For best experience please turn on javascript and use a modern browser!

The Intertemporal Path of U.S. Criminal Antitrust Enforcement

Event details of May 30, 2005: Vivek Ghosal (Georgia Institute of Technology)
Date 30 May 2005
Time 11:45 -13:15


Firms’ conduct related to price-fixing, bid-rigging, output-restriction and market allocation schemes are considered illegal per se and these constitute the criminal antitrust conspiracies. We study criminal enforcement by the Antitrust Division of the U.S. Department of Justice during the post-war era,1948-2003, and the data reveal substantial short-run fluctuations in enforcement as well as conspicuous long-run drifts.

The first part of the paper examines the broad driving forces for the total number of criminal prosecutions. Four key findings emerge from our time-series analysis which controls for shorter-run and longer-run driving forces. First, criminal enforcement appears to follow a distinct counter-cyclical pattern with the number of criminal cases prosecuted by the Antitrust Division increasing about two years after an economic downturn. We tie this finding to the rich literature on the stability of cartels. Second, criminal enforcement shows an increase a year or two after increases in civil enforcement (monopolization cases, merger control and restraints of trade). We link this finding to the process via which the Antitrust Division pursues criminal investigations and our results suggest that civil investigations unearth criminal conduct leading to an increase in criminal cases prosecuted. Third, there is evidence of a clear structural-break in criminal enforcement starting around 1979. We relate this result to the shift in U.S. doctrine of emphasizing antitrust enforcement in those areas with clearer loss of welfare. We also discuss the role played by amnesty programs, leniency laws and increases in fines. Fourth, our results show that post-structural-break, Republican administrations have placed greater emphasis on criminal enforcement as compared to Democrats. An explanation of this finding is that Republicans have attempted to target business conduct with incontrovertible harm to consumers without compensating gains in efficiency.

The latter part of the paper examines more detailed data and employs the Vector Autoregression methodology to study the dynamic interrelationships between the number of civil prosecutions, the number of grand jury investigations, the number of criminal prosecutions, and the number of firms and individuals convicted for price-fixing and related allegations. Our analysis uncovers several interesting relationships which point to complex feedback effects between investigations, court cases and convictions of firms and individuals.


Oudemanhuispoort 4-6
1012 CN Amsterdam