How Loyalty Discounts Can Perversely Discourage Discounting: Comment
Assaf Eilat, Jith Jayaratne, Janusz Ordover, Greg Shaffer, Apr 16, 2010
The academic literature on loyalty discounts and exclusive dealing demonstrates that the welfare effects of these practices are ambiguous and that market details determine the direction of the effect. However, in his recent paper entitled How Loyalty Discounts Can Perversely Discourage Discounting, Professor Einer Elhauge argues that exclusive contracts with loyalty discounts offered by a single incumbent seller can create anticompetitive effects in a broad range of settings. Moreover, he claims, "anticompetitive effects are exacerbated if multiple sellers use loyalty discounts."
To motivate his analysis, Professor Elhauge defines two types of customers of the incumbent seller, namely: "committed buyers," who signed an exclusive contract with the incumbent, and "uncommitted buyers," who did not. The "uncommitted buyers" are free to purchase from the entrant, if there is entry, or from the incumbent. The contractual benefit from signing an exclusive contract with the incumbent is the promise that the committed buyer will always pay $d less than the price that the incumbent offers to the uncommitted buyers. Put another way, $d is the assured spread between the incumbent's prices offered to these two groups. From this it follows that if the incumbent chooses not to compete for the uncommitted buyers then the incumbent can, in effect, charge committed buyers the monopoly price.
Professor Elhauge departs from key papers in the literature on exclusive dealing by defining his exclusive contracts to include a Most Favored Nation (MFN)-like feature, which actually does a great deal of work in his model. The MFN-like feature comes through the assumption that the price paid by committed buyers depends not only on the fact that they committed to an exclusive contract, but also on the price that the seller offers to uncommitted buyers (the "list" price). This link between the prices offered to committed and uncommitted buyers is not found in any of the key papers in the exclusive-dealing literature that Professor Elhauge cites. This linkage is crucial because--as exposited by Professor Elhauge--it greatly increases the incumbent's cost of competing with the entrant for sales to the uncommitted buyers. In fact, in Professor Elhauge's model, the incumbent seller sets its list price so high that no uncommitted buyer will buy from it in equilibrium, which means that along the equilibrium path, committed buyers never actually receive the loyalty discount d they are promised; or, stated more precisely, they receive their "discount" d off of a phantom price which nobody pays! As we show below, the qualitative results in his model are unchanged even if there are no loyalty discounts and buyers do not commit to buy exclusively from the incumbent. Thus, it is the MFN-like feature of the contracts, not the loyalty discounts, that drives his results--and the potential impact of MFN provisions on competition has already been extensively analyzed in other papers.