a loyalty discounter’s equally efficient rival might not be able to attract purchases by matching the discounter’s above-cost loyalty rebate if the rival’s “regular” base of sales is substantially smaller than that of the discounter.
If that is indeed what Commissioner Wright is saying, he has a point. Suppose, for example, that the market for tennis balls consists of two brands, Penn and Wilson, that current market shares, reflective of consumer demand, are 60% for the Penn and 40% for Wilson, and that retailers typically stock the two brands in those proportions. Assume also that it costs each manufacturer $.90 to produce a can of tennis balls, that each sells to retailers for $1 per can, and that minimum efficient scale in this market occurs at a level of production equal to 35% of market demand. Suppose, then, that Penn, the dominant manufacturer, offers retailers a 10% loyalty rebate on all purchases made within a year if they buy 70% of their requirements for the year from Penn. The $.90 per unit discounted price is not below Penn’s cost, so the loyalty discount would come within my safe harbor.
Nevertheless, the loyalty discount could have the effect of driving Wilson from the market. After implementation of the rebate scheme, a typical retailer that previously purchased sixty cans of Penn for $60 and forty cans of Wilson for $40 could save $7 on its 100-can tennis ball requirements by spending $63 to obtain seventy Penn cans and $30 to obtain thirty Wilson cans. The retailer and others like it would thus have a strong incentive to shift purchases from Wilson to Penn. To prevent a loss of market share that would drive it below minimum efficient scale (35% of market demand), Wilson would need to lower its price to provide retailers with the same total dollar discount, but on a smaller base of sales (40% of a typical retailer’s requirements rather than 60%). This would require it to lower its price below cost. For example, Wilson could match Penn’s $7 discount to the retailer described above only by reducing its $1 per-unit price by 17.5 cents ($7.00/40 = $.175), which would require it to price below its cost of $.90 per unit. Viewed statically, then, it seems that even an above-cost loyalty discount could occasion competitive harm by causing rivals to be less efficient, so that they could not match the discounter’s price.
In light of dynamic effects, though, I’m not convinced that examples like this undermine the case for a safe harbor for above-cost loyalty discounts. Had the nondominant rival (Wilson) charged a price equal to its marginal cost prior to Penn’s loyalty rebate, it would have enjoyed a price advantage and likely would have grown its market share to a point at which Penn’s loyalty rebate strategy could not drive it below minimum efficient scale. Moreover, one strategy that would prevent a nondominant but equally efficient firm from being harmed by a dominant rival’s above-cost loyalty rebate would be for the non-dominant firm to give its own volume discounts from the outset, securing up-front commitments from enough buyers (in exchange for discounted prices) to ensure that its production stayed above minimum efficient scale. Such a strategy, which would obviously benefit consumers, would be encouraged by a liability rule that evaluated loyalty discounts under straightforward Brooke Group principles (i.e., that included a safe harbor for above-cost discounts) and thereby signaled to manufacturers that they must take steps to protect themselves from above-cost loyalty discounts.
