Does Ramadan Exposure In Utero Affect Long-term Labor Market Behavior? Evidence From Indonesia

This paper utilizes the Islamic holy month of Ramadan, a period of religious fasting, as a natural experiment to identify the effects of in utero nutrition on the labor market behavior of children in Indonesia when they become adults. Approximately seventy five percent of all pregnancies overlap with Ramadan, suggesting that in 2010 alone, more than 1.2 billion Muslims globally and 155 million Muslims in Indonesia were potentially exposed to their mother’s fasting. In contrast to previous research, this paper uses household and biological sibling fixed effects for sub-samples of adults, to control for any bias associated with the selective timing of pregnancies as well as to control for time invariant differences across families in rates of compliance (fasting) during Ramadan. In this sense, this study estimates average treatment effects, rather than intent to treatment effects, for sub-samples of household members. Using the Indonesian Family Life Survey (IFLS), the results show that those exposed work 4.5% fewer hours and are 3.2% more likely to be self-employed, with average treatment effects for sub-samples even larger. Moreover, by making use of novel religiosity data from the latest wave of the IFLS, this study shows that these effects are strongest in religious families. No effects are found on non-Muslims. When suggestive channels are explored, results show that children aged 7-15 score 5.9% lower on Raven’s Colored Progressive Matrices assessment and 7.8% lower on math test scores. These estimates generally increase in magnitude when biological sibling fixed effects are used. Exploiting the panel feature of the IFLS, this paper further identifies increased probability of child labor and fewer hours of study during elementary school as two other deeper behavioral channels. In addition, those exposed report lower birth weight, which indicates that these effects are indeed driven by in utero effects. When days of potential exposure during pregnancy are explored, 6 to 18 days of exposure seems to be a critical window during which the marginal effects of fasting are maximized, after which the effects are stabilized.