"Economic reform is sometimes seen as damaging to a government's re-election chances, but anecdotal evidence from OECD countries would not seem to strongly support this perception. This paper tests this hypothesis on a sample of 21 OECD countries over the period 1985-2003, controlling for other economic and political factors that may affect re-election. It is found that the chances of re-election for incumbent governments are not significantly affected by their record of pro-market reforms. However, the electoral impact of reform is found to differ strongly depending on which types of policies are considered. In particular, reform measures that are more likely to hurt large groups of 'insiders' seem electorally more damaging. A series of framework conditions appears to affect the impact of reforms on re-elections. Reformist governments in countries with rigid product and labour markets tend to be voted out of office, suggesting the existence of a 'rigidity trap'. While fiscal stimulus is not an effective instrument to 'sweeten the pill' and raise the odds of re-election, the presence of liberal financial markets appears to soften electoral resistance to structural reform. The latter finding is of particular relevance in the current financial crisis: forward-looking governments should not rush to over-regulate financial markets in order not to compromise the feasibility of product and labour market reforms". Copyright (c) CEPR, CES, MSH, 2010.