South Africa has one of the highest per-capita crime and murder rates of any country in the world. However, the majority of violent crime in South Africa is localized in a small proportion of low-income, marginalized communities. Increasing law enforcement in areas like these is a common strategy for reducing violent crime–but the Western Cape Government in South Africa asked ideas42 to identify other, behaviorally informed solutions for improving safety in South Africa’s low-income communities.

Working with the Western Cape’s Department of Community Safety and researchers from the University of Cape Town, we sought to understand the behavioral and psychological factors that might be perpetuating the incidence of violence in these communities. We found that young people were sticking with the status quo by spending time in unsafe places, environments where – because of situational factors – they were more likely to both perpetrate crime, and become a victim of crime.

To try and overcome this tendency to spend time in unsafe environments – especially on weekend evenings when most crime was committed – we designed a computer-based activity-planning tool to help nudge them toward less risky activities. This tool drew on behavioral insights that people are:

• Less likely to go with the status quo option when they are forced to make an active decision about all available options.
• More likely to follow through on their intended actions when they make public commitments and plan ahead.

We developed a prototype tool and tested it with young people from Cape Town’s Cape Flats neighborhoods. We wanted to assess whether it was more effective than the traditional approach of improving safety awareness in two respects:

• Improving how “safe” the young people reported feeling.
• Reducing the number of violent events they experienced in the past week.

The tool had powerful effects. A randomized controlled trial found that young people in the treatment group were half as likely as the control group to participate in unsafe activities by the end of the study. The treatment population was also almost half as likely as the control group to report feeling unsafe, and half as likely to report experiencing violence in the past week.

These results have significant implications for how we think about improving safety and reducing crime. In many parts of the world violent crime is a serious problem, with policy-makers and practitioners alike looking to increased investment in enforcement as a way of mitigating the problem. This pilot project illustrates that supporting targeted decision-making and planning for both potential victims and perpetrators has the potential to significantly reduce violent crime. Read the full brief from this project here.

Interested in learning more about this work applying behavioral science to a crucial social problem? Reach out to us at info@ideas42.org or tweet at @ideas42 to join the conversation.