Although there exists a voluminous literature on the causes and effects of electoral disproportionality, the establishment of the main determinants of this phenomenon is still an open empirical question. Do voters react in a completely strategic way when a majoritarian reform of the electoral system is introduced? Do they, in contrast, need a couple of elections conducted under the new rules in order to learn about to the change in the incentives generated by the electoral institutions? Which is the modifying impact, if any, played by ethnic fractionalization? Taking advantage of our own dataset, we aim to establish the main determinants of electoral disproportionality in European democratic countries since World War II. Identification is achieved by using the appropriate econometric techniques (basically, cross-sectional time-series). We first hypothesize that permissive changes in the electoral system lead to less disproportionality, whereas the effect of restrictive reforms is the opposite. However, in order to provide an accurate picture of this phenomenon, we need to take into account the modifying effect of three additional variables such as the stage in the process of democratic consolidation, the number of elections conducted under the old/new rules, and the level of ethnic fractionalization. To test these arguments, we use a new dataset that contains observations for elections from both new and consolidated democracies.