Section 2 Symposium: Howard Marvel on Safe Harbors for Short Term Exclusive Dealing Contracts

Howard Marvel —  6 May 2009

marvelHoward P. Marvel is Professor of Economics in the Department of Economics and Professor of Law in the Moritz College of Law, both at The Ohio State University.

Exclusive dealing prevents the bait-and-switch behavior by dealers who convert customers drawn by one brand to the products of its rivals. Despite the red flag of “exclusive” in its title, the practice is ordinarily uncontroversial, indeed innocuous. Automobile manufacturers often pay incentives to encourage dealers to deal exclusively in their vehicles. Business format franchising ensures that large swathes of distribution are exclusive. And when exclusive dealing is denied to suppliers, the results can be catastrophic, as when the FTC massacred those manufacturers of hearing instruments that relied on exclusive dealers. The FTC sued five such firms for exclusive dealing, four of which agreed to drop it, and promptly died.

When does exclusive dealing represent a problem? When distributors that sign exclusive dealing arrangements are the only way for rival suppliers to get to market.  Thus, the Section 2 Report says that the analysis of exclusive dealing begins with a determination “whether the arrangement has the potential to harm competition and consumers. In situations where competitive harm is implausible-for instance, where other efficient distribution methods are available in sufficient size and number to rivals-courts appropriately uphold the arrangement.” (at 140)

There is a shortcut available for this evaluation. Suppose that B can bring in customers of its own, as opposed to hijacking customers that A brings in. Then B can offer dealers a better deal, and thereby can buy distribution to compete effectively with A. But B’s efforts to amass a distributor network can be hamstrung by A’s contracts, particularly if A’s contracts are of long duration and have staggered expiration dates. A prospective rival who wants to obtain distribution in the absence of such contracts has merely to provide a combination of lower prices (not hard to do if the incumbent is a monopolist) and promotion that brings customers of its own to dealers. The absence of long term contracts has generally been sufficient to allay fears about exclusive dealing.

But not any more. The DOJ won its challenge to Dentsply’s exclusive dealing despite the absence of long-term contracts. (Disclosure: I was an expert witness for Dentsply in this case.) Make that despite the absence of any exclusive dealing contract at all. Dentsply simply refused to sell its teeth to any dealer who sold a rival’s teeth (other than grandfathered rivals). A dealer who wanted to walk away from Dentsply was free to do so.

But on the grounds, apparently, that a win is a terrible thing to waste, the Section 2 Report sweeps aside the possibility of a safe harbor for short duration contracts saying that “the legality of exclusive dealing should not depend solely on its length.” (at 141) A firm with a high share must defend its use of exclusive dealing with a showing that rivals had available “alternative distribution channels” that would “allow competitors to ‘pose a real threat’ to [the] continued existence [of the defendant’s market power].”

That’s a tough test for any firm that holds a large share. The rebuttal is to say simply “if you’re so threatened, why are you still here?”

Denstply’s rivals in the premium tooth end of the market (that’s where the money was) were two large European firms, Ivoclar and Vident. Each sold teeth successfully in Europe. Teeth in the U.S. are sold to dental laboratories that fabricate dentures. Dentists can specify teeth, but as removable dentures have declined relative to fixed solutions, dentists increasingly rely on labs to choose teeth . When dentists selected teeth, Dentsply’s promotion to dentists and dental schools gave it preferred status. But with reliance on labs, bait-and-switch by the dealers selling teeth to labs became a possibility. This switch in distribution resulted in Dentsply’s exclusive dealing policy for dealers.

Each rival had a permanent appliance business that exceeded Dentsply’s.  Rival representatives were in the labs on a regular basis. All they needed was an appealing product line and a compensation structure that encouraged them to take time out from pushing crowns to pushing teeth.

The rivals had neither. European teeth are realistic in appearance, which isn’t necessarily a plus. Having observed the results of European dentistry, it may be that the US market for such realism peaks around Halloween but is modest otherwise. Rivals teeth were also tough labs to use. The relatively flat biting surfaces on American-style teeth produced comfortable dentures more easily.

