Tweet
http://www.economist.com/business-fi...ry_id=15867693
Economics focus
The mean streets of Guildford
Spending more on education and private security are cost-effective ways of cutting crime
Apr 8th 2010 | From The Economist print edition
GUILDFORD is a prosperous town in London’s commuter belt and an unlikely setting for a seminar on crime. But one of the best sessions at the recent annual conference of the Royal Economic Society (RES), held at the University of Surrey, teased out two of the big themes on the economics of crime—deterrence (what it is that prevents crime?) and incentives (what it is that makes for law-abiding citizens?).
To start, Philip Cook of Duke University unveiled a new paper, written with John MacDonald of the University of Pennsylvania, on private contributions to public order. Their paper is based in part on a study of Business Improvement Districts (BIDs) in Los Angeles. BIDs are not-for-profit bodies that provide services, such as private security guards and sanitation, on behalf of local firms. The marvel of such public goods is that one firm’s use of them does not make them less useful to others. The flaw is that businesses can benefit even if they do not bear the costs. A change to Californian law in the 1990s addressed this problem by forcing all businesses to join a collective scheme if enough local firms had signed up to it. It helps, too, that the city authority collects the levies that finance the BIDs.
Private security turns out to be a cost-effective way of cutting crime. The study compares crime rates in 30 BIDs set up after 1995 with those in neighbouring districts. BIDs tend to be high-crime areas, so the authors adjust for this. They find that each $10,000 spent by an average BID resulted in 3.4 fewer crimes per year.
To work out whether this was money well spent, the authors surveyed the public to put a cash value on each crime prevented. People were asked whether they would vote for a scheme that reduced a particular crime by 10% at a particular cost in tax dollars (the range of “offers” varied from $25 to $225). The authors used the responses to assign a social cost to different crimes. They calculate that preventing a robbery is valued at $263,000, an assault at $79,000 and a burglary at $21,000. Given the reduction they bring about in each sort of crime, every $10,000 spent by the average BID bought some $200,000-worth of crime prevention.
A benefit-to-cost ratio of 20 to one is impressive, though it does not tell us whether the schemes directly benefit the firms that pay for them—through increased custom, fatter profits, higher property values, and so on. It is clear, however, that private security is good value for society as a whole. So good, in fact, that one suspects that some of the costs have been left out. Perhaps crime was not stamped out but merely shifted elsewhere. Or perhaps BIDs work so well because they draw on extra support from the police. That is not the case, says Mr Cook. Establishing a BID leads to fewer arrests and so reduces the cost of policing. Nor is there evidence that crime increases in neighbouring districts after BIDs are set up, he says. If anything, crime nearby falls.
Why is private security apparently so cost-effective? One reason, says Mr Cook, is simply that guards are paid less than police officers. Another is they are dedicated to a single district and are directly responsible for making it safe. Guards can specialise. They know which shifty characters to look out for and which policing works best in their area. Unlike policemen, they are not called away to supervise a parade or protect a dignitary.
The research also shows how effective “target hardening” (ie, self-protection against crime) can be. Mr Cook noted that there were twice as many cars in America in 2008 as in 1989, but fewer car thefts. Steering locks, engine immobilisers and tracking systems have made newer cars harder to steal. In a similar vein, a paper presented by Ben Vollaard of Tilburg University showed that newly built homes are harder to burgle. Mr Vollaard and his co-author, Jan van Ours, reckon that homes put up after a change in the Dutch building code in 1999 were 26% less likely to be broken into than those built beforehand. To comply with the code, builders had to fit high-quality locks and burglar-proof windows and doors. These may not put off a determined thief but are enough to slow down an opportunist, said Mr Vollaard.
Too cool for school
Are there ways to prevent people from becoming criminals in the first place? In principle, a lengthier education ought to reduce crime by raising people’s future earning power from legitimate work, making a criminal career less attractive. School also keeps would-be criminals in touch with the right sort of peers and social attitudes. There is plenty of evidence that a lack of education goes hand in hand with criminal behaviour. Studies of America’s jail population in the 1990s showed that most inmates had not finished high school. But few studies have established that less education is actually a cause of crime.
A third paper presented at the conference, written by Stephen Machin of University College London, Olivier Marie of Maastricht University and Suncica Vujic of the London School of Economics, uses a statistical trick to find a causal link between low education and crime. The authors looked at the crime rates of a cohort of British school-leavers, some of whom were forced to stay in school for longer because of a legal change to the school-leaving age. They found this group was less likely to engage in criminal behaviour than an earlier cohort. The authors calculate that one year of extra education reduces property crimes by 1-2%, and that the cost of the extra schooling is outweighed by the benefits of reduced crime. These results echo a study of American crime by Lance Lochner of the University of Western Ontario and Enrico Moretti of the University of California, Berkeley, which found the biggest benefit from extra education was in fewer violent crimes. That result is not replicated in the newer study, perhaps because there are too few murders in Britain to show up statistically. After all, Guildford is a long way from Los Angeles.
The article refers to the following papers:
“Mobilising Private Inputs for Crime Prevention.” NBER working paper (forthcoming)
“Does Regulation of Built-in Security Reduce Crime? Evidence From a Regression Discontinuity Approach.” Univesity of Tilberg (mimeo)
“The Crime-Reducing effect of Education.” Center for Economic Performance working paper (forthcoming)
"The Effect of Education on Criminal Activity: Evidence from Prison Inmates, Arrests and Self-Reports" by Lance Lochner and Enrico Moretti, American Economic Review (2004)
The presentations from the RES special session on the economics of criminal behaviour can be found at: http://www.resconference.org.uk/inde...d=29&Itemid=36
Economics focus
The mean streets of Guildford
Spending more on education and private security are cost-effective ways of cutting crime
Apr 8th 2010 | From The Economist print edition
GUILDFORD is a prosperous town in London’s commuter belt and an unlikely setting for a seminar on crime. But one of the best sessions at the recent annual conference of the Royal Economic Society (RES), held at the University of Surrey, teased out two of the big themes on the economics of crime—deterrence (what it is that prevents crime?) and incentives (what it is that makes for law-abiding citizens?).
