The effect of COVID-19 restrictions on routine activities and online crime

Theories of crime emphasize the role of different causal factors. Some focus on what psychological (or similar) factors motivate offending, ignoring the role of crime opportunity, while others focus on the latter, taking offender motivation as a given. Testing such theories is difficult absent significant events that influence both, but the COVID-19 pandemic has created such conditions. As a consequence of government containment policies, we have seen dramatic changes to people’s day-to-day activities, with people spending more time at home (ONSa, 2020; ILO 2020) and more time online (Kemp et al., 2021), reducing opportunities for crime in urban environments but increasing them online. Equally, by restricting mobility, containment policies have had the potential to cause economic and emotional strain that would affect offender motivation – in the short-term at least. Here, we test if and how COVID-19 containment policies impacted upon different types of fraud committed online and in a physical context. According to offender-focused explanations, the strain caused by the pandemic would be expected to lead to increases in crime, which should largely persist regardless of subsequent variation in macro level changes to people’s routine activities. According to opportunity theories of crime, we would also expect crime to change, but for levels to closely follow changes to mobility and online activity – increasing as restrictions are imposed and declining as they are relaxed. Using data for online crime and doorstep fraud, online sales, mobility data and an Autoregressive Conditional Hetroskedasticity (ARCH) statistical modelling framework, our findings provide striking evidence consistent with routine activity theory (Cohen & Felson, 1979), with oscillations in people’s day-to-day activities appearing to shape peaks and troughs in levels of crime committed online.