Dictatorships are generally characterized by their use of repression, which is higher than that of democratic systems. Repression is, jointly with the mobilizing of some political support, one of the two basic instruments dictators use in order to stay in power. There is broad empirical evidence proving that loyalty and cooptation help dictators survive in office. Yet, there is not systematic investigation analyzing the role of repression. Does repression help dictators retain power? This paper is aimed at filling this important gap. In particular, I address the simultaneous relationship existing between survival and repression by implementing a two-stage estimation method. I use data on authoritarian rulers’ survival and modes of exit. Data on repression are from the Political Terror Scale, the Cingranelli and Richards’ Physical Integrity Rights Index, and Freedom House’s civil liberties score. The results show that an increased probability of exit causes important and significant increases in repression levels; and that repression helps increasing the likelihood of dictators’ survival. Further, I also analyze if repression affect differently survival once the modes in which rulers’ can be replaced are disaggregated, and vice versa. Curiously, political terror is found to be only significantly helpful in preventing non-violent and regular exits, but not violent or irregular ones. Instead, restrictions on civil liberties are effective in deterring both types of threats. Similarly, it is only non-violent threats which trigger significant increases in political terror, while the risk of a violent or irregular exit leads the regime to increase restrictions on civil liberties.