The District Court found Dentsply’s explanation for its rivals’ failures plausible. The Third Circuit reversed on the grounds that Dentsply was big, and therefore shouldn’t be messing around with any practice that had “exclusive” in its name. The appellate opinion swept aside the District Court because of Dentsply’s success: its rivals’ inability to grab share was “a reflection of the effectiveness of Dentsply’s exclusionary policy.” Only by failing to compete could Dentsply demonstrate that its market was competitive. The Section 2 Report makes this ‘success is enough to condemn’ result more likely in future cases. But who was right? Was the problem for a rival that it simply had to have access to dealers, or was it something else?

Since Dentsply, Ivoclar and Vident have followed separate paths. Ivoclar has introduced a new line of teeth, sold direct, priced below premium levels, and designed for US tastes. They should not have succeeded in the view of DOJ. In contrast, Vident has obtained distribution for its teeth, at least for some periods of time, through some of Dentsply’s largest dealers. Thus we have a test-if the DOJ was correct that exclusive dealing was holding back rivals, Vident should have surged, while if the District Court’s finding that rivals had available adequate distribution was right, Ivoclar had a shot. Who wins? Ivoclar has gained share. A search for Vident  on the websites of the two big mail-order dealers that picked up Vident teeth after Dentsply yields shade guides at both locations, but no teeth.

It is unfortunate that Dentsply, a decision that the Court of Appeals got wrong by going simply with its gut instinct that big plus exclusive is bad, is the basis for the Section 2 Report’s position on exclusive dealing. A better outcome would have been simply to handle exclusive dealing through the statute under which it is treated explicitly, Section 3 of the Clayton Act. There we find that “exclusive-dealing contracts terminable in less than a year are presumptively lawful under section 3.” (Roland v. Dresser, 749 F.2d 380 (7th Cir., 1984)).


2 responses to Section 2 Symposium: Howard Marvel on Safe Harbors for Short Term Exclusive Dealing Contracts

  1. 
    Michael Salinger 6 May 2009 at 3:57 pm

    The duration of exclusive contracts is a factor to consider, but short contracts should not be a safe harbor. Certainly, there are valid reasons for companies to want exclusivity from their trading partners. But the distinction Tim Brennan drew is the right one. Section 2 challenges about competing too hard – charging too low a price, improving products, investing in capacity, adding product features – should be judged under standards that entail low risks of false positives. Exclusionary behavior – in effect paying firms not to deal with rivals – is more problematic. Because firms presumably have to pay in some way for any exlcusivity they get, exclusive dealing is the sort of practice where a no economic sense or disproportionate harm test is inappropriate.

    Should Microsoft (or Intel) be allowed to refuse to sell to OEMs who sell machines with Linux (or AMD chips)? There are serious arguments on both sides of the question, but they do not turn on the duration of the contracts.

  2. 
    Tim Brennan 6 May 2009 at 8:32 am

    Howard’s sensible recommendation regarding timing would be more likely to be followed if the analysis of exclusive dealing focused on the market that matters–the complement providers, such as distributors, dental labs, retail outlets, input suppliers–with whom the exclusive dealing contracts were made. One could and should treat multiple instances of exclusive dealing as tantamount to a merger of those complement providers, as far as dealing the product is concerned.

    That, in turn, invites applying the Horizontal Merger Guidelines to assessing the competitive effects in that complement market. One aspect of the HMGs is entry–if the exclusive dealer ties up the complement market, how long will it take new competition to enter? If entry is sufficiently rapid, there’s no problem with the merger.

    It is but a short step to say that if the complement providers who signed the exclusive contract could and would bolt within that HMG-specified time limit, there’s unlikely to be a competitive problem with the exclusive dealing arrangement. I’ll leave it to Howard and others to debate that principle would apply to Dentsply.