Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorKerber, Wolfgangen_US
dc.contributor.authorKretschmer, Jürgen-Peteren_US
dc.contributor.authorvon Wangenheim, Georgen_US
dc.description.abstractAlthough both in US antitrust and European competition law there is a clear evolution to a much broader application of rule of reason (instead of per-se rules), there is also an increasing awareness of the problems of a case-by-case approach. The error costs approach (minimizing the sum of welfare costs of decision errors and administrative costs) allows not only to decide between these two extremes, but also to design optimally differentiated rules (with an optimal depth of investigation) as intermediate solutions between simple per-se rules and a fullscale rule of reason. In this paper we present a decision-theoretic model that can be used as an instrument for deriving optimal rules for a sequential investigation process in competition law. Such a sequential investigation can be interpreted as a step-by-step sorting process into ever smaller subclasses of cases that help to discriminate better between pro- and anticompetitive cases. We analyze both the problem of optimal stopping of the investigation and optimal sequencing of the assessment criteria in an investigation. To illustrate, we show how a more differentiated rule on resale price maintenance could be derived after the rejection of its per-se prohibition by the US Supreme Court in the Leegin case 2007.en_US
dc.publisher|aUniv., Dep. of Business Administration & Economics |cMarburgen_US
dc.relation.ispartofseries|aJoint discussion paper series in economics |x2008,16en_US
dc.subject.keywordLaw Enforcementen_US
dc.subject.keywordCompetition Lawen_US
dc.subject.keywordAntitrust Lawen_US
dc.titleOptimal sequential investigation rules in competition lawen_US
dc.type|aWorking Paperen_US

Files in This Item:
233.13 kB

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.