This paper tests empirically the strategic explanation of budget deficits suggested by Tabellini and Alesina [13] and Persson and Svensson [12]. Tabellini and Alesina suggest that governments with different political orientation provide different public goods. The model predicts that: a) public good provision follows a political pattern; b) the incumbent that anticipates her defeat at the next election runs budget deficits to tie the hands of the future government. Persson and Svensson suggest that liberal governments prefer more public good provision than conservative ones. The model predicts that: a) the conservative (liberal) incumbent that anticipates her defeat at the next election runs budget deficits (surpluses); b) budget imbalances have a political color. Using U.S. and pooled data for sixteen OECD countries, we find little evidence that the incumbent’s probability of being voted out of office explains budget deficits, that the provision of public goods follows a political pattern or that budget imbalances have a political color.