We provide new cross-country evidence on smoking persistence in Europe, which can be due to both true state dependence and individual unobserved heterogeneity. We distinguish between the two by using semi-parametric panel data selection methods, applied to both the smoking participation and the cigarette consumption decision. We find that for both smoking decisions true state dependence is generally much smaller, but still important, when unobserved individual heterogeneity is accounted for. We uncover large differences in true state dependence across countries and relate them to discrepancies in regulations, social norms on and tolerance towards smoking, and awareness of its health risks.