Realism About Theory

So, the point of this post can be made effectively simply by quoting at length from a paper by Louis Kaplow and Carl Shapiro. What I'm getting at is the importance of empirical work and a sense of the appropriate level of confidence in our theoretical results. Here's the passage:

Economists have long recognized that there exist certain prerequisites to successful collusion. The classic modern reference is Stigler (1964). Green and Porter (1984) embed these issues in a supergame context. The key elements are (1) reaching consensus: some understanding must be reached among the otherwise-competing firms regarding what conduct is permitted under the terms of the collusive agreement, such as the prices that the firms will charge; (2) detection: some reliable means must exist by which departures from the agreement can be detected; and (3) punishment: some credible mechanism must be established by which such departures are punished if and when they are detected. Specifically, the prospect of detection and punishment must be sufficient to deter individual firms’ proclivity to cheat on the agreement, typically by cutting prices in the short-term, hoping to reap greater profits through a higher market share, at the expense of the other firms, before they can respond. Related to the need to reach an agreement is the problem of (4) inclusion: a means of inducing participation by a sufficiently large number of incumbent suppliers so that competition from non-participants does not undermine the profitability of the collusive agreement. Lastly and relatedly, the incumbent firms must be protected by (5) entry barriers: there must not be so much competition from quickly arriving new entrants so as to undermine the effectiveness of collusion.

Some economists consider these requirements to be so daunting that cartels are either unable to form or quick to collapse, even in the absence of antitrust laws designed to stop collusion. For example, when OPEC first arose, some confidently predicted its immediate demise. However, the experience with OPEC and empirical evidence on price-fixing more broadly does not support this optimistic view. For example, in the past decade, the Antitrust Division of the U.S. Department of Justice has broken up many large, international cartels that had operated for years (despite the fact that they were illegal under the antitrust laws) and successfully reaped hundreds of millions if not billions of dollars in profits.