Hot Tempered: The Effect of Temperature and Extreme Heat on Child Maltreatment

A burgeoning literature points to rising temperatures as a cause of disparities in cognitive and non-cognitive ability, birth rates, aggression, health, and other correlates of human capital (see e.g. Carleton & Hsiang, 2016). Children in particular are uniquely vulnerable to the insults of climate change. Extreme heat not only affects children directly in terms of infant mortality and early health, parents of young children are increasingly disconnected from protective resources and benefits, cultivating a stressful environment ripe for maltreatment when temperatures rise. Whether temperature directly predicts violence against children is unknown. To that end, in this paper I estimate the causal effect of temperature on child abuse by linking daily temperature data to the census of administrative Child Protective Services (CPS) reports for 48 states aggregated to two-week county-level periods 2010 through 2017. In addition to Linear Probability Models, I employ a Regression Discontinuity (RD) framework to examine CPS report rates around the start of heatwaves, testing whether Low Income Home Energy Assistance Program (LIHEAP) state-level cooling support programs and air conditioning (a/c) penetration rates mitigate this effect. My preliminary findings imply that a 10 degree Fahrenheit increase in temperature increases CPS reports by up to 4.6 percent. I detect much larger effects among reports alleging neglect, implying that extreme temperatures affect both parenting and reporting behaviors. The effects of cumulative exposure to extreme heat appear nearly as harmful. LIHEAP cooling programs and a/c penetration rates partially mitigate this relationship, suggestive of large gains from expanding access to such programs. This work contributes to nascent knowledge on the effects of climate change on parenting behaviors and child health, unveiling potential long-term effects of extreme heat among the most disadvantaged populations.