Conventional wisdom holds that voters punish governments that implement fiscal austerity. Yet, most empirical studies, which rely on ex-post yearly austerity measures, do not find supportive evidence. This paper revisits the issue using action-based, real-time, ex-ante measures of fiscal austerity as well as a new database of changes in vote shares of incumbent parties. The analysis emphasizes the importance of the 'how'--whether austerity is done via tax hikes or expenditure cuts--and the 'who'--whether it is carried out by left- vs. right-leaning governments. Our main finding is that tax-based austerity carries large electoral costs, while the effect of expenditure-based consolidations depends on the political-leaning of the government. An austerity package worth 1% of GDP, carried out mostly through tax hikes, reduces the vote share of the leader's party by about 7%. In contrast, expenditure-based austerity is detrimental for left- but beneficial for right-leaning governments.

We also find that the electoral cost of austerity--especially tax hikes--can be contained if it is implemented during good economic times.