Despite many claims by high-ranking policy-makers and some scientists that climate change breeds violent conflict, the existing empirical literature has so far not been able to identify a systematic, causal relationship of this kind. This may either reflect de facto absence of such a relationship, or it may be the consequence of theoretical and methodological limitations of existing work. We revisit the climate–conflict issue along two lines. First, at the theoretical level we specify the mechanism through which climate change is likely to affect the risk of armed conflict. We focus on the causal chain linking climatic conditions, economic growth, and armed conflict, and also argue that the growth–conflict part of this chain is contingent on political system characteristics. Second, at the methodological level, we develop an approach that takes care of endogeneity problems in the climate–economy–conflict relationship. We test our theoretical argument on a global data set for 1950-2004. The results show that the climate change–conflict hypothesis rests on rather shaky empirical foundations: we do find some negative effects of climate change on economic growth, while stronger economic growth is associated with a lower probability of civil conflict. However, the climate change effect on growth is not robust to changes in climate indicators and samples. Our results also indicate that non- democratic countries are more likely to experience armed conflict when economic conditions deteriorate. Our results suggest that investing in climate-friendly economic growth and democracy can qualify as a no-regrets strategy.