In recent decisions involving fast-moving consumer goods, the European Commission has relied on merger simulation models to predict the likely effect of horizontal mergers. Is this the beginning of a new trend? The merger simulation label contains a variety of models, and if all merger simulations look the same to the layman, they do not all have the same probative value. In this article I discuss some of the limits of employing merger simulation, and question the benefit of relying on some categories of simulation models.
1. Is merger simulation back in the practitioner’s toolbox? Merger simulation models have been around since the mid-1990s, but the European Commission has seldom used them to assess the competitive effects of horizontal mergers. Recently, in a number of high profile merger cases involving fast-moving consumer goods, the Commission relied on merger simulation to assess unilateral effects. In support of the Kraft/Cadbury (2009) decision, the Commission cites simulation results submitted by the parties’ economic advisers. In Unilever/Sara Lee (2010), the Commission developed its own merger simulation model to support its finding. And in the SCJ/Sara Lee (2011) case, the parties withdrew the notification late in phase II, but during the merger review both the Commission and