Do COVID-19 Restrictions on Activities Reduce Crime?

A natural experiment is taking place that relates the decline in urban crime to the restrictions on most businesses and residents imposed to reduce the spread of the COVID-19 virus. What matters for this study is that, in this time period, the exposure to most types of urban crime for most people has been reduced substantially and exogenously. 

Balkin and McDonald in Journal of Urban Economics, 1981 developed a market model for street crime in which potential victims choose the amount of time to expose themselves to the risk of victimisation. At the same time criminals choose the amount of time needed to search for victims. The model can be framed in a set of five simple equations. The measured crime rate can increase, decrease, or remain the same when the real crime rate increases. The impact on measured crime rate will depend on how much people reduce their hours of exposure to the possibility of crime when their real risk increases. Empirical evidence to support the model based on people’s exposure to street crime was reported by McDonald and Balkin in Urban Studies, 1983.

The model is depicted as a stylised supply and demand diagram.  The demand for exposure time is a negatively sloped function of risk of victimisation per unit of exposure time, and the supply of offender time is a positively sloped function of the probability of finding a victim per unit of search time. Due to the virus mandates, we show a large exogenous reduction (leftward shift) in the demand for exposure time, which results in declines in exposure time and the risk of victimisation and a large decline in the nominal (reported) crime rate per month. 

Simultaneously, the closure of all non-essential businesses means that people are not working, or not working as much as before the COVID-19 restrictions, and likely are losing income. Some of those workers who lose their jobs may be driven into some form of crime, and some criminals who also hold legitimate jobs may devote more time to crime.

Combining both an exogenous reduction in the exposure to crime by potential victims and a potential offender exogenous increase in the supply of time searching for victims, results in a decrease in potential victim exposure time and a decrease in reported crimes but an ambiguous effect on the real victimisation risk. 

We look at recent data for five major cities: New York, Los Angeles, San Francisco, Chicago and Philadelphia. The data show that reported crimes, in general, have declined, but the net effect on the risk of crime is not measured and may be negative or positive. 

The City of Philadelphia is of particular interest. Overall incidents of crime fell by 25% from March 16 to March 22, 2020 compared to the average week. But there was an increase in auto theft which is considered an anomaly by some observers, but it may be a case of crime substitution.  Many autos in Philadelphia are parked on the street, not in garages and are now less observed by natural surveillance. Burglary and theft are in the set of reported crimes that are lower. They are property crimes that often involve entering a home or its territory. Most people are staying at home most of the time, so criminals are more likely to encounter people in the attempt to commit these types of crime. Burglars and thieves wish to avoid people, so these crimes have declined as well.

Several questions arise. Will the decline in most urban crimes (and increase in auto theft) continue? Will other cities experience the same crime pattern? Will offenders find and focus on new types of targets compatible with social distancing? Will crime increase when the restrictions on business and residents end? Can something be learned that will mitigate any increase in crime when the activity restrictions are lifted?

While many may judge these crime changes desirable, we doubt that society is overall better off, holding aside the health impacts of the COVID-19 virus. We think this for two reasons: (1) The quality of life is diminished by reduced access to the full range of life’s activities, and (2) while the criminal justice caseload is reduced, the real risk of crime may not have declined. 

This blog post is based on a larger paper: The COVID-19 Virus and the Decline in Crime which can be obtained on the website of the Social Science Research Network.

Read McDonald and Balkin's Urban Studies paper Citizen Demand for Exposure to Street Crime on Urban Studies OnlineFirst.