Loyalty discounts, exclusive dealing and bundling: Rule of reason, quasi-per-se, price-cost test, or something in between?

Assaf Eilat, David Gilo, Guy Sagi

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

The article reviews loyalty rebates, target rebates, exclusive dealing, and bundling, and argues that these are analogous practices that deserve similar competitive analyses and rules. In particular, in the case of all of these practices, at least some marginal units are typically sold below cost. The article shows that the analyses and rules that should apply to all of these practices ought not to depend on their labels, but rather on the monopoly power of the supplier engaged in the practice; whether, in the particular case, exclusion is costless or almost costless; the size of the sanction that the buyer suffers from being disloyal to the monopolist, and whether the sanction makes it impossible for the monopolist's as efficient rivals to compete for the buyer; the degree of market foreclosure, including its effective duration; the presence or absence of any efficiency justifications, and whether the discount is expected to be passed on to consumers. The analysis further highlights how exclusion may well be costless, or almost costless and can be achieved when the monopolist has non-price means of coercing buyers to be loyal. Further, intermediate cases are explored, in which exclusion, though not entirely costless, is nevertheless cheaper to the monopolist than ordinary predatory pricing.

Original languageEnglish
Pages (from-to)345-380
Number of pages36
JournalJournal of Antitrust Enforcement
Volume4
Issue number1
DOIs
StatePublished - 1 Apr 2016

Keywords

  • Bundling
  • Exclusionary practices
  • Exclusive dealing
  • Loyalty rebates
  • Target rebates
  • Tying

Fingerprint

Dive into the research topics of 'Loyalty discounts, exclusive dealing and bundling: Rule of reason, quasi-per-se, price-cost test, or something in between?'. Together they form a unique fingerprint.

Cite this