There’s a old joke about antitrust law and the dangers of pricing decisions.  I’ve heard it several ways, but this as good a re-telling as a few internet searches have turned up:

Three American prisoners are sitting around comparing stories. Eventually the subject turns to what crimes they committed to end up in their predicament.

The first prisoner says, “I charged higher prices than everyone else and they accused me of profiteering and price gauging.”

The second prisoner says, “I charged lower prices than everyone else and they accused me of predatory price cutting and cut-throat competition.”

The third prisoner says, “I charged the same prices as everyone else and they accused me of collusion and cartelization.”

The point, obviously, is to make American antitrust laws look stupid.  Doing my duty as an antitrust practioner, I’ve defended antitrust enforcement from that sort of cynicism.  Robinson-Pattman is rarely enforced (and not at all by the agencies, really).  ”Price gauging,” on it’s own, really isn’t an antitrust theory.  There’s nothing wrong with pricing the same as long as it’s not the result of an agreement.

But it seems that the Commissioner Wright (and the Third Circuit) wants to revive the second part of the joke.  In a recent speech he argued that loyalty discounts should be judged for their exclusive effects rather than falling within a safe harbor if they result in above-cost prices.  In effect, he’s arguing that companies should face the possibility of civil lawsuits alleging that they are harming competition by pricing too low.

I don’t agree.  And here’s something you won’t hear me say very often: I don’t even care if he’s got the economic analysis right.

Sometimes, you need bright line rules.  Pricing decisions are one of those times.  Price-cutting should be presumptively legal as long as prices remain above cost (and in many circumstances, even below cost).  Any other rule creates an incentive not to discount, which harms consumers.

Is there some marginal loss of antitrust enforcement from such a presumption?  Probably.  There will be some cases in which above-cost pricing will foreclose some amount of competition.  But it’s going to be extraordinarily rare, and the cost of bringing it within the scope of antitrust enforcement will be the loss of perfectly legal discounting resulting in higher prices to customers.

I thought this discussion was settled with Brooke Group and the near-consensus among the antitrust bar that Robinson-Pattman should be repealed.  Commissioner Wright tells us that the concern with loyalty discounts is that they can “impair competitors’ access to distribution and other sales outlets.”  While that’s certainly true, it really matters how they impair access.  And how they do it is offering lower prices to consumers, which is what antitrust enforcement is supposed to be about.

Commissioner Wright frames the discussion as being about “partially exclusive” (how’s that for an oxymoron?) contractual arrangements, which I don’t think is helpful. Beginning with those descriptors puts a thumb on the scale on the side of suspicion about these practices.  What we’re really talking about is price cutting, not using other coercive means of locking in customers.   Price cutting as a means of maintaining customer relationships is competition on the merits.  In fact, it’s one of the hallmarks of competition on the merits (along with quality and service improvements).  There is no reason for it to be inherently suspect.

His analysis views the loss of a discount as a “cost” of switching to a new supplier for marginal purchases.  Hm. In certain narrow circumstances, perhaps it is.  If, for example, the new supplier is capacity constrained such that can’t meet all of a given customer’s needs, it might be impossible to offer a price low enough to win any business if switching would cause that customer to lose its loyalty discount with its incumbent supplier. In effect, that would increase the minimum efficient scale of the new supplier to at least be the full requirements of at least one customer.

But the “raising rivals costs” view of loyalty discounts seem to rely on what seems to be a questionable unstated assumption about distribution channels. That is, it seems to assume that the existing distribution network is fixed and not subject to meaningful entry.  Or, in other words, that it is possible to lock up “all” the distributors.  But if you’re offering a better product and a lower price, why can’t you find a distributor?  Are there no customers who are interested in your better deal? Surely there is a set of distribution outlets that are just the right size to switch all of their business to you so that you aren’t facing any higher costs via lost discounts.  Maybe it’s one distribution outlet.  Maybe it’s a small handful of them.  Or maybe you can vertically integrate with little additional cost (for example, over the internet).

The question really is whether there is room for a monopolist to price below it’s profit maximizing price but above a competitive price and in so doing keep competitors from entering.  But for that to be the case, doesn’t there have to be something else preventing entry, and isn’t that something else the actual source of the monopolist’s market power?  After all, there is still a margin between the competitive price and the monopolist’s discounted price that should be attractive to entrants.

Obviously, Commissioner Wright understands all of this (and more).  He describes nearly all of the assumptions that need to fall in place for loyalty discounts to have any foreclosure effects.  The difference is one of policy judgment.  He chooses “accuracy” over administrability.  I’m skeptical that there is much additional realistic accuracy in his approach, and I’m certain there is a lot of litigation cost, so I’d make the opposite choice.

Of course, Commissioner Wright sits on the Federal Trade Commission, and I’m just a guy with a blog. If you’re a firm with a large market share, perhaps you want to be careful with your loyalty discounts.

There is quite a bit more on this topic on the Truth on the Marketblog, which I intended to include my reactions to, but time is running short so perhaps another day.