The community spillover effect of violent crime
This week’s guest blog is from Panka Bencsik from the University of Sussex on the results from her research on crime’s costs to communities, specifically how violent and sexual crimes increase stress for locals in the area.
Crime in our neighbourhoods can be scary. Research suggests that it can leave people feeling less safe, it can change the face of the streets, and it can even lessen the value of property. However, separating the negative effect of ongoing crime – such as annual crime rates – and the impact of each individual criminal event on locals’ stress levels has not been possible before.
Impact of crime on individual wellbeing
Research I conducted on the Thames Valley region of England set out to measure the impact each violent or sexual crime had on the neighbourhood it took place in, beyond the effect of the ongoing crime rate. Taking into account the unique characteristics of the neighbourhood, in my paper, Stress on the Sidewalk: Mental health costs of close proximity crime, I found that violent and sexual crimes have an additional negative mental health effect. Specifically, violent and sexual crimes increase stress levels for those in the vicinity for three days after they were committed.
Violent and sexual crimes are often considered some of the most severe crimes because they directly impact a victim, and endanger physical safety. They are also the most feared crimes in England and Wales (based on results from the Crime Survey for England and Wales). People in areas with higher levels of these crimes tend to face various negative outcomes, like decreased amount of nighttime work and outdoor exercise due to not feeling safe.
However, it was not previously clear whether each individual crime takes a toll on people beyond the victim, or if it is a compilation over time that has an effect. By joining two extremely micro-level datasets, I found that violent and sexual crimes do indeed each take a toll on entire neighbourhoods, not just individual victims.
Measuring stress and crime levels
Specifically, I measure individuals’ stress levels using the Mappiness survey, which was created by George MacKerron, and is a voluntary daily panel survey. This means that those who join begin to receive a ‘beep’ twice a day, and at these times they are asked how happy they are, relaxed they are, what they are doing, and who they are with. After a few days of participation, the respondents start to get feedback on their own happiness, such as about when and with whom they are the happiest. Since 2011, the survey attracted more than 3 million responses from more than 30,000 individuals, and it continues to be open and free to download for anyone interested.
I combine this Mappiness data on stress levels with data on crime for years 2010 to 2017 using highly secure crime data from the Thames Valley region. Thames Valley, which sits just west of London, has 2.1 million residents, and includes cities such as Oxford, Reading, and Slough. Crime data from this region is provided by Thames Valley Police, who have taken a leading role in collaborations with academics to improve policing outcomes.
Using the uniquely detailed geographical information available in both datasets, I was able to measure crime levels at the smallest statistical area in England and Wales: the Output Area. Output Areas are usually about the size of a single street, they on average contain 131 households. Joining the crime and the stress data at this specific locality level meant I could identify when a violent or sexual crime happened recently on the very street the Mappiness respondent was standing. I found that each violent or sexual crime increases stress for the following three days. About 250 adults are impacted by each reported crime of this type, and their stress levels increases substantially.
Interestingly, it is not crimes the day of or the day before a response that appear to cause the increased stress, but crimes reported to police two days prior. This lag between the crime being committed and the local stress levels going up suggests that there might be a mediator present, such as word of mouth or the media. To test this, I scraped news data, and found that nationwide stress levels increased in response to the number of articles written on crimes in the domestic news section of leading UK daily newspapers.
Rethinking the cost of crime
We are aware that crime causes social harm in its negative impact on victims; on businesses; through lost wages; through transforming neighbourhoods; and leading to altered behaviour and local mood in the long-term. The findings suggest that crime also has an additional, immediate negative wellbeing effect, which should be added to the complete cost of crime.
Bryson, Alex and George MacKerron. How does terrorism affect individuals’ wellbeing?. Report, 2018.
Cornaglia, Francesca and Andrew Leigh. Crime and mental wellbeing. CEP Discussion Paper No 1049, 2011.
Dustmann, Christian and Francesco Fasani. The effect of local area crime on mental health. The Economic Journal, 126(593):978–1017, 2016.
Hamermesh, Daniel S. Crime and the timing of work. Journal of Urban Economics, 45(2):311–330, 1999.
Janke, Katharina Marie, Carol Propper and Michael A. Shields. Does violent crime deter physical activity? IZA Working Paper Series. No. 7545, 2013.
Kirchmaier, Tom and Carmen Villa Llera. Murders in London. Report, 2018.
Linden, Leigh and Jonah E. Rockoff. Estimates of the impact of crime risk on property values from Megan’s Laws. American Economic Review, 98(3):1103–27., 2008.
What's the evidence on social connection and place?
How do the places we live influence individual and community wellbeing?