Strong Ties, Weak Ties, and Bad Ties in Antitrust Caselaw
Abstract
Traditionally, network analysis of legal doctrine has focused on strong ties within caselaw—for instance, by mapping citations to a decision as a proxy for the decision’s influence. Taking a somewhat different tack, this Article highlights the weak links within a body of cases to assess overall doctrinal coherence. Specifically, this Article focuses on the doctrinal coherence of tying caselaw.
In antitrust, tying is the sale of one product on the condition that the consumer also purchase another product. Due to doctrinal and economic complexities, tying jurisprudence suffers from extraordinary incoherence. To gauge this incoherence, this Article applies topic modeling (with novel modifications) to a corpus of nearly every federal tying decision since the passage of the Sherman Act and builds visualizations that depict the tightness—and thereby the internal consistency—of topics within the tying caselaw.
This application yields three results. First, tying cases comprise a distinct and pronounced cluster (i.e., bound by “strong ties”) among all antitrust cases. Second, within tying caselaw, topics are diffusely connected (i.e., “weak ties”). Finally, the most prominent tying decisions involve franchise tying, which has low precedential value (i.e., “bad ties”) but dominates the corpus nonetheless. These findings suggest that, unlike other antitrust cases, tying decisions are not as broadly influential across fact patterns.