Thursday 14 July 2011

Crime nudge

Posted by Dr Steve Gibbons, LSE and SERC

The Government is a fan of open data, arguing that transparency keeps the public sector on its toes. Ministers also like ‘nudges’, light-touch interventions that push people towards the ‘right’ decision.

The two came together unexpectedly this week in a survey from insurance firm Direct Line. Earlier this year Oliver Letwin unveiled street-level crime maps, a flagship open data initiative. However, Direct Line found 11% of people had since not reported a crime – because they were worried about the effects of mapping higher crime on local house prices.

It’s always worth checking exactly how survey questions like this are phrased. The question actually asked was:

“Have you been affected by a crime that you did not report to the police because you feared it would show up on an online police crime map, making it more difficult to rent/sell your house?”

Of 3000-odd people polled, 11% of respondents said yes. Around 75% of these had turned a blind eye to anti-social behaviour, and 45% to vehicle crime. Smaller fractions had either witnessed, or been a victim of assault or street robbery – but had not reported this.

We will have to wait until more recent official data is available on trends in recorded crime – relative to the victimisation survey data from the British Crime Survey – to see whether the Direct Line research really stacks up.

Assuming it does, two points follow.

First, for spatial and real estate economists none of this is very surprising. In 2004 I published this paper in the Economic Journal, based on identical data to that used in the street-level crime maps, showed that incidents of criminal damage (vandalism, graffiti etc.) had big effect on house prices across London. A 10% increase in crime pushed house prices down by 1.5%.

Second, for policymakers this is an unwelcome reminder that it’s easy to ‘nudge’ people in the wrong direction. The Home Office told the FT that providing local information on crime will push crime down, by encouraging people to ‘hold their local police to account … pushing police to tackle crime’. Perhaps. But the Direct Line survey suggests that detailed online crime data could have some serious unintended consequences, given the economic incentives for home-owners to conceal the true level of crime in their neighbourhood. It also follows that policies which attempt to bring crime down by publicly revealing its true extent may not be very successful.