Commissioner Wright maintains that all this discussion of price-cost comparisons is inapposite because the theoretical harm from loyalty discounts stems from market exclusion (and its ability to raise rivals’ costs), not from predation. He says, for example:
- “[T]o the extent loyalty discounts raise competition concerns, the concerns are about anticompetitive exclusion and, as a result, the legal framework developed to evaluate exclusive dealing claims ought to be used to evaluate claims relating to loyalty discounts.” [p. 12]
- “[P]redatory pricing and raising rivals’ costs are distinct paradigms of potentially exclusionary conduct. There simply is not a stable relative relationship between price and cost in raising rivals’ cost models that form the basis of anticompetitive exclusion, and hence it does not follow that below cost pricing is a necessary condition for competitive harm.” [pp. 19-20]
- “When plaintiffs allege that loyalty discounts … violate the antitrust laws because they deprive rivals of access to a critical input, raise their costs, and ultimately harm competition, they are articulating a raising rivals’ cost theory of harm rather than price predation.” [p. 24]
- “Raising rivals’ costs and predation are two different economic paradigms of exclusionary conduct, and economic models within each paradigm establish the necessary conditions for each practice to harm competition and give rise to antitrust concerns. Loyalty discounts and other forms of partial exclusives … are properly analyzed under the exclusive dealing framework. Price?cost tests in the predatory pricing tradition … simply do not comport with the underlying economics of exclusive dealing.” [p. 33]
I must confess that I’m baffled by Commissioner Wright’s oddly formalistic pigeonholing. Why must a practice be one or the other—either pricing too low or excluding rivals and thereby raising their costs? That seems like a false dichotomy. Indeed, it seems to me that a problematic loyalty discount is one in which the discounter excludes its rivals from a substantial portion of the distribution network (and thereby raises their costs) via the mechanism of conditional price cuts. It’s “both-and,” not “either-or.” And if that’s the case, then surely it makes sense to limit which price cuts may occasion liability—i.e., only those that could not be matched by equally efficient rivals. [It is important to note here that I don’t advocate a price-cost test as an alternative to a foreclosure-based analysis. Rather, a plaintiff should have to establish below-cost pricing (to show that the plaintiff was deserving of antitrust’s protection via the highly disfavored prohibition of discounts) and demonstrate that the discounting at issue resulted in substantial foreclosure from distribution outlets (the latter showing is necessary to prove harm to competition rather than simply to a competitor).]
Throughout his speech, Commissioner Wright emphasizes that the primary competitive concern presented by loyalty discounts is the possibility of “anticompetitive exclusion.” He writes on page 8, for example, that “[t]he key economic point is that the antitrust concerns potentially arising from loyalty discounts involve anticompetitive exclusion rather than predatory pricing….” On page 12, he reiterates that “to the extent loyalty discounts raise competition concerns, the concerns are about anticompetitive exclusion.” He then apparently assumes that loyalty discount-induced exclusion is “anticompetitive” if it is sufficiently substantial—i.e., if the discounter’s rivals are foreclosed from so many distribution outlets that they are driven below minimum efficient scale so that their costs are raised relative to those of the discounter.
I would dispute the notion that discount-induced exclusion is anticompetitive simply because it’s substantial. Rather, I’d say such exclusion is anticompetitive only if it is substantial and could not have been avoided by aggressive pricing. Omitting the second requirement creates the possibility that antitrust will be used by a laggard rival to prevent a more aggressive rival’s consumer-friendly price competition. (LePage’s anyone?)
Suppose, for example, that there are two producers of widgets, A and B, which both produce widgets at a marginal cost of $.79 and, given their duopoly, charge $1.00 per widget. A, whose market share has hovered around 50%, institutes a loyalty rebate of 20% for retailers that purchase 70% of their requirements from A. If B offers the same deal, or simply cuts its price to $.80, it should lose no market share. But suppose B doesn’t do so, A captures 70% of the market, and B falls below minimum efficient scale. Would we say that B’s exclusion is “anticompetitive” because A’s discount scheme resulted in such substantial foreclosure that it raised B’s costs? Should B be able to collect treble damages for based on its “anticompetitive exclusion”? Surely not.
Commissioner Wright, from whom I have learned more about “error costs” than anyone else, seems oddly unconcerned about the chilling effect his decidedly pro-plaintiff approach to loyalty discounts will produce. Wouldn’t a firm considering a loyalty discount—a price cut, don’t forget!—think twice if it knew its rivals could sit on their hands, claim “exclusion” if the discount successfully moved substantial market share toward the discounter, and collect treble damages? The safe harbor Hovenkamp, Crane, and I have advocated would provide assurance to potential discounters that they will not face liability if they charge above-cost prices, prices that could be matched by equally efficient, aggressive rivals. Isn’t that approach more likely to minimize error costs?
Two closing points. First, despite my disagreement with Commissioner Wright on this issue, I share the widely held view that he is one of the most brilliant antitrust thinkers out there. He’s taught me more about antitrust than anyone (with the possible exception of the uber-prolific Herb Hovenkamp). His questioning of my views on loyalty discounts really makes me wonder if I’m missing something.
Second, to those who think Commissioner Wright has “drifted” or “turned,” let me assure you that he’s long held his views on loyalty discounts. As you can see here, here, and here, we’ve been going round and round on this matter for quite some time.
Perhaps one day one of us will persuade the other.