In the Heat of the Moment: An Investigation of Routine Activity Theory and the Urban Heat Crime Effect

Abstract

Evidence suggests that heat is positively correlated with increased crime. However, the mechanisms behind this are not clear. One explanation is physiological aggression, which states that increased temperatures induce a biological response that leads to more aggressive behavior, violence, and crimes. Another theory is Routine Activity Theory, or RAT. RAT states that crimes are a function of likely offenders, suitable targets, and the absence of capable guardians. According to RAT, warmer temperatures, and more pleasant weather increase chances of being outside. This leads to more “social opportunities” for crime and more suitable targets.

In this thesis, I attempt to empiraclly test the theoretical frameworks behind the heat crime relationship. To do so, I study the top 5 most populated U.S. cities (New York City, NY; Los Angeles, CA; Chicago, IL; Houston, TX; and Phoenix, AZ). For each city, I obtain detailed crime records over the last decade and create a dataset that is disaggregated by day and district. I then link this with socioeconomic data and daily weather data. The final dataset is in panel form to control for police district fixed effects. My goal is to study how temperatures affects crime rate and determine how this is distributed across poverty.

I hypothesize that RAT is the primary driver of the heat crime relationship. RAT is the only theory that suggests that there will be differential heat/crime effects among socioeconomic status, since poverty increases the chances that an individual is a motivated offender. Stronger impacts in higher poverty districts will be evidence that RAT is a driver of this effect. To further test RAT, I introduce two other indicators. The first is whether the heat/crime effect drops off in extreme temperatures. The second is if there are differences between crime types.

Apart from Phoenix, I find a substantial heat/crime relationship in all cities. In Phoenix, there is still a strong heat and violent crime relationship. These effects are much stronger in higher poverty neighborhoods. In Chicago, the richest quartiles experience a 2.2% decrease in homicides on days above 85°F while the poorest quartiles experience a 92% increase. In New York City, the highest poverty communities saw a 88.1% increase in homicides on days over 85°F. In Los Angeles, high poverty districts saw a 20.4% increase in violent crimes associated with heat, 5 times larger than in low poverty districts. In Phoenix, the impact of an over 85°F day is linked to a 21.5% increase in violent crimes

I find substantial evidence for RAT. Besides differential socioeconomic effects, I find that there are drop off effects in all cities. Further, the presence of heat crime effects across both violent and property crimes indicates that RAT is present.

Policymakers should seek to understand the inequalities facing low-income communities which make them more vulnerable for crimes. They should address structural barriers that may lead high poverty individuals to become motivated offenders. Policy should also seek to introduce guardians that can intervene and deescalate altercations.