What can be learnt from econometric studies in cartel cases?

Cartel econometrics has become a well-developed field of research that can contribute to the efficient implementation of competition policy. Beyond the most traditional type of analysis aiming to quantify the impact of cartels on consumer welfare, other methods have been developed lately. They allow economists to check whether the observed behaviour (in terms of public price announcements, price dispersion, or bids in response to tenders) is more likely to have been caused by collusion or by normal competitive interaction. This can facilitate the delineation of the periods effectively affected by coordination, as well as the interpretation of the available evidence when there is no obvious smoking gun. All these methods require great caution in order for their results to be robust and convincing. The currently ongoing effort by competition authorities' economists to define best practices for the submission of economic evidence should make the use of econometric techniques more efficient.