Over the last 20 years, the economics profession has moved away from the estimation of structural economic models and toward the use of experiments, either controlled experiments or so-called "natural" experiments, for the purposes of estimating the causal effects of programs, policies, and other interventions. While industrial organization economists have been slower to adopt the experimental approach, it is increasingly commonly seen in industrial organization settings. This is particularly so in the area of antitrust analysis. For example, experimental approaches have been used to analyze the effect of a firm's presence in a local market on another firm's prices in that market in the context of the proposed Staples/Office Depot and Whole Foods/Wild Oats mergers, the effects of consummated mergers on prices, and overcharges associated with price-fixing conspiracies.
However, the experimental approach recently has been subject to somewhat of a backlash in the economics literature. Among the criticisms are that experiments, and particularly natural experiments, frequently answer a very limited question that is not the question of real interest and that the conditions required for the reliability of the results are often not well-articulated and, more importantly, are often not satisfied.
In light of the prominence of experimental approaches in economics, the increasing application of such approaches in industrial organization, and the emerging criticisms, this area of economic research is likely to have an impact on the way that antitrust analyses are performed. In this paper, I review the use of natural experiments in the antitrust context.