VoxEU Column Frontiers of economic research

Public safety through private action

In the US, more people work in private security than in all police forces combined, yet public debate about crime prevention typically looks at the use of public resources to deter, incapacitate, or rehabilitate criminals. This column calls for more discussion of how private action can make policing more effective and reduce the profitability of crime. One such experiment – “business improvement districts” in Los Angeles – has generated remarkable social benefits.

With respect to crime at least, we’re far safer today than a generation ago. In the US, property crime rates (estimated from victimisation surveys) have dropped a remarkable 80% since 1980. In Britain, there has been a similar decline, with the British Crime Survey recording a 50% drop in crime since 1997. Elsewhere in Western Europe, the picture is not so clear, but at least for property crime (as opposed to assault and other violence) there is a strong downward trend.

To what do we owe this remarkable change in public safety? There is no lack of commentary among economists and criminologists on the subject of the great crime drop, including by such notable scholars as Steven Levitt (2004), Alfred Blumstein (Blumstein and Wallman 2006), and Franklin Zimring (2007). Levitt, for example, credits abortion legalisation, increased imprisonment, hiring more police, and the waning crack epidemic as the primary causes of the crime drop in the US.

Policy debates about how to sustain the crime decline typically focus on how best to use criminal justice resources. But typically missing from the discussion is any account of what has always been the most pervasive and important influence on crime – the efforts of households and businesses to prevent victimisation and assist the authorities in solving crimes (van Ours and Vollaard 2011a). In our view, increasingly effective private action to prevent and deter crime deserves much of the credit for the crime drop.

Among the relevant trends are the growing sophistication of alarms, monitoring equipment and locks; the decline in the use of cash; and the growing employment of private security, including within the context of “business improvement districts”.

Alarms, locks, and other technological advances

The technological advance in crime prevention is perhaps most evident in the case of motor vehicle theft. In the US, fewer vehicles were stolen in 2009 than 30 years earlier despite a doubling of the number of vehicles on the road. Part of the credit must go to the fact that modern vehicles are much easier to lock up (with a single button push), and come equipped with a device called an immobiliser that makes it virtually impossible to start the engine without the appropriate electronic chip – the one embedded in sophisticated keys. Gone are the days when a savvy teenager could hotwire a car and go for a joy ride.

Since 1998, all new cars sold in the European Union have been required to be equipped with an immobiliser, and research by Jan van Ours and Ben Vollaard finds that these devices have reduced theft by 80% in England and Wales (van Ours and Vollaard 2011b). While there is no such requirement in the US, most new vehicles there also have an immobiliser installed.

Another technological advance of some importance are the electronic tracking devices such as LoJack. In cities that support LoJack, vehicle owners can buy and install the electronic tracking device that can be switched on and traced by the police if the vehicle is reported stolen. Since the device is hidden on the vehicle, and vehicles with LoJack are indistinguishable from others, LoJack has a general deterrent effect. Thieves are more likely to be caught, and perhaps more important, undercover businesses that strip parts from stolen vehicles and sell them are more likely to be discovered and shut down.

This is a classic case of ‘co-production’ between private action (purchase of the device) and public enforcement. Research by Ian Ayres and Steven Levitt (1998) finds the social benefit from these devices to be an order of magnitude greater than the cost.

Other technological changes have made life difficult for thieves as well – electronic tags to prevent shoplifting, modern security systems for residences and businesses, improved tracking systems for recovering stolen merchandise, and the replacement of cash with electronic payment methods (which have elabourate defences against fraudulent use). It’s not that these innovations are foolproof, but they surely deserve some of the credit for the remarkable drop in property crime during the last three decades.

Also potentially important is the advent of cell phones, which provide a widely distributed technology for eyewitness reports of crime to the police in real time. Preliminary research suggests that US states that were earlier adopters of cell phones had greater declines in violent crime between 1999 and 2007 (Klick et al. 2011). Our research has documented a long upward trend in the percentage of crimes reported to the police, and we are willing to bet that cell phones get part of the credit (Cook and MacDonald 2011b).

The lack of systematic research on these trends has perhaps contributed to their neglect in policy debate. Too often the choice in crime prevention is defined by alternative schemes for using public resources to deter or incapacitate or rehabilitate criminals.

We want to add to this discussion the possibility of private action to make policing more effective (LoJack, crime reporting), and to reduce the profitability of crime. In the US, immobilisers are not required and 15% of new vehicles are sold without them. Insurance regulations provide little or no incentive for owners to buy LoJack or other crime prevention devices. Victims and witnesses who cooperate with the police are often treated shabbily, rather than rewarded for providing a public service. Redressing these problems should be as much a part of the crime policy discussion as sentencing or police deployment.

Private security and public order

What about private security services? In the US, there are more people working in private security than in all police forces combined, and private security employment has been growing faster over the last three decades. Yet once again the public discourse on crime control largely ignores this elephant in the room, and once again the evaluation research has little to say. One exception is new research evaluating private security activities by “business improvement districts” (BIDs). Our research has documented the cost-effective results of security investments by BIDs (Cook and MacDonald 2011a).

A BID is a non-profit organisation created by commercial property owners that provides local public goods, usually including public safety. To create a BID requires that most of the property owners in the relevant neighbourhood sign on and agree to pay fees to the organisation. By hiring private security and working closely with the police, BIDs ‘co-produce’ crime control.

Our research investigates the crime-prevention performance of 30 BIDs that were created in Los Angeles during the 1990s. The statistical evaluation demonstrates that BIDs reduce crime – that is, that crime rates in BIDs are lower than what they otherwise would have been, as indicated by crime trends in non-BID areas of the city.

Furthermore, we find a clear connection between the “dose” (the amount of BID spending on private security) and the “response” (crime reduction). Every $1,000 spent in a BID neighbourhood generates $20,000 of social benefit in the form of crime reduction – a remarkable 20-to-1 benefit-cost ratio.

The city of Los Angeles also saves money because the arrest rate actually decreases in BIDs. And there is no evidence that crime is displaced to neighbouring areas. Thus, the social cost savings from BID security expenditures is an order of magnitude greater than expenditures and provides a direct payoff to the streets where they provide extra private security.

It appears very much in the interest of Los Angeles to continue providing organisational support for its BIDs. On the other hand, the analysis does not imply that all other neighbourhoods in Los Angeles should organise, since it is reasonable to suppose that many of the areas with the highest potential payoff have already organised.

The same can’t be said for all other cities in the US. There are currently upwards of 1,000 BIDs in US cities, but many are handicapped by inadequate legal infrastructure. What is needed to create an effective system of BIDs is a clear legal framework that includes a solution to the “free riding” problem – some merchants will inevitably want to enjoy the benefits of BID services without actually paying the annual fees. In Los Angeles, the city collects those fees as an addition to the property tax bill, with the same legal compulsion.

At a time of severely constrained public finances, it is especially appealing to harness private security to the public cause of crime control. BIDs, by organising entire neighbourhoods, can accomplish just that purpose. They have begun to appear in Britain and warrant consideration elsewhere.


Ayres, Ian and Steven D Levitt (1998), "Measuring Positive Externalities from Unobservable Victim Precautions: An Empirical Analysis of Lojack", Quarterly Journal of Economics, 113(1):43-77.

Blumstein, Alfred and Joel Wallman (2006), "The Crime Drop and Beyond", Annual Review of Law and Social Science, 2:125-146.

Cook, Philip J and John MacDonald (2011a), “Public Safety through Private Action: An Economic Assessment of BIDs”, The Economic Journal,121 (May): 445-462.

Cook, Philip J and John MacDonald (2011b), “The Role of Private Action in Controlling Crime”, in PJ Cook, J Ludwig, and J McCrary, Controlling Crime: Strategies and Tradeoffs, University of Chicago Press, forthcoming.

Klick, Jonathan, John MacDonald, and Thomas Stratmann (forthcoming), “Mobile Phones and Crime Deterrence: An Underappreciated Link” in Keith Hylton and Alon Harel (ed.), Handbook on the Economics of Criminal Law, Edward Elgar.

Levitt, Steven D (2004). "Understanding Why Crime Fell in the 1990s: Four Factors That Explain the Decline and Six That Do Not", Journal of Economic Perspectives 18(1):163-190.

van Ours, Jan and Ben Vollaard (2011a), "Reducing the invitation to crime", VoxEU.org, 7 July. 

van Ours, Jan and Ben Vollaard (2011b), “The engine immobilizer: a non-starter for car thieves”, mimeo, Tilburg University.

Zimring, Franklin E (2007), The Great American Crime Decline, Oxford University Press.

630 Reads