The results of studies exploring the long-term consequences of famine during foetal or infant development are inconsistent. We tested the hypothesis that selection forces occurring during a famine change the distribution of frailty in the affected cohorts, possibly hiding negative long-term effects. Using mortality data for Finland, gathered from the Human Mortality Database, we explored the effect of being born during the Great Finnish Famine of 1866-68 by comparing mortality at age 60 and over for the 1850-89 births, taking into account unobserved cohort heterogeneity. Contemporaneous Swedish cohorts, unexposed to the famine, were used as an additional control group. When cohort heterogeneity is accounted for, a lower life expectancy at age 60 for male cohorts in Finland born during the famine is observed. The results for females are less conclusive. No substantial changes are seen in the Swedish cohorts. In order to provide consistent estimates of the long-run effects of famines, selection forces need to be considered. © 2013 © 2013 2013 The Author(s). Published by Taylor & Francis.