Early-life famine exposure and later-life outcomes: Evidence from survivors of the greek famine

This chapter examines education and labor market outcomes for cohorts with early-life exposure to the Greek 1941-42 famine. Given the short duration of the famine, we can separately identify effects for cohorts exposed as fetuses, infants and one-year-olds. Our empirical analysis uses data from the 1971, 1981, 1991, and 2001 Greek National Population Housing Censuses. For our main specification that includes birthplace controls, we find negative cohort effects on the likelihood of completing upper secondary school for the cohorts exposed as infants and one-year-olds. Because the famine was more severe in urban areas, we also estimate separate models for urban- and rural-born individuals. Consistent with our prediction, the negative cohort effects for the early-life famine exposed cohorts are larger in the urban-born subsample. The negative cohort effects increase in specifications without birthplace controls. We attribute a part of this increase to a rising share of individuals from areas with negative education and labor market prospects in the cohorts with early-life famine exposure. The cohort effect difference between specifications with and without birthplace controls is largest for the 1942 cohort, a large part of which was conceived during the famine. We suggest that this finding is due to the fact that negative birthplace selection into this cohort occurred not only through famine mortality, like in the other cohorts with early-life exposure, but also through famine-related falls in fertility.