To start, Philip Cook of Duke University unveiled a new paper, written with John MacDonald of the University of Pennsylvania, on private contributions to public order. Their paper is based in part on a study of Business Improvement Districts (BIDs) in Los Angeles. BIDs are not-for-profit bodies that provide services, such as private security guards and sanitation, on behalf of local firms. The marvel of such public goods is that one firm’s use of them does not make them less useful to others. The flaw is that businesses can benefit even if they do not bear the costs. A change to Californian law in the 1990s addressed this problem by forcing all businesses to join a collective scheme if enough local firms had signed up to it. It helps, too, that the city authority collects the levies that finance the BIDs.
Private security turns out to be a cost-effective way of cutting crime. The study compares crime rates in 30 BIDs set up after 1995 with those in neighbouring districts. BIDs tend to be high-crime areas, so the authors adjust for this. They find that each $10,000 spent by an average BID resulted in 3.4 fewer crimes per year.
To work out whether this was money well spent, the authors surveyed the public to put a cash value on each crime prevented. People were asked whether they would vote for a scheme that reduced a particular crime by 10% at a particular cost in tax dollars (the range of “offers” varied from $25 to $225). The authors used the responses to assign a social cost to different crimes. They calculate that preventing a robbery is valued at $263,000, an assault at $79,000 and a burglary at $21,000. Given the reduction they bring about in each sort of crime, every $10,000 spent by the average BID bought some $200,000-worth of crime prevention.
A benefit-to-cost ratio of 20 to one is impressive, though it does not tell us whether the schemes directly benefit the firms that pay for them—through increased custom, fatter profits, higher property values, and so on. It is clear, however, that private security is good value for society as a whole. So good, in fact, that one suspects that some of the costs have been left out. Perhaps crime was not stamped out but merely shifted elsewhere. Or perhaps BIDs work so well because they draw on extra support from the police. That is not the case, says Mr Cook. Establishing a BID leads to fewer arrests and so reduces the cost of policing. Nor is there evidence that crime increases in neighbouring districts after BIDs are set up, he says. If anything, crime nearby falls.
Why is private security apparently so cost-effective? One reason, says Mr Cook, is simply that guards are paid less than police officers. Another is they are dedicated to a single district and are directly responsible for making it safe. Guards can specialise. They know which shifty characters to look out for and which policing works best in their area. Unlike policemen, they are not called away to supervise a parade or protect a dignitary.
The research also shows how effective “target hardening” (ie, self-protection against crime) can be. Mr Cook noted that there were twice as many cars in America in 2008 as in 1989, but fewer car thefts. Steering locks, engine immobilisers and tracking systems have made newer cars harder to steal. In a similar vein, a paper presented by Ben Vollaard of Tilburg University showed that newly built homes are harder to burgle. Mr Vollaard and his co-author, Jan van Ours, reckon that homes put up after a change in the Dutch building code in 1999 were 26% less likely to be broken into than those built beforehand. To comply with the code, builders had to fit high-quality locks and burglar-proof windows and doors. These may not put off a determined thief but are enough to slow down an opportunist, said Mr Vollaard.
Too cool for school
Are there ways to prevent people from becoming criminals in the first place? In principle, a lengthier education ought to reduce crime by raising people’s future earning power from legitimate work, making a criminal career less attractive. School also keeps would-be criminals in touch with the right sort of peers and social attitudes. There is plenty of evidence that a lack of education goes hand in hand with criminal behaviour. Studies of America’s jail population in the 1990s showed that most inmates had not finished high school. But few studies have established that less education is actually a cause of crime.
A third paper presented at the conference, written by Stephen Machin of University College London, Olivier Marie of Maastricht University and Suncica Vujic of the London School of Economics, uses a statistical trick to find a causal link between low education and crime. The authors looked at the crime rates of a cohort of British school-leavers, some of whom were forced to stay in school for longer because of a legal change to the school-leaving age. They found this group was less likely to engage in criminal behaviour than an earlier cohort. The authors calculate that one year of extra education reduces property crimes by 1-2%, and that the cost of the extra schooling is outweighed by the benefits of reduced crime. These results echo a study of American crime by Lance Lochner of the University of Western Ontario and Enrico Moretti of the University of California, Berkeley, which found the biggest benefit from extra education was in fewer violent crimes. That result is not replicated in the newer study, perhaps because there are too few murders in Britain to show up statistically. After all, Guildford is a long way from Los Angeles.
The article refers to the following papers:
“Mobilising Private Inputs for Crime Prevention.” NBER working paper (forthcoming)
“Does Regulation of Built-in Security Reduce Crime? Evidence From a Regression Discontinuity Approach.” Univesity of Tilberg (mimeo)
“The Crime-Reducing effect of Education.” Center for Economic Performance working paper (forthcoming)
"The Effect of Education on Criminal Activity: Evidence from Prison Inmates, Arrests and Self-Reports" by Lance Lochner and Enrico Moretti, American Economic Review (2004)
The presentations from the RES special session on the economics of criminal behaviour can be found at: http://www.resconference.org.uk/inde...d=29&Itemid=36