Since the introduction of the EU merger control mechanism in 1990, the European Commission has always denied merging parties the option of using a defence based on efficiency considerations to show why an otherwise anticompetitive merger should be cleared. At the end of 2002, however, the Commission made an apparent U-turn in this regard. In fact, the reform package it is currently proposing includes a draft of the horizontal merger guidelines in which a whole section is dedicated to the treatment of efficiencies. Williamson's ‘efficiency defence’ model, despite being considered naïve by its own creator, provides a simple but powerful analytical framework through which enforcement agencies can analyse the social desirability of a concentration. With some adjustments, which are necessary to take account of many ‘real world’ factors, the trade-off suggested by Williamson's compelling model appears to be robust in theory and feasible in practice. Examples of this may be found in the two most advanced antitrust jurisdictions in which such a defence has been contemplated: the United States and Canada. Although there are doubts concerning the practical feasibility of introducing efficiency considerations into the EU merger control process, the Commission has in the past consistently demonstrated that it is able to assess, estimate and set off the same trade-off terms suggested by Williamson's refined model in neighbouring areas of antitrust law, including the full-function cooperative joint ventures that fall under Article 81(3) EC.