As a part of a broader revival of Section 2 of the Sherman Act, the anti-trust division of the Department of Justice, under Obama appointee Christine Varney, has opened a review of potentially anti-competitive practices by the dominant telcom providers.  One specific issue that has received attention is exclusionary contracting between wireless carriers (AT&T) and handset manufacturers (Apple iPhone.) The FTC is reportedly also exploring these contracts.  Exclusive contracts bind a manufacturer’s handsets to specific carriers thereby hindering or preventing end-users from migrating to other carriers.  The widespread nature of these contracts may create a barrier against entry by new, smaller wireless providers who cannot offer their users handsets that compete with the top models.

The review is reported to be at an early stage and may not lead to a formal investigation, but as this develops there are a few basic economic arguments to keep in mind.  To start with, there is the benchmark “Chicago School” view which starts with the observation that exclusionary contracts require the voluntary agreement of the handset manufacturers.  The manufacturers internalize the costs of the entry barrier because without entry they will have fewer competitive carriers to sell their phones to.  Therefore, exclusive contracts must compensate manufacturers for this loss impying that these contracts will be in place only when the total surplus from exclusion exceeds the cost, i.e. when it is efficient. The Chicago argument is a longstanding pillar of regulatory policy that still holds sway today.  From the article:

Jon Muleta, former wireless bureau chief of the FCC, said exclusive handset deals won’t be an issue the government can pursue on antitrust grounds unless major handset makers say they’re being forced into the deals. “The equipment providers enter into these deals willingly,” Mr. Muleta said.

The Chicago argument ignores the costs to end users from reduced competition in wireless service.  It would apply only if manufacturers internalize all of the benefits to consumers from increased competition. But under any reasonable model of the wireless market structure, end-user consumer surplus would increase with more competition for wireless service and this becomes an externality relative to the parties in the Chicago bargain.

Secondly, the Chicago argument has been discredited as it takes a naive view of the way contract negotiation would work.  Implicitly, the Chicago argument assumes that handset manufacturers must be compensated at least what they would earn if entry were to occur.  But scale economies imply that a new carrier will enter only if sufficiently many, or sufficiently large, manufacturers remain free of exclusive deals.  The dominant carriers can use a “divide and conquer” strategy which exploits the difficulty for handset manufacturers to coordinate severing their exclusive deals.  Without this coordinated threat, manufacturers cannot extract the compensation envisioned in the Chicago argument, and again efficiency breaks down.

The definitive references here are Rasmusen, Rasmeyer,  and Wiley “Naked Exclusion” and a follow-on comment by Segal and Whinston, both in the American Economic Review.

There is a separate defense of exclusive contracts, often cited and also reflected in the article.

Paul Roth, AT&T’s president of retail sales and service, told Congress last month that the billions of dollars the company invests in its network and services would be put at risk if government were to “impose intrusive restrictions on these services or the way that service providers and manufacturers collaborate on next-generation devices.” Mr. Roth said there is plenty of competition and innovation in the wireless industry.

AT&T’s tremendous investment in its 3G network will pay off only because of its exclusive deal with Apple to market the iPhone.  Thus, it is often argued that exclusive contracts are in fact pro-competitive as they reward investment with profits that would otherwise be subject to hold-up or competed away.  I will take up this argument in a subsequent post.