It is common practice to summarise the economic performance of countries in terms of four dimensions (real growth, inflation, unemployment and the external account), which are visually captured by the magic diamond of the OECD. In this paper we present a synthetic performance measure which merges the four separate indicators into one single statistic. The relative importance of each indicator, representing another macroeconomic objective, may vary across countries and over different subperiods. Therefore we want to construct an indicator which allows unequal weighting of its components, using a data envelopment analysis (DEA)-inspired linear programming model which exhibits ‘benefit of the doubt weighting’. These synthetic macroeconomic performance scores reveal interesting information. They confront measurement with perception. In this paper we use our measure to check empirically whether the strict Maastricht convergence criteria actually have led to a relative economic performance deterioration of the EU-candidates compared to the rest of the world. This viewpoint is often articulated in the theoretical literature. In particular, we investigate the performance of twenty OECD countries, half of which belongs to the EU, in the quinquennial period before and after the Maastricht Treaty.de unions are country specific. Second, a change in competition policy is likely to affect labour markets differently than a change in trade policy. The results also indicate that apart from location, market structure and the level at which wages are bargained over (firm or sector level) matter. In a further step the theoretical predictions we derive, are tested on Belgian company accounts data supplemented with data from a postal survey.