Association for Public Policy Analysis & Management

There is a large body of literature that argues that prenatal and postnatal periods are sensitive stages for child development. Negative shocks during these years can impair skill formation, and there may be long term consequences since past human capital investments influence both future skills and investments (Heckman and Cunha, 2007). Exposure to air pollution is one form of negative health shock because of the vulnerability of children’s respiratory organ system. This paper analyzes the effects of exposure to forest fires in Indonesia on children’s human capital formation and parental responses to these shocks.

Studies have found that exposure to air pollution in utero affects health at birth and increases the likelihood of neonatal and infant mortality  (i.e: Chay and Greenstone, 2003; Arceo-Gomez et. al, 2012; Foster et al., 2009; Currie et al., 2011; see Currie et al., 2013 for a review). However, there is more limited evidence on the effects of early life air pollution on medium and long-term human capital outcomes. Exposure to pollution has been linked to negative effects on cognitive skills, education, and labor market outcomes (Almond et al., 2009; Bharadwaj et al., 2013; Currie et al., 2013b; Peet, 2014; and Black et al.,  2013). Other studies find that exposure during the first year of life is associated with lasting negative effects on adult outcomes (Isen et al., 2013; Peet, 2014; Sanders, 2012; see Currie et al., 2013 for a review).

In this paper, we explore the medium-term effects of in utero and early childhood exposure to air pollution in Indonesia. We exploit the geographical variation of Indonesia’s forest fires and cohort variation in exposure as a natural experiment. Between August 1997 and November 1997, during the El Nino phenomenon, Indonesia and Southeast Asia experienced the most intense and long lasting series of forest fires in its history. Prior literature has exploited this variation to study child mortality and adult health.  Jayachandran (2009) uses data from the 2000 census combined with pollution measures from the aerosol index from the Earth Probe Total Ozone Mapping Spectrometer (TOMS). She finds that higher levels of air pollution due to the 1997 fires led to a decline of 1.2 percent in cohort size. Frankenberg, Mckee and Thomas (2005) study the consequences on adult health using data from the Indonesian Family Life Survey (IFLS) and pollution data from TOMS. They find that the fires had a negative impact on adult health.  Similarly, using data from the IFLS and TOMS, we exploit the sharp timing of the forest fires and the geographic variation in the spread of the haze to estimate the effects of the fires on children’s human capital formation. We find that early life exposure to air pollution is associated with lower height for age among children under 5, and the health effects persist. Affected children tend to be shorter and have lower lung capacity when they are older. However, we find no significant effects on cognitive skills, which suggests that parental investments may mitigate the affected children’s early disadvantage.