In the heat of the moment: Economic and non-economic drivers of the weather-crime relationship

Can Social Protection Reduce Environmental Damages?

“…The mean value of the daily homicide risk in the data is 0.066%, implying that effect size of a one-degree Celsius increase in temperature amounts to a 2.1% increase in daily homicide risk. These results are consistent with earlier work in the United States (Ranson, 2014) and concurrent work in Mexico (Baysan et al, 2019;Cohen and Gonzalez, 2018) and India (Blakeslee et al, 2018). The results are similar when we include locality-by-month-of-year fixed effects (Column 2).…”

“…uses monthly data and finds similar effects to ours. Three concurrent papers also examine the relationship between temperature and violence in Mexico(Baysan et al, 2019; Cohen and Gonzalez, 2018) and India(Blakeslee et al, 2018).7 Other plausible mechanisms include increases in interactions with others and increased alcohol use(Cohen and Gonzalez, 2018), although we provide suggestive evidence that these are unlikely to be the dominant mechanisms.…”

“…Empirical evidence shows that the effect of same-day temperature on homicides varies by levels of income and economic development, both across and within countries. Longitudinal studies on this relationship have been conducted, to our knowledge, in three countries -United States, India and Mexico; effects in Mexico and India (Blakeslee et al, 2018) are at least an order of magnitude larger than the effects in the United States (Ranson, 2014). Within Mexico, the effect of temperature on homicides declines as income and levels of electrification rise (Figure 2).…”

Weather and Crime—Cautious evidence from South Africa

“…While we do not have the same timely resolution, our crime records at a monthly level also allow us to investigate in detail how crime rates fluctuate over the agricultural cycle in response to drought shocks. It hence provides strong evidence in support of agricultural income shocks as an underlying driver of criminal activity, which is in line with what Blakeslee and Fishman's (2018) and Blakeslee et al's (2021) analyses suggest. 5 Moreover, our paper speaks to the growing literature on the determinants of crime in South Africa.…”

“…In the context of the Global South, we complement Blakeslee and Fishman (2018), who use detailed annual district-level crime data from India and find that heat and drought increase all categories of crime in their records, with a larger impact on property crimes than on violent crimes. Blakeslee et al (2021) examine daily crime statistics and weather data in India. They observe that violent crimes respond to both daily and seasonal temperature variation.…”