In the past century, more people have perished from famine than from the two World Wars combined. Many more were exposed to famine and survived. Yet we know almost nothing about the long run impact of famine on these survivors. This paper addresses this question by estimating the effect of childhood exposure to China's Great Famine on adult health and labor market outcomes of survivors. It resolves two major empirical difficulties: 1) data limitation in measures of famine intensity; and 2) the potential joint determination of famine occurrences and survivors' outcomes. As a measure of famine intensity, we use regional cohort size of the surviving population in a place and time when there is little migration. We then exploit a novel source of plausibly exogenous variation in famine intensity to estimate the causal effect of childhood exposure to famine on adult health, educational attainment and labor supply. The results show that exposure to famine had significant adverse effects on adult health and work capacity. The magnitude of the effect is negatively correlated with age at the onset of the famine. For example, for those who were one year old at the onset of the famine, exposure on average reduced height by 2.08% (3.34cm), weight by 6.03% (3.38kg), weight-for-height by 4% (0.01 kg/cm), upper arm circumference by 3.95% (0.99cm) and labor supply by 6.93% (3.28 hrs/week). The results also suggest that famine exposure decreased educational attainment by 3% (0.19 years); and that selection for survival decreased within-region inequality in famine stricken